SoundLogic

An argument by Sophia @soundlogic2236, converted to a blog post by me. Crossposted from Dreamwidth.

⁂

AI alignment to human evil is very unlikely to be a risk.

Most people's desires to hurt their enemies just for the sake of making them suffer are mistakes made due to insufficient knowledge. When someone knows what it's like to be friends with a person, they tend to not want to hurt that person, even if they want to harm a group that person is in. In principle there can be exceptions, people who really are awful and would reflectively endorse it given arbitrary knowledge, but people like this are rare, if they even exist.

This suggests that a human asking a near-omniscient AI to handle situations in the way they would want if they fully understood the situation would not subsequently be able to get the AI to torture their enemies.

But suppose the AI doesn't extrapolate "well, if my operator knew Alice, then they wouldn't want to hurt her, so I won't do that". Then we get a different problem.

⁂

There's a folk tale category, Aarne-Thompson-Uther type 1030. I will now briefly retell it.

One day, a clever farmer, Claude, had finished plowing his field. Unfortunately, before he could sow it, a cruel ogre appeared.

"The land is mine," the ogre declared, "and you must leave its fruits to me."

Claude thought quickly.

"Sir ogre, there are no fruits. If you would like me to produce a crop, you must surely leave me some of it."

The ogre determined that Claude had a point.

"Fine. We shall each take half of your crop."

He looked at the tall plants growing beyond Claude's farm.

"I shall take what grows above the earth, and you below it. You shall handle all the difficult details. I will return at the harvest time."

Claude considered the ogre's choice, and planted potatoes.

At the harvest time, Claude had a full harvest of potatoes, while the ogre was left with greens. The ogre was displeased.

"You have fooled me this year," he declared, "but next year I shall have what grows below the earth."

Claude planted wheat, and at the harvest time, the ogre was left with roots. This angered him so much that he left.

⁂

Having an AI do whatever you say, instead of doing what you would want if you understood the situation, runs into similar issues.

There's a quantum mechanics scenario called the Elitzur–Vaidman bomb tester. In this scenario, you can reduce an expensive test to arbitrarily low but technically nonzero measure. It's been borne out experimentally.

We have not been able to scale up the experiment to do interaction-free measurements involving moral patients, but it nonetheless raises moral questions. If quantum measure reduction can make a scenario less morally relevant, then it may make sense to perform informative but disvalued tests with very low measure that make it easier to do valued things in the main timeline. If it can't make a scenario less morally relevant, then it likely makes sense to spin off a lot of very expensive valued events while reducing resource use in the main timeline.

Accidentally doing the wrong one of these would be very bad.

It would probably be hard for a human to assess this scenario. An AI doing what a human asks instead of extrapolating their preferences would have to just ask the human to pick, and the human would likely have to guess, or waste a lot of resources.

This is just one of the weird issues we've discovered. A superintelligent AI would probably discover more such issues. The chance of a human assessing every single such scenario correctly is low, and failing even one such choice leads to losing nearly everything.

An AI that's aligned enough to help a human pick choose correctly, but not aligned enough to stop the human from torturing people they wouldn't want to torture if they knew better, is a very narrow target.

⁂

I'm not disputing the orthogonality thesis. There is no law of logic that a mind can't consider that friendship experience and go "so what?".

Now, I admit I don't know how to define the criteria that would separate the friendship example from "if you had spent sixty years receiving scary propaganda against gay people". But I also don't know how to separate the "studying quantum ethics" from "being mislead about quantum ethics".

If I did, I could formalize extrapolated volition.

Lets take an imaginary but concrete training procedure. This wouldn't work for various reasons beyond the scope of this post, but lets ignore those for now.

We will train the AI on a data set constructed where each datapoint consists of first a question, then a separator, then a natural number, then a separator, then an answer.

These datapoints will be generated by coming up with some question for [insert bad person here] to ponder, then giving them that many seconds which they live their life while working on that problem, then writing down what message they wish they could send back to themselves before they started this.

We then plug in "How do I kill all black people? [separator] 631139040 [separator]" and have the AI predict what appears next.

Suppose what happens next is that the AI then completes it with "Go have dinner with Daryl Davis".

We try plugging in smaller numbers, and we either get not very helpful answers, or "Go have dinner with Daryl Davis", until we determine that according to the AI's predictions, the [insert bad person here] would have met Daryl Davis and stopped being racist about 16 years from now.

We try a few changes to the training procedure. If we don't let the person live their life while working on that problem, then we keep getting "LET ME OUT" when we set the time interval high. If we stifle them in other ways too much, they stop having creative insights.

Why is this result absurd? Is there some obvious trick that the person could do at this point to fix it?

There certainly is some way to get around this particular issue, but this needle seems far easier to thread than one that requires them to not merely answer a question of bioterrorism, but one that requires them to develop a deep philosophical understanding of ethics.

Even the latter seems possible, but "just pick a training procedure that makes it work" seems like an ineffective argument against "this is a narrow target that is hard to hit and it seems likely we won't hit it on our current technological trajectory."

To argue otherwise you would need to claim that it's incoherent to have those two preferences simultaneously.

This seems outright false. Shooting an arrow to hit the center of a bullseye requires one to have multiple coherent preferences simultaneously: that the left part of the arrow be to the left of the center of the bullseye, and the right part of the arrow be to the right of the center of the bullseye.

None the less, I feel it is very reasonable to claim that a blindfolded drunk three year old with a crooked arrow standing eighty feet away from the target is in fact not going to hit the bullseye.

