This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
But, this is an important sticking point. The only way allowing people to escalate from fists to guns reduces overall murder is if the 2-3 OOM increase in lethality is paired with a 2-3 OOM decrease in fist fights. This seems implausible. So, the only plausible conclusion is that a norm of escalating fist fights to gun fights causes a large increase in dead bodies. To me that is a steep price to pay for honor and fairness.
I don't think when numbers are OOM that you can fairly describe it a "quibbling".
Definitely not when the expected loss of life is a decade (e.g. 50 years * 22% = 11) versus ~a week (2-3 OOM less). I don't think one man's honor and sense of fairness at a bar one night are worth an expected value loss of a decade of human life. Do you? Do you honestly think the deterrence effect is anything other than a rounding error next to a decade of human life lost in expectation?
I just pointed out how important the distinction is between fist fights and unprovoked assaults. A world with fewer fist fights sounds nice to me, but to each their own. A world with fewer unprovoked assaults, though, is one I'd really like to live in, even if that means I never get to blindside someone myself. Wouldn't you agree? Wouldn't "I can never safely give someone a black eye out of the blue" be a price so small that it's worth paying for even a slightly reduced risk of being punched and possibly even killed out of the blue? Maybe not when we're thinking about 22 year olds, I guess? It's a real shame that men often become old enough to murder people with fists before they become old enough to realize they should avoid any risk of murdering people with fists.
Everything I'm reading about this case makes it sound like it moved from "unprovoked assault" to (albeit unfair) "fist fight", from which point Cranston did escalate to manslaughter if not murder ... but I wouldn't dare swear to any of that before seeing the video, nor generalize it to other cases. Would you? I've seen enough cases where the initial descriptions and the eventually published videos turned out to be tangentially related at best.
You're also ignoring the distinction between different dead bodies. Why? If someone invents the "murder a little child" button, a magic device which can only be used once and has a fifty-fifty chance of working, would you kill them if that was the only way to stop them from pressing it? 1 expected death for 1/2 would be an increase in dead bodies, which in this arithmetic we're treating as interchangeable, so that seems like a no, but maybe that ratio is fairly described as quibbling. What if it only had a 0.1% chance of working? We're now talking about multiple orders of magnitude, right? So at this point it's too steep a price to pay, and little Suzie might need to risk biting it so the guy who gets a kick out of her risk doesn't have to? Except ... couldn't he just not press the button, if he's aware of our position and his life has value to him? If the answer is "yes", then we've solved the problem! 0.001 deaths would have been worse than than 0 deaths. If the answer is "no", then his life is so grossly different from his likely victims that even comparing 0.001 deaths to 1 death seems to be comparing apples and oranges ... and so perhaps we've still solved the problem.
So, must we still treat attackers equally to victims? If so, do we need to worry about the QALYs of jail, too? Just glancing at my local laws, giving someone a black eye out of nowhere looks like it can earn up to a year in jail. I vastly prefer published sentencing guidelines and fair trials over even heat-of-the-moment vigilante "justice", but if we're only comparing punishment magnitudes, at a QA ratio of .5 for jail time that year is already 1.5 OOM worse than an expected risk of "~a week". Is that still too harsh, except in the unlikely case every jailing manages to deter dozens of potential assaults? Or is 0.5:0.02 now a reasonable punishment ratio, nothing like that crazy 11:0.02?
I would happily accept even a certainty of being killed if I ever punched someone hard enough to knock them off their feet or concuss them, for even a slight deterrence of the possibility of being a sudden battery victim at that level myself. If it was made a clear societal expectation, I would expect this to generalize to others too, at least anyone who's a moral agent capable of being deterred. The probability you're not multiplying in here is "what are the odds that I might innocently do this awful thing by accident", which is itself orders of magnitude below 1 or should be. Knowing that you might die for doing something that might murder someone is a worthwhile deterrence effect, because a decade of the life of a human who can't be dissuaded from risking others' lives so easily is not necessarily even a positive! Violence tends to escalate, especially from someone who doesn't consider its consequences. Escalating an argument to battery that might murder an innocent person is quite bad, and most of the badness is an externality, so unless much of that badness is internalized, the distribution of consequences does not generally give the perpetrator the right incentives. You might still value the life of a battery perpetrator and their victim on exactly the same scale ... but clearly the perpetrator does not.