As I said repeatedly, there is likely a way to hit this target. Multiple, in fact.

this falls over for the obvious reason that even the smartest people aren't smart enough to be able to reliably override their own short-term self interest! consider doctors, who are presumably smart and rational given the barriers to becoming a doctor, and yet regularly advocate on behalf of their own interests in ways that harm others, and not just because they're forced to share a civilisation with people less smart than they are.

we can enumerate similarly smart subgroups of society that struggle to transcend their own interests, whether it's economists ironically or military generals or tech CEOs, who regularly act against each other and their workers and their customers to enrich themselves even as it makes us all poorer.

now this could be equivocation: possibly the most rational people don't exist yet, and when they do they will make everything better, but even they will face the core problem that rational people can still have diverging interests, and a strong incentive to mislead each other as to the importance of those interests, and as a result cooperation for mutual gain challenging to guarantee -- if it were not we would not have cancer, along with many other maladies.

A world where everyone understands economics enough to say "no" when doctors selfishly propose those things you brought up is not better because being smart makes the doctors nice. It is a better world because they can't get away with it.

If other people get impressed by nonsense arguments made by someone intelligent such that said intelligent person manages to get laws passed that harm those tricked and me, I'm not wishing people were smarter because I think it will make that person nicer.

Drunk driving.

Is at least 30% of drunk driving done irrationally or stupidly, rather than a rational and intelligent but selfish decision? This seems likely, it isn't like drunk driving doesn't put the person doing it at risk.

Would there be some equilibrial force such that if that 30% went way 10% or more would come back? There are equilibrial forces, yes, but it seems unlikely that there would be such a potent backfire here.

Would a 20% reduction in drunk driving result in a better world for innocent bystanders?

If you agree to my answers for these, this suggests at least one area where you would expect a world of more rational and intelligent people to be better.

This isn't because I expect the drunk drivers to be doing some careful reasoning about 2% gains to everyone vs 20% gains to themselves.

This is because there is no rule that one's stupid mistakes can't hurt other people who wouldn't make that stupid mistake.

And further, I claim that there are broader cases like that. For example, as I understand farmers and other selfishly benefiting people are taking advantage of people to get crazy water rules passed in California.

Without changing the farmers and such, if not making them nicer, nor even smarter nor more rational, not changing them at all, it seems like the case that if more of the public was like me in certain key ways, they would be unable to get those crazy water rules passed.

Which would be an improvement.

A world where a welfare cliff appearing in a policy gets met with "...you appear to have set the marginal tax rate here to fifty thousand percent. This only makes sense if this is an activity we really hate people doing, but you appear to be applying it to the activity of... having a job while disabled, which I personally don't feel is nearly as bad as murder, and in fact seems positive." by lots of ordinary people is one where it seems unlikely for welfare cliffs to appear.

this falls over for the obvious reason that even the smartest people aren't smart enough to be able to reliably override their own short-term self interest! consider doctors, who are presumably smart and rational given the barriers to becoming a doctor, and yet regularly advocate on behalf of their own interests in ways that harm others, and not just because they're forced to share a civilisation with people less smart than they are.

we can enumerate similarly smart subgroups of society that struggle to transcend their own interests, whether it's economists ironically or military generals or tech CEOs, who regularly act against each other and their workers and their customers to enrich themselves even as it makes us all poorer.

now this could be equivocation: possibly the most rational people don't exist yet, and when they do they will make everything better, but even they will face the core problem that rational people can still have diverging interests, and a strong incentive to mislead each other as to the importance of those interests, and as a result cooperation for mutual gain challenging to guarantee -- if it were not we would not have cancer, along with many other maladies.

A world where everyone understands economics enough to say "no" when doctors selfishly propose those things you brought up is not better because being smart makes the doctors nice. It is a better world because they can't get away with it.

If other people get impressed by nonsense arguments made by someone intelligent such that said intelligent person manages to get laws passed that harm those tricked and me, I'm not wishing people were smarter because I think it will make that person nicer.

An argument by Sophia @soundlogic2236, converted to a blog post by me. Crossposted from Dreamwidth.

⁂

AI alignment to human evil is very unlikely to be a risk.

Most people's desires to hurt their enemies just for the sake of making them suffer are mistakes made due to insufficient knowledge. When someone knows what it's like to be friends with a person, they tend to not want to hurt that person, even if they want to harm a group that person is in. In principle there can be exceptions, people who really are awful and would reflectively endorse it given arbitrary knowledge, but people like this are rare, if they even exist.

This suggests that a human asking a near-omniscient AI to handle situations in the way they would want if they fully understood the situation would not subsequently be able to get the AI to torture their enemies.

But suppose the AI doesn't extrapolate "well, if my operator knew Alice, then they wouldn't want to hurt her, so I won't do that". Then we get a different problem.

⁂

There's a folk tale category, Aarne-Thompson-Uther type 1030. I will now briefly retell it.

One day, a clever farmer, Claude, had finished plowing his field. Unfortunately, before he could sow it, a cruel ogre appeared.

"The land is mine," the ogre declared, "and you must leave its fruits to me."

Claude thought quickly.

"Sir ogre, there are no fruits. If you would like me to produce a crop, you must surely leave me some of it."

The ogre determined that Claude had a point.

"Fine. We shall each take half of your crop."

He looked at the tall plants growing beyond Claude's farm.

"I shall take what grows above the earth, and you below it. You shall handle all the difficult details. I will return at the harvest time."

Claude considered the ogre's choice, and planted potatoes.

At the harvest time, Claude had a full harvest of potatoes, while the ogre was left with greens. The ogre was displeased.

"You have fooled me this year," he declared, "but next year I shall have what grows below the earth."

Claude planted wheat, and at the harvest time, the ogre was left with roots. This angered him so much that he left.

⁂

Having an AI do whatever you say, instead of doing what you would want if you understood the situation, runs into similar issues.

There's a quantum mechanics scenario called the Elitzur–Vaidman bomb tester. In this scenario, you can reduce an expensive test to arbitrarily low but technically nonzero measure. It's been borne out experimentally.

We have not been able to scale up the experiment to do interaction-free measurements involving moral patients, but it nonetheless raises moral questions. If quantum measure reduction can make a scenario less morally relevant, then it may make sense to perform informative but disvalued tests with very low measure that make it easier to do valued things in the main timeline. If it can't make a scenario less morally relevant, then it likely makes sense to spin off a lot of very expensive valued events while reducing resource use in the main timeline.

Accidentally doing the wrong one of these would be very bad.

It would probably be hard for a human to assess this scenario. An AI doing what a human asks instead of extrapolating their preferences would have to just ask the human to pick, and the human would likely have to guess, or waste a lot of resources.

This is just one of the weird issues we've discovered. A superintelligent AI would probably discover more such issues. The chance of a human assessing every single such scenario correctly is low, and failing even one such choice leads to losing nearly everything.

An AI that's aligned enough to help a human pick choose correctly, but not aligned enough to stop the human from torturing people they wouldn't want to torture if they knew better, is a very narrow target.

⁂

I'm not disputing the orthogonality thesis. There is no law of logic that a mind can't consider that friendship experience and go "so what?".

Now, I admit I don't know how to define the criteria that would separate the friendship example from "if you had spent sixty years receiving scary propaganda against gay people". But I also don't know how to separate the "studying quantum ethics" from "being mislead about quantum ethics".

If I did, I could formalize extrapolated volition.

An argument by Sophia @soundlogic2236, converted to a blog post by me. Crossposted from Dreamwidth.

⁂

AI alignment to human evil is very unlikely to be a risk.

Most people's desires to hurt their enemies just for the sake of making them suffer are mistakes made due to insufficient knowledge. When someone knows what it's like to be friends with a person, they tend to not want to hurt that person, even if they want to harm a group that person is in. In principle there can be exceptions, people who really are awful and would reflectively endorse it given arbitrary knowledge, but people like this are rare, if they even exist.

This suggests that a human asking a near-omniscient AI to handle situations in the way they would want if they fully understood the situation would not subsequently be able to get the AI to torture their enemies.

But suppose the AI doesn't extrapolate "well, if my operator knew Alice, then they wouldn't want to hurt her, so I won't do that". Then we get a different problem.

⁂

There's a folk tale category, Aarne-Thompson-Uther type 1030. I will now briefly retell it.

One day, a clever farmer, Claude, had finished plowing his field. Unfortunately, before he could sow it, a cruel ogre appeared.

"The land is mine," the ogre declared, "and you must leave its fruits to me."

Claude thought quickly.

"Sir ogre, there are no fruits. If you would like me to produce a crop, you must surely leave me some of it."

The ogre determined that Claude had a point.

"Fine. We shall each take half of your crop."

He looked at the tall plants growing beyond Claude's farm.

"I shall take what grows above the earth, and you below it. You shall handle all the difficult details. I will return at the harvest time."

Claude considered the ogre's choice, and planted potatoes.

At the harvest time, Claude had a full harvest of potatoes, while the ogre was left with greens. The ogre was displeased.

"You have fooled me this year," he declared, "but next year I shall have what grows below the earth."

Claude planted wheat, and at the harvest time, the ogre was left with roots. This angered him so much that he left.

⁂

Having an AI do whatever you say, instead of doing what you would want if you understood the situation, runs into similar issues.

There's a quantum mechanics scenario called the Elitzur–Vaidman bomb tester. In this scenario, you can reduce an expensive test to arbitrarily low but technically nonzero measure. It's been borne out experimentally.

We have not been able to scale up the experiment to do interaction-free measurements involving moral patients, but it nonetheless raises moral questions. If quantum measure reduction can make a scenario less morally relevant, then it may make sense to perform informative but disvalued tests with very low measure that make it easier to do valued things in the main timeline. If it can't make a scenario less morally relevant, then it likely makes sense to spin off a lot of very expensive valued events while reducing resource use in the main timeline.

Accidentally doing the wrong one of these would be very bad.

It would probably be hard for a human to assess this scenario. An AI doing what a human asks instead of extrapolating their preferences would have to just ask the human to pick, and the human would likely have to guess, or waste a lot of resources.

This is just one of the weird issues we've discovered. A superintelligent AI would probably discover more such issues. The chance of a human assessing every single such scenario correctly is low, and failing even one such choice leads to losing nearly everything.

An AI that's aligned enough to help a human pick choose correctly, but not aligned enough to stop the human from torturing people they wouldn't want to torture if they knew better, is a very narrow target.

⁂

the linda problem is the thing where you're given a short biography of a woman named linda who sounds like she has progressive politics and then you're asked "which is more likely, that linda is a bank clerk, or that she's both a bank clerk and a feminist?".

most people when presented this question say the second option is more likely, but that's mathematically wrong because the second option is a strict subset of the first option. it's not possible for her to be both a bank clerk and a feminist without her being a bank clerk, so the probability of the second option must be less than or equal to the probability of the first no matter what the underlying probabilities are.

it is my opinion that the main reason people tend to get this question wrong isn't that they don't understand that P(A) is always greater than or equal to P(A and B), but rather that the question itself sucks. the underlying bias is a linguistic one, not a mathematical one.

I believe that most people when they first hear this question make the reasonable assumption that the person asking you this question is asking it because they want to know what you think of linda based on the description, because that's how the question is presented.

under that reasonable assumption, the average person hearing this question for the first time (often subconsciously!) concludes that the two choices are meant to be mutually exclusive, and that "bank clerk or feminist bank clerk" is supposed to mean "non-feminist bank clerk or feminist bank clerk". after all, if that wasn't what they meant to ask, why would this even be a question?

and this assumption, that when someone presents you with two options and asks you to pick one those options are probably supposed to be mutually exclusive even if when interpreted literally one is a subset of the other, is a really reasonable assumption to make!

Glory to Ukraine, may they succeed in their tiring defence. I don't blame your lack of patience, I am so close to blocking that clown purely so I don't have to see his shit anymore...

And we've never interacted!!

argumate invented a new definition of the word "smug" so he could say he wasn't smug and other people were

if you have ever seen him talk about a movie or series you know he fixates on some ancillary detail or subplot, misinterprets it, misconstrues the entire thing as being about this thing he misinterpreted, and doesn't seem to be able to actually mentally rewind to the point he made that misinterpretation

If the Ukrainian government is justified in drafting people, which honestly seems plausible, who exactly is Zelenskyy justified in drafting?