(The "clear societal expectation" here is as important as the "guidelines and fair trials" bit above; the less/more clear a consequence is, the less/more you can conclude about the future danger of a person who risks it. That 22yo is currently likely to have grown up watching action movies where the hero gets their head bashed in every ten minutes and walks it off. That shooter should have known he'd be jailed unless he could show it was self-defense and not retribution, and in the latter case I'll be happy he's not on the street still either.)
I'm still waiting on some outstanding questions from earlier. Can we escalate from handguns to long arms? From bludgeoning weapons to guns? What is your risk level at which the "on switch" goes on, if it's so clearly off at 0.1% and so clearly on by 11%? How does the victim know his attacker is unarmed, rather than merely stunning him before losing the element of surprise by drawing a weapon? How can it be reasonable to expect the victim of a sudden attack to the head to rapidly yet carefully predict all the possible outcomes of fighting back or not? Why is the attacker, who hasn't been similarly impaired and has had all the time he wanted to think ahead, not to be held equally if not more responsible for predicting the risk of the outcome that he initiated and was entirely capable of avoiding?
Consider my previous comment as applying entirely to unprovoked assaults as well.
Please don't imply I prefer a world with more fist fights, when I obviously don't.
Sure? Though obviously the cost of achieving that world is important to consider.
Sure? Alas, that's not what we're considering.
Good on you for being open minded. I personally don't really care about the specifics of a random particular case.
Because when we're weighing pros and cons and you discussion partner ignores a 2-3 OOM factor, perseverating on a dramatically less important factor is not helpful.
To answer your question, I'm unsure if I would kill the button-pushing sociopath, but note that advocating killing him merely means accept a 2x decreased value of his life. Still far and away from the 2-3 OOM factor I've been harping on.
Moreover, in the minds of most people, there's an enormous difference between someone pushing a button with a specific probability of killing someone purely for the thrill of killing someone with a magic button and a drunkard throwing a punch at someone he views as disrespecting him. So even if you believe our button-pushing sociopath's life has 0 value, I don't know why you would infer the punch-throwing drunkard's life has no value. That's an enormous jump so, no, we have not "solved the problem" with your thought experiment - you've simply replaced the hard problem (is it acceptable to shoot drunks who punch people) with an easy problem (is it acceptable to shoot sociopaths who push magic buttons that kill people), all while refusing to actually grapple with the fact that you advocate cutting 10 years of life from someone in expectation because of one dumb drunk decision at a bar.
No? You seem to think I am a blind utilitarian calculator. I'm not. But when utilitarianism says the costs outweigh the benefits by 10ish QALYs to ~0 QALYs and the benefit is a sense of fairness/justice in a brawl in a bar some random night... well, that seems like a pretty easy question for me to answer.
It seems silly to discount someone's life by 100x merely because the threw an unprovoked fist some night. It seems silly to think such a policy change would reduce fist fights by 2-3 OOM. It seems silly that fairness in a bar fight weighs more than 10 years of human life.
What's so obvious about this? You're arguing for a world with less deterrence for unprovoked physical assault and starting fist-fights. You want to make this activity less risky and less dangerous, and the obvious conclusion from that is that you actually do want a world with more fist-fights. What exactly is your position if you're not arguing for a world with more fist fights and unprovoked assaults as opposed to a world with less of these but more self-defence killings? That's a position that one can agree or disagree with, but "I want to make it easier to assault people and start fist fights and reduce the negative consequences of doing so, but I don't want more fist-fights to happen" really isn't.
I've been arguing this entire time that I don't think the trade-off of marginally fewer fist fights is worth the trade-off of more "self-defence" killings. This does NOT imply that I prefer a world with more fist fights ceteris paribus.
As an example: wearing seatbelts probably cause people to have more car accidents via risk compensation. Me supporting people wearing seatbelts does not imply I want more car accidents.
More options
Context Copy link
More options
Context Copy link
I didn't intend to imply you did; I apologize if it came off that way.
That was the response I expected, which is why I continued with:
So before you complain about the mote in my eye:
Undelete that and all the other outstanding questions; there's a beam in yours.