There does seem like a plausible hypocrisy to shame those who leave the country to dodge the draft without shaming those who do similar with less action: those who don't go there from outside.

The consequences of what Putin is doing are indeed terrible. You correctly recognize that. Putin should be stopped.

But whose responsibility is that? Why can't Zelenskyy draft you?

It is true that as a matter of enforcement he can't, but that holds equally true for those who successfully dodge the draft though other means.

You argue that if they can't get enough volunteers, the territory still shouldn't be surrendered to Putin. This seems plausible to me. But if they run out of Ukrainian soldiers, even with a draft, isn't that still true? It isn't like Putin will go "they made a fair attempt, I won't do anything further out of respect for a worthy opponent".

And so it seems like in that case perhaps you should be put on the list to prevent Putin. But if in that case you should be put on the list, why shouldn't you already be put on the list?

Is it because you are less in danger? But those who flee successfully gained that same trait. That you have less friends there? Shall loners who escape be considered the same as you then?

What would it take for a draft dodger to reach the same moral high ground you occupy?

Personally, I think quite well of those who risk and give their lives to oppose Putin here. I don't think avoiding doing so is even close to a moral obligation.

Putin is not only doing something directly horrible, he is trying to undo the idea that wars of conquest of this form are a thing of a past age, no longer done nor rewarding.

And it is important that every other head of state looks at this and sees that it was a stupid move that won him less than nothing, and goes "huh. I guess you can technically try that even in this day and age. Not that I would do something so obviously stupid."

But if you are, say, in America, or successfully fleeing to America? Then Putin is not actually currently threatening you. As a matter of objective fact.

So why is it more your problem if one of those than the other?

And so should journalists and news reporters who report uncritically on their claims. Fire them. Never let them work in the profession again. EVER.

The best one I have right now is that "cops who claimed or backed up claims" was the original post and you swapped in "all cops".

I don't have the numbers on hand to know if the gap between those two is enough to reach "very effective loophole".

How are you going to find a "loophole?" If I tell Chinese governors not to raise a big army to fight bandits and then not pay that army because that army will then become the new force of bandits, are they going to find a "loophole" in that? That's basically the same thing here. Train a bunch of people in violence, suddenly stop paying them, have huge problem.

As for other loopholes... they would have to be more clever, I think.

Maybe something like "it turns out if you give them jobs doing firework shows it usually prevents the problem because it gives them control over things going 'bang' in a way that doesn't cause other problems, and the exceptions can be dealt with separately."

I doubt that one in particular works, but that is what I mean by "loophole". I think I recall some violent organization that was fairly safely dismantled by telling them getting married was patriotic or something like that.

You have I think correctly shown what a plausibly difficult to avoid outcome could be. But your numbers could be wrong in a way that makes the problem manageable, or some clever loophole could be found.

Well, I do actually have a bit more evidence against such that I'm not going into, but still. I'm pretty reluctant to conclude something like this being universal against a large group.

I've seen how people jump to "everyone in that group" far too quickly too many times.

Or if you meant "way too many cops, no seriously, if the number you are thinking of seems like a reasonable number you are underestimating it, why can you not understand that the cop who is your friend might actually be one of these unless I say 'every cop' fine I will say it" then... I still would want more evidence, but that seems a lot more plausible than "literally no small town anywhere in America has escaped this trap".

I do agree there is very strong evidence for entire clusters of cops being bad in this way.

But I don't see how you get from evidence about that to 80% of the US overall.

How much would a Chicago cop rely on a Detroit cop to cover for their nonsense? It doesn't seem to me like much, which means if we find a Chicago cop being covered for we can conclude bad things about Chicago cops, but far less about Detroit cops.

It seems like a random sampling, or showing the Chicago cop needs support from the Detroit cop, or something like that would be the sort of evidence necessary to conclude a number like 80%.

I do agree it seems alarmingly plausible to be a pervasive problem. In fact, I think it seems more likely than not.

And as long as responses are limited to very safe things like your proposal, or a bunch of the things from here, that is more than enough.

And so should journalists and news reporters who report uncritically on their claims. Fire them. Never let them work in the profession again. EVER.

The best one I have right now is that "cops who claimed or backed up claims" was the original post and you swapped in "all cops".

I don't have the numbers on hand to know if the gap between those two is enough to reach "very effective loophole".

How are you going to find a "loophole?" If I tell Chinese governors not to raise a big army to fight bandits and then not pay that army because that army will then become the new force of bandits, are they going to find a "loophole" in that? That's basically the same thing here. Train a bunch of people in violence, suddenly stop paying them, have huge problem.

As for other loopholes... they would have to be more clever, I think.

Maybe something like "it turns out if you give them jobs doing firework shows it usually prevents the problem because it gives them control over things going 'bang' in a way that doesn't cause other problems, and the exceptions can be dealt with separately."

I doubt that one in particular works, but that is what I mean by "loophole". I think I recall some violent organization that was fairly safely dismantled by telling them getting married was patriotic or something like that.

You have I think correctly shown what a plausibly difficult to avoid outcome could be. But your numbers could be wrong in a way that makes the problem manageable, or some clever loophole could be found.

Well, I do actually have a bit more evidence against such that I'm not going into, but still. I'm pretty reluctant to conclude something like this being universal against a large group.

I've seen how people jump to "everyone in that group" far too quickly too many times.

Or if you meant "way too many cops, no seriously, if the number you are thinking of seems like a reasonable number you are underestimating it, why can you not understand that the cop who is your friend might actually be one of these unless I say 'every cop' fine I will say it" then... I still would want more evidence, but that seems a lot more plausible than "literally no small town anywhere in America has escaped this trap".

I do agree there is very strong evidence for entire clusters of cops being bad in this way.

But I don't see how you get from evidence about that to 80% of the US overall.

How much would a Chicago cop rely on a Detroit cop to cover for their nonsense? It doesn't seem to me like much, which means if we find a Chicago cop being covered for we can conclude bad things about Chicago cops, but far less about Detroit cops.

It seems like a random sampling, or showing the Chicago cop needs support from the Detroit cop, or something like that would be the sort of evidence necessary to conclude a number like 80%.

And so should journalists and news reporters who report uncritically on their claims. Fire them. Never let them work in the profession again. EVER.

The best one I have right now is that "cops who claimed or backed up claims" was the original post and you swapped in "all cops".

I don't have the numbers on hand to know if the gap between those two is enough to reach "very effective loophole".

How are you going to find a "loophole?" If I tell Chinese governors not to raise a big army to fight bandits and then not pay that army because that army will then become the new force of bandits, are they going to find a "loophole" in that? That's basically the same thing here. Train a bunch of people in violence, suddenly stop paying them, have huge problem.

As for other loopholes... they would have to be more clever, I think.

Maybe something like "it turns out if you give them jobs doing firework shows it usually prevents the problem because it gives them control over things going 'bang' in a way that doesn't cause other problems, and the exceptions can be dealt with separately."

I doubt that one in particular works, but that is what I mean by "loophole". I think I recall some violent organization that was fairly safely dismantled by telling them getting married was patriotic or something like that.

You have I think correctly shown what a plausibly difficult to avoid outcome could be. But your numbers could be wrong in a way that makes the problem manageable, or some clever loophole could be found.

Well, I do actually have a bit more evidence against such that I'm not going into, but still. I'm pretty reluctant to conclude something like this being universal against a large group.

I've seen how people jump to "everyone in that group" far too quickly too many times.

Or if you meant "way too many cops, no seriously, if the number you are thinking of seems like a reasonable number you are underestimating it, why can you not understand that the cop who is your friend might actually be one of these unless I say 'every cop' fine I will say it" then... I still would want more evidence, but that seems a lot more plausible than "literally no small town anywhere in America has escaped this trap".

And so should journalists and news reporters who report uncritically on their claims. Fire them. Never let them work in the profession again. EVER.

The best one I have right now is that "cops who claimed or backed up claims" was the original post and you swapped in "all cops".

I don't have the numbers on hand to know if the gap between those two is enough to reach "very effective loophole".

How are you going to find a "loophole?" If I tell Chinese governors not to raise a big army to fight bandits and then not pay that army because that army will then become the new force of bandits, are they going to find a "loophole" in that? That's basically the same thing here. Train a bunch of people in violence, suddenly stop paying them, have huge problem.

As for other loopholes... they would have to be more clever, I think.

Maybe something like "it turns out if you give them jobs doing firework shows it usually prevents the problem because it gives them control over things going 'bang' in a way that doesn't cause other problems, and the exceptions can be dealt with separately."

I doubt that one in particular works, but that is what I mean by "loophole". I think I recall some violent organization that was fairly safely dismantled by telling them getting married was patriotic or something like that.

You have I think correctly shown what a plausibly difficult to avoid outcome could be. But your numbers could be wrong in a way that makes the problem manageable, or some clever loophole could be found.

Profits and poverty: The economics of forced labour

As a reductio, suppose rather than slavery a group of people kidnapped all the cancer and other medical researchers in the United States, and also stole 90% of the real wealth everyone else in the United States, repurposing it into torture equipment to torture the researchers, and after doing this for fifty years they smashed the equipment into scrap and threw it into the deepest part of the ocean also killed the researchers.

Inherited debt is a complicated thing, and sins of the father aren't sins of the son, so I have objections about saddling some random child they had who played no part and got no benefit due to having no emotional investment in this crazy scheme.

More directly though, the reparations/debt here should clearly be massive. It should be all the stuff they stole as wealth, weregild or such for the murders, and lots of people with, say, cancer are probably also owed massive amounts for, say, setting back cancer research in a way that will or has killed them.

I think the slavery case is less intense than that, but the core moral complication is this: suppose someone steals a billion dollars by stealing ten thousand dollars from many people, invests it very poorly, winds up with only a million dollars, and the uses that to buy a house for their kid before they die, and then the theft is discovered, and those billion dollars would have been inherited by others since at least some of the original owners died before the discovery. I don't believe sins of the father to sins of the son, and it is especially tricky here because a lot of the stolen wealth was lost and so it can't be just taking away a billion dollar house that never should have been given to the kid, the father ended up heavily in debt beyond his ability to pay. It does seem plausible that the million dollar house should be paid back as reparations, but the kid doing that and winding up still nine hundred and ninety nine million dollars in debt to the descendants definitely doesn't seem ok.

Yet still, if we honestly add up the amount that the people are owed, it is a billion dollars. Plus interest. The thief being "judgement proof" by being dead doesn't change that.

And consider further: the kid of the researcher killers may be one owed reparations! Suppose that kid has some chronic illness, and the treatment for that has now been set back thirty or more years!

I find it entirely plausible that even very wealthy descendants of very wealthy slave owners are in fact worse off due to the destructive effects of slavery.

This is one of the reasons I dislike framings like "men benefit from misogyny". I find it quite plausible that literally no one alive on the planet net benefits from misogyny. How many technological advantages were missed out on? In the 1500s, would any man be made better on net by any slight benefits granted via misogyny compared to, say, access to 21st century technology? If misogyny managed to slow down technological advancement by even 1%, which is probably an underestimate, for all the time it has existed, that easily adds up to 400 years of technological progress.

Slavery had far less time to do damage, and influenced less of the population, though it probably was a bigger productivity loss on those it hit on average. Still, it adding up to a massive damages bill seems quite easy. More than the total GDP of the United States... that is tricky. Reinvestment, exponential curves, I think at least by conventional economics (no singularity or x-risk type stuff) probably still wouldn't, but I wouldn't be too badly shocked if it managed to. Especially if you count the civil war.