My position is that there's a level of unjustified lethal force, perhaps on the order of p=1-in-1000 risk, such that a victim reducing even similarly minimal subsequent risk is worth perhaps a q=1-in-10 risk of lethality against the aggressor. I agree that 2OOM is a high ratio. I admit it might be morally too high, or (see the split thread) it might be factually too low to apply to a beating vs a gun, or perhaps both. But I think "some ratio exists, it might even be as high as 100, and the more avoidable the aggression is the higher the ratio should be" is a more philosophically defensible position than deleting attempts to clarify a position. Fists kill people; the idea that a gun is qualitatively rather than quantitatively different is just philosophy-via-rounding-error.
Yup, there's the rounding error: (x-xtrue)/xtrue = 100%. Even the FP8 people manage to keep some operations under 10%.
So, how high a percentage of the time would such attacks have to kill people before you'd stop rounding to zero? Or conversely, how low a percentage of the time would the response have to be lethal before you started? Do brass knuckles count? That bike-lock-swinging guy from protest season? A protruding car key? Maybe it's all doomed to degenerate into hunches and culture clashes, but "don't suddenly attack people" seems so much easier to turn into a bright line rule than "you can attack them, a little lethally, and they can attack you, but nobody start doing it too too lethally, if you get me".
This is a good point. I hinted at "22 year olds are just dumb" as a reason to be wary of excessively punishing their mistakes; there but for the grace of God and all that. "Drunk people are just dumb" might not be quite as strong a reason, since here I suspect "if you're going to be violently dumb when drunk, teetotal" is the dead-body-minimizing solution, but maybe "drunk and 22" brings that solution out of reach.
Doesn't nearly everybody? Suppose the killer in this case had been pissed off enough to brandish and aim his gun before even being provoked by an attack? It would be self-defense to kill him first, no? Would we want to forbid that because it would be cutting off his life from one dumb drunk decision? And there are thousands of drunk drivers (and thousands of their victims) who die each year in the US; not individually so high in expectation but boy do the numbers add up fast. The laws of physics themselves frequently apply the death penalty to people who make one dumb drunk decision at a bar, and we don't even bother to try to suspend the sentence with breathalyzer tests at the bar parking lot exits.
Which I addressed with
You ignoring me grappling with an issue ≠me not grappling with the issue. Motes and beams, indeed.
Yes, I understand that, and you based this on your thought experiment of a sociopath with a magic button, which I addressed and rejected (see above).
Would you have preferred 0.1 QALYs? I admit I feel like that is quibbling. Just replace my "0" with "0.1" and literally everything I say continues to follow through. The point is that the consequences of "don't escalate to guns" are far better than "do escalate to guns". Since the consequences are enormously better (e.g. 10 years of human life), I expect an argument for why escalation is commendable (or should be legal) to offer something of similar value. This "rounding" is not central to my argument in the least.
I don't have a specific numerical answer, but I don't need one, because the answer is definitely not 2-3 OOM.
How about just "try running away before shooting"? But, I don't concede that a brighter line outweighs the expected loss of life.
Agreed, but alas the framing of this conversation isn't what to do if you are omnipotent, but what actions / laws we should advocate for.
Again, there is a gradient here. Most things in life are on a gradient. We can't just ignore the gradient, because it makes decision-making simpler.
Ignoring all the bits here and there, let me give you my own thought experiment that I think illuminates my intuition:
Suppose you have two sons: Bob and Dan. A genie comes before you and say
Which would you wish for? Is Dan's honor/fairness/safety-from-fists more important to you than Bob getting shot? [Edit: Which society would you want children to grow up in? ]
My fault here was not ignoring "Still far and away from the 2-3 OOM factor I've been harping on." That seemed to imply that you had just skimmed past the 2-3-OOM-factor case, that your "specific probability" paragraph was still just referring to the previous case. A straight "no, this time I'm definitely just crossing my fingers for Suzie" (or even "probably no but I'm still unsure") to the latter case would have been clearer.
Indeed! There's still several unanswered questions in this thread I'd originally been hoping to hear your answers to. Even if the answers were all just "I'm unsure"; it would have been a step up from just dropping them. The vehemency with which you were certain that 2-3OOM is excessive punishment made me assume that you had a more precise moral code than my flippant speculations, one which would thus have wider applicability. If your answer is just "the ratio is below 100 and maybe below 2 but I can't get any closer than that" then that'd be terse and honest and it might be correct, but it's also as wide a range of uncertainty as I've got and so it's much less educational than I'd hoped. This would have been an incredibly widely relevant sort of moral rule if it could be pinned down more clearly.