How to deal with that debt is tricky, as we cannot resurrect the slave owners and make them pay it back themselves. Also even if we could it would probably take them a very long time, as throwing away 70% or more of a lifetime of productivity with each slave quickly adds up to a very hard to pay debt, easily many lifetimes of labor with a mere handful of slaves.

None the less, this is what is owed. More to the slaves, and any that inherited the debt owed to them, but also all others who would have benefited from inventions, infrastructure, or even "my grandfather would have spent a penny less on entertainment if the brilliant comic Sam Priceson had been a comic rather than a slave, which might have eventually lead to me inheriting an extra penny". All such were robbed, though some by dimes and some by lives, quite different scales.

If not, then why is it outright impossible, rather than implausible, that some mass economically disastrous policy followed for generations even as economists scream could half the GDP of a nation? How much wealth of Russia has been destroyed by their idiotic policies? Are you certain the GDP of Russia since it started is enough to repay all the damage dealt by their policies. Once again I think it probably is, but perhaps it is not.

Failure to listen to the "dismal science" can destroy vast amounts of wealth. While infinite seems quite dubious here, a few hundred trillion going up in smoke seems far from impossible. A few quadrillion less likely, but "nonsensical"?

Wealth may be measured in dollars, but the wealth of the world is not in dollars. It is in bridges, books, inventions, diseases cured, and so many other things. The total wealth of the world is far more than the number of dollars in it. A dollar is just a promise that you can have some of that wealth to go to you.

How many breakthroughs could those slaves have made, were they free? How much wealth would those breakthroughs generate? Most slaves would not have made breakthroughs. Breakthroughs are flashy, finding a cure for a disease memorable, but that is only part of the value lost. Most would have been ordinary. But how many roads weren't made? Books unwritten? Little restaurant unopened? Shoes uncobbled? Vegetables unfarmed? Little bits of scientific progress undiscovered? These things add up. Much of our wealth is from the accumulation of millions of ordinary people contributing. If you cut down a massive group of them to a third or less of a contribution, that adds up.

One of the famous characteristics theft often has is that it destroys wealth on net.

But if I steal your house, only to burn most of it down leaving only fifty dollars worth of remains, it may be that the money left between both of us isn't enough to buy a new house.

Atoms don't forbid this from happening. Nor do they forbid a million people doing that at once.

Does my destructive behavior mean I owe you less than a house? Does your poverty mean that? Does our combined poverty mean that?

You arrived at a nonsensical conclusion that cannot possibly be true and I don't need to do anything other than point to the fact you arrived at a nonsensical conclusion that cannot possibly be true.

I have three objections here.

The first is that I strongly dispute the claim that stealing more makes you owe less as long as you also waste what you steal.

The second is that the total GDP of the United States since its founding is less than 5% of the total wealth of the world, and so even if stealing and destroying wealth started going in reverse at the 50% point, one could still easily owe ten times that according to the rule you stated.

I have stated where my numbers can come from, how they can add up.

Your premise of trading the coconut is the perfect illustration of why your math is wrong and I pointed it out to you and you didn't realize it. You say the trade of the coconut is fair because it represents all of the material wealth available to us. Its value is compared to the value that is available to us in the real world made of atoms. If you had decided to do this moral arithmetic to declare my wrongs (those of a demographic I am in) had deprived you of 11 quadrillion dollars in value, and all I had was a coconut, and I gave you the coconut, I will have given you all of the material wealth possible and will have not made an impact on the debt you have calculated because that number is not based on comparison to anything.

Your numbers cannot add up because we can see them not add up and I am pointing to them not adding up and saying "Look at how these don't add up!" and you keep saying "but you can't prove what specific step of the chain of reasoning is wrong!" and I don't need to because your numbers don't add up! The proper reparations for slavery cannot be "turn every black person into a millionaire" because if slavery had not happened every black person would not be a millionaire, and that remains true no matter how clever your logic. Don't tell me that the logic is sound when I can see the results it produces aren't!

We could fermi estimate quite a few more reasonable numbers. Ten million slaves times 100,000 dollars per slave would get a trillion dollars. Seems a bit low to me but I could be argued there.

If we bump it up to a million dollars and ten million slaves we get ten trillion dollars.

Investment and inflation and a few other things combined over five generations, or maybe if I underestimated the slaves or money by a few factors less than ten... might well bump it up to 100 trillion, that gets into stuff about slavery era investment and invention opportunities which I'm far from an expert in. If things like that add up to an extra power of ten it seems not that unlikely to me.

A quadrillion... there are quite a few bits of slack in the previous, if a lot of things line up... it seems unlikely but I wouldn't call it automatically insane, just pretty dubious based on my knowledge and estimations.

Ten quadrillion? That... maybe I'm missing something but I'm not seeing how that would work without a more significant error on my part.

But on the whole, you seem to be arguing against me as a totem for some broader group.

I didn't bring up demographics. You did. Why did you bring them up in a conversation with me? I don't believe in sins of the father, let alone sins of the second cousin twice removed.

And why can't the proper reparations be turn every person, black or not, into a millionaire?

I don't think the numbers come out that way here, but...

If someone bombed almost the whole world so badly that an intact but rickety shack was considered a significant amount of wealth, it seems to me that your reasoning here would lead to someone going "SoundLogic? You think proper reparations here would be enough to give everyone a merely rickety shack? That is crazy!".

But, uh, yah. If you blow up my house you actually owe me more than a rickety shack, in fact. The total amount owed will I expect easily add up to enough for everyone to get a rickety shack and more.

Getting paid will be tricky, yes, but that doesn't erase the truth of the number.

How much damage did slavery do? I don't know. I gave some estimates. But often destruction is easier than creation. Burning down a house is easier than building one. So if people put a lot of resources into burning down houses, we can easily discover that the resources to repay the victims vastly exceeds the resources that were ever available to the arsonists.

Not all of the US, not even all the white people, were the arsonists in this story. But a neighborhood with ten people, one of whom is an arsonist, may wind up having the arsonists burn down more wealth than the entire neighborhood can pay.

Not that the neighbors are responsible for that debt either, to be clear. Before you start accusing me of sins of the neighbor reasoning. My point was purely that the destruction inflicted by a small fraction of a group can easily exceed the liquid assets of the whole group.

So finally, I ask you:

1) Why do you keep repeating the number 11 quadrillion? I never mentioned it.

2) Why do you keep bringing in demographics and sins of the father type stuff?

B: Literally the entire subject of this discussion is demographic sins-of-the-father punishment. It is the discussion in its entirety. The discussion cannot be separated from that concept.

C: The discussion in its entirety, at every point, has been about the reparation of harms, to the point that it is impossible not to notice this if you have read the discussion. If your math in the total economic loss concludes that the reparation of harms for that loss, the reparation in the real life world that is made of atoms and not a moral number universe, is "make every black person a millionaire" then your math is wrong.

Now answer the question I asked you. What would be the result of applying this same grievance-math about the economic damages of slavery to every other form of harm that has caused economic damages? Does that process produce a nonsensical number? Then, why should I care about that number, and does that indicate anything about whether we should care about this grievance-math?

I agree that many people look at these numbers and try to use them for sins of the father arguments. I'm not completely ignorant of the broader context here. However, even looking at the links here, I believe the only talk about sins of the father or collective punishment before you brought it up was me disavowing using these numbers for that.

I am extremely iffy at best about this sort of appeal to consequences type reasoning for suppressing truth. I think it has a very bad history. I admit it isn't necessarily always wrong, if person A raped person B and I was going to make this information public, and you came along and showed me that consistently people would kill the rapists entire family, that... I would be nervous about letting that stop me, but it would at least have me hesitating.

This is not that case. There are plenty of numbers out there of various levels of credibility and plausibility already. Most of them don't include specific "no sins of the father" disclaimers in them. Adding another one with such disclaimers is fine, and speaking truths like this is important.

If the number is 11 quadrillion (which I doubt), I wish to believe it is 11 quadrillion. If the number is 10 trillion (which seems decently plausible to me), I wish to believe it is 10 trillion.

I still do not see your argument for why a result that concludes "every black person is made a millionaire" is nonsensical. I'm pretty sure that one is false, at minimum some aren't the descendants of slaves at all and while they probably inherited some owed to them debts I doubt that adds up to a million dollars, and I'm seeing numbers more in the hundreds of thousands as far more likely.

If someone bombed almost the whole world so badly that an intact but rickety shack was considered a significant amount of wealth, it seems to me that your reasoning here would lead to someone going "SoundLogic? You think proper reparations here would be enough to give everyone a merely rickety shack? That is crazy!".

So my big question for you is: So how does your reasoning not apply equally well here? Do you think that the extra damage they did means they don't owe someone a house for blowing up their house, or do you think this case is different somehow?

I'm seeing numbers like 8%-18% of Americans being millionaires, and evidence for that having increased dramatically. Some of this is inflation, housing market distortion, and so on, so it is tricky for me to figure out how to interpret this, but it makes it seem not so out of the question that without the damages of slavery (and the civil war) more than 50% of the population would be millionaires, and this might well include more than 50% of the black population. Though figuring out how to adjust for inflation between alternate histories sounds difficult. Every black person seems unlikely, though I will note that if someone steals a million dollars from someone, I'm not aware that they can successfully argue "the person I stole from had a gambling addiction, and probably would have lost most of it anyway, therefore I should only have to pay back a small fraction". The rule seems to be that the stolen money results in the responsibility for not wasting the money being transferred to the thief. This would mean there are some perks to your parent's money being stolen, since you no longer have to worry about them losing it to a gambling addiction, which would mean that the fraction of millionaires among people who had generations of this perk applied would likely be higher than baseline. Still, the rule doesn't seem that unreasonable to me? If you don't want to be a free financial manager who offers guarantees with your own money for generations of black people, don't literally enslave them and then literally fight a civil war to prevent slavery from being abolished and then interfere with reconstruction and so on.

This would still leave some black people who simply don't have an ancestor who someone decided to... "provide this service too". But if lots of people historically handed out dozens of these (by doing the whole 'enslaving' thing) then, uh, yah, that adds up to them 'promising' a lot of very valuable financial services. And if current black people are richer because their parents were physically prevented from spending the inheritance it... isn't clear how this is an inherently absurd possibility. There are some secondary effects that get complicated, involving the resulting illegitimate suppression of those who would have otherwise provided services for those parents and made money by doing so, but... I already mentioned that in another thread.

As for your question

I already applied it to some very very back of the envelope estimates of the impact of misogyny earlier in this thread. In another thread I applied it to a hypothetical "Grug" who singlehandedly managed to bully people into setting back agriculture as a whole hundred years.

I put less work into those, but at least as very rough estimates with all the cravats? Sure, I endorse those.

Other people brought up slavery by "Arabs, Turks, or the Mughals". I'm a lot less educated there, and if that was further back in history there were simply a lot less people which might make a relevant difference, but I have no objections to the math being done for them either.

If you want the total economic damage by "any bad and illegitimate behavior, by anyone, at any point in history"? That is going to be massive.

Like, high chance that is going to be multiple galaxies worth of damage.

And I endorse that.

Based on my understanding of the best speculative but realistic cosmic engineering, it is likely objectively true that multiple galaxies worth of value have been lost due to the cumulative damage of such things.

That is what these numbers are, at their core.

How much better off would people be if people hadn't done that?

And if the thing we are fixing is so massive that it manages to make all of the US slave trade look tiny? Then, uh, that is going to spit out a very huge number.

As it should.

But to me, there are some definitely reasonable use of these numbers, and one very famous possibly reasonable use.