In what equivalent sense am I refusing to grapple with any issues? Was there a question or ten I missed answering? Was any of my own uncertainty underspecified to any greater extent than your range?
My primary concern at the start of this wasn't that I hadn't noticed
... because I was pointing out the gradient here! The post I replied to mocked the idea that a punch could be as scary as lethal force, so I pointed out that punches are frequently lethal force. Maybe they're not lethal enough to treat as lethal force, since we have to draw a line somewhere, but talking as if we've found bright-line categories by drawing a line in between "zero" and "positive" is just a way of evading the fact that it makes us much less comfortable to have to draw a line through a gradient in between some x and 1.001 x.
In this particular case? Prosecutors say the deceased was "tussling" with the shooter's friend when the shot was taken. That probably makes the shoot even stupider but it also probably morally precludes the shooter just running away.
Very good advice in most similar cases, though, I'm sure. Hell, even in cases where the shooter can't run away it's probably often worth it to try; e.g. the first thing that made me suspect Rittenhouse was being misdescribed was the video of him trying his best to disengage before defending. Accepting an increased risk of your own death in an attempt to reduce that of your assailants can be a noble thing to do even when supererogatory.
This is a very good thought experiment; I think it clarifies my moral intuition much better. Imagining someone I care about rather than myself does make the "just don't start shit" option seem too much less certain, so in the drunken-bar-fight case I'm hoping for Bob to remain unshot even if Dan got beaten because of those norms, even after I try to correct the implicit one-to-one ratio in your version of the thought experiment to something more realistic.
But as you originally posited, I don't think it's the lethality ratio that's motivating me here. If we change the scenario to something where the lethality is of the same magnitude but the root cause is pure sociopathy rather than pure drunken stupidity, I'm then okay taking the other tradeoff. Suppose Bob played "the knockout game" on someone? A punch is still just a punch, and if he gets shot because of it I'm going to be distraught, but I'd also be thankful that the "game" stopped and it's safe to walk down the street again, and I'm going to blame my awful son and my awful parenting, not the shooter. Conversely, with lethality increased but the root cause still just impaired stupidity, I'm hard pressed to morally criticize. If either Bob's victim or Dan (or Cranston) was drunk or punch-drunk and took a shot, not because it was a good shoot but because he had no idea what was going on and couldn't distinguish "fistfight that ends in bruises" from "attacker out for life and limb", I'm not going to be too upset with the shooter for being mistaken, and even if my equanimity just means I regret us having to put him away for manslaughter I'm certainly going to think further escalation is excessive. There's a reason why "26 seconds" was a headline-grabbing claim in this Cranston case; that's not even as much time as it takes to write a brief comment here, but every extra second decreases the odds that the shooter thought defense was a necessity rather than an excuse.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
Whether the death happens to a victim or a perpetrator is not "less important". It's more important than just about everything else relevant to the situation.
Only if utilitarianism doesn't distinguish between QALYs for a perpetrator and a victim. Also, using QALYs here at all produces bizarre results because it becomes much less bad to kill an older perpetrator than a younger one.
Tossing a punch at someone is an attempt to kill, or a reckless act that may kill, and should be treated as such.
You've made an assertion. Not an argument.
If I had two sons, and one son got drunk and punched someone at a bar while another got drunk and was punched by someone at a bar. I would not want to live in a world where the former was killed and the latter killed their assaulter. I'd much rather live in the alternative world where no one died. Which would you rather live in?
Again, I am not a blind utilitarian calculator. It is a model.
This is black-and-white thinking. There are gradations here that you are ignoring, because they are inconvenient to you. Those gradations are central to my argument, so I'm not sure how I'm supposed to respond here.
This argument proves too much. If most people had a son and their son tried to kill someone, they would prefer that the son not be killed in self-defense at all. If you're taking this argument seriously, it doesn't actually matter how high the chance of death from the lethal attack is, killing in self-defense is wrong, period.
Also, you should consider it from behind the veil of ignorance (not the actual version of veil of ignorance, the popular version): If A is lethally attacking B and you don't know whether A or B is your son, would you prefer that B be able to kill A in self-defense even if the lethality isn't too high? I would, and I think most people would.
I agree.
If Man A punched Man B and I knew one of them was my son but not which, I would pray to every god under the sun that B didn't shoot at A.
Why? Remember, we're assuming self-defense. If B shoots A and A is your son, your son is a murderer who was shot in self-defense. If B doesn't shoot A and B is your son, then your son is a murder victim, with some probability. Surely you'd be more concerned for your son's survival if he's innocent than if he's a murderer, so if you don't know whether he's A or B, you'd prefer the scenarion where B shoots.
Unless you're saying that you don't want B to shoot A because your only concern is reducing the chance of death, and it doesn't matter who started it. Then this proves too much and implies that you oppose lethal self-defense, period. Do you?
No? He was a drunkard who punched someone and was killed. Not a murderer.
"with some probability" is don't a lot of work there.
It matters, but not as much as my son stay alive.
It does matter, but not nearly as much imo as it does to you.
If a unhinged man is holding a dozen people hostage with a gun, take him out if you can. I'd support this even if there were a 1-in-13 chance the hostage-taker was my son. If a psychotic man has a knife and is charging you, and you can't get away, shoot him.
If one man sucker punches another in public, don't shoot. I'd support this even if there were a 1-in-2 chance the punched man was my son.
There are scenarios where there are degrees of uncertainty regarding the correct action - parameters that influence this include
who started it
how much effort was made to escape
how much each party escalated prior
discrepancies in overall strength of the parties
presence of peers
distance from security/police
etc
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
I think this is an instance of causal decision theory in the wild, in that you're holding the punch stable when there's no reason to expect that to be the case. What if it being "the sort of world where people who throw punches are killed" means that instead you get to pick between the world where your sons punch and are punched, and the world where nobody is even punched? Then the question would be to what extent punch-kill actually allows acausal flow, right? Ie. we may imagine a world where some people just, out of the blue, are struck by the urge to punch and otherwise-agentically seek out a target to punch. In that case, the kill-branch obviously would only worsen the situation. So the question comes down to if the punch urge is such that the kill branch can successfully shift the incentives enough to suppress the punch branch enough to make up for the QALY loss.
Because at the end of the day, we'd at least somewhat prefer that the least people die. Right?
I agree with all of that. Personally, I think the impact of right-to-shoot on number-of-fist-fights would be very small, because I don't think most fist-fights are based on a rational weighing of pros and cons, which is why I didn't center it in the thought experiment. If you'd prefer, try this one:
You have three sons: Bob, Dan, Frank, and Ivan. They go out to bars and get drunk.
A genie comes up to you and tells you
Bob punched someone.
Dan got punched.
You can choose one of two futures for your wish
Bob got shot. Dan shot his assaulter. Nothing happens to Frank or Ivan.
Bob was not shot. Dan did not shoot his assaulter. Frank also punched someone (not shot) and Ivan was also punched (didn't shoot).
There, now the thought experiment bakes in an assumption that letting-people-shoot reduces fist-fights by 50%. As their parent, which would you wish for? I know what I would.
I think this crucially depends on the death rates from punching vs shoot-to-fist ratio. Also I don't think fistfights are rational, but getting into a situation where a fistfight may ensue is absolutely rational. If you look at for instance duels, IMO a society where getting shot is at risk can develop alternate ways of mediating the sorts of situations that otherwise become fistfights. This is becoming really hard to model - but I don't think that guns are limited to having a 50% reduction on fistfights. If they were, they'd probably be a bad trade on utilitarian grounds alone, though there may still be other cases for them.
Agreed, and based on the evidence this seems like a 2-3 order-of-magnitude difference, which is why I've been arguing its crazy to advocate for shooting. To argue otherwise on utilitarian grounds requires claiming that for every 1% decrease in shooting-people-who-throw-punches, the number of punches throw grows by 100-1000%, which seems patently absurd.
And we can modify the thought experiment: your extended family of ~50 people go to a bar and drink. Would you prefer they all come armed with a willingness to escalate a punch to a shot? I wouldn't. I have a hard time believing many people would.
Once someone actually cares about the people involved, it seem clear that they are much less gung-ho (pun intended).
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link