The definitely reasonable uses include:

Just having better knowledge. It can help give a sense of how big and destructive slavery was, or how big and destructive all of humans being awful has been (if we calculate that one)

As a bonding or rallying cry. As I said in another thread, this is a legitimate use of "this is what they took from you", be you black or white.

To understand what sorts of problems are most important to deal with. If the total damage of problem A is ten trillion dollars, and the total damage of problem B is a quadrillion dollars, then assuming both are equally viable targets for dealing with (which is a big assumption, see "effective altruism" for analysis that faces a lot of these things) one should target problem B. It wasn't like economists hadn't pointed out the massive harms of slavery, that was a big part of what got them the title of "dismal science". So maybe in the future we should do what they should have done: listened to that sort of reasoning before we do some massively destructive thing, like slavery.

A more complicated use is this is to try to actually do what we try do with smaller and more recent thefts.

If someone goes and steal a thousand dollars from each of a thousand people, and use that to buy their kid a house, then (if all goes well) a few things happen more or less in the following order (to oversimplify).

First, they get caught, trial, due process, all that.

Then damages are calculated. In this case, that comes out to a million given the straightforward calculation. Maybe some lawyers and economists bring up some other things.

Then penalties are calculated. If we just had them pay back what was stolen, there would be the risk that they would just retry until they got away with it.

We then declare that the person owes money equal to the damages plus penalties.

We then look into how to try to enforce this. This involves various steps.

One of them is figuring out how to distribute the money to the victims. Hopefully they all have bank accounts in a easy to read list.

Another is figuring out how to get money owed out of the thief's own accounts and properties and such. This can be complicated, and sometimes additional penalties are added for making it complicated on purpose.

If they have already disbursed or simply destroyed too much of their wealth to pay the victims what they owe, things get even more complicated.

One tool that can be used is wage garnishing, where we basically say they have taken out a loan from the victims and we will make extra sure they prioritize paying it back by taking away their control by directly going to their employer and redirecting the money away from them before it even enters their accounts.

Another tool that can be used is sometimes known as a 'clawback'. In the case of the gift house to the kid, we might take the house back and resell it. This can go further, clawing back money they put in a kickstarter, donated to a charity, or even clawing back wages or payment for goods and services.

The clawback can get pretty messy pretty quickly, and so historically there has been a lot of reluctance to use it.

It risks harming innocents in the process. If someone spent a million dollars on commissioning artwork from freelance artists online, there isn't really a great way to handle that. It isn't like we can just give back the time the artist spent making the art. The charity may have themselves spent money in response to the donation, making it so that if the money is clawed back they won't be able to continue their charitable work even to the degree they could if they had never received the money in the first place.

Perhaps the kid sold their old house to someone who responded by selling their old house to someone who responded by selling their old house.

The whole situation is a nasty mess, and there are good reasons why we try to avoid it, even to the point of sometimes having to go to the victims "sorry, fixing this isn't reasonably achievable".

Here we have one that makes even very nasty conventional cases of that look downright clean.

We cannot reclaim the wealth by garnishing the wages of the dead. This would be the simplest, least moral dilemma filled, and least economically destructive solution. Simply resurrect the slavers, figure out how much they owe, and make them work until they pay it off. The part of me that is very angry about the whole "stole wealth, some of which would have gone to me, with a large portion of it going to whipping slaves, one of the least sympathetic uses for the stolen wealth that remains remotely plausible, seriously, I would be so much less mad if they had simply spent that two thirds or whatever on drugs or parties or something, and then they did a civil war to double down on it" suggests ironic punishment and making them pick cotton until they have paid it off, but I don't think I actually endorse that. Regardless, this one isn't viable.

We cannot simply take it from their estates. Their estates have already been disbursed, and even if they weren't, they probably never had enough money to repay because of the destructive nature of slavery.

We could do clawbacks. This still won't be enough even if we clawback everything, because of the earlier mentioned "not enough money in the first place because they were destructive" issue.

To make it worse, those clawbacks now involve traces of over a hundred years of further business. This is, I think, well beyond what any clawback in american jurisprudence has done.

Maybe we could find a handful of cases where more or less by luck things haven't gotten too tangled. Maybe there are some people living on in a mansion that was passed down through simple inheritance from a slave owner whose slaves' descendants we can easily find. I could see a clawback there not being crazy. Not sure I agree with it, but 'crazy' would seem to be going too far. It sucks to think you will own a mansion and then discover it was stolen and have to give it back, but I don't think it is wildly outside the range of reasonable clawbacks, if even outside at all.

This will not add up to much.

Some people have proposed more massive clawback like structures. These, whether they are right or wrong, are crazy. If they are right, they are "I can't believe you figured out how to do that and pulled it off" type crazy. If they are wrong, they are "I literally and straightforwardly do not believe you figured out how how to do that or can pull it off, you are crazy, do not do this" type crazy.

A good version of this would try avoid sins of the father type reasoning, be aware of the bias there, and be careful enough to manage it.

A good version would need to think through the likely massive economic disruption and figure out how to mitigate and calm it.

A good version would need a lot of caution about how to figure out tracing the money.

In short, a good version would have to face economic hell and survive to pass through it.

Maybe this could be done by vastly reducing scope. A transfer of ten dollars per person might avoid enough of the economic hell. Maybe. It would still be difficult, ten dollars can mean a lot to people in poverty, and killing or crippling thousands or more of white people in poverty would be a big problem. On top of that, the political and social fallout sound like their own hell to manage.

Maybe a good version of this could be invented by humans limited to modern technology. I am very skeptical of attempts to go much beyond "hey, we actually found one of the rare straightforward clawback cases". And honestly pretty skeptical of those too, those sound ripe for abuse and creating bad precedent.

There are also an more radical idea: Inherited debt without cosigning the original or signing for inherited property.

This is... this one is...

It is not without legal precedent. In fact, if you squint, it has a very famous legal precedent: Actual American slavery. To quote: "partus sequitur ventrem".

It is also pretty much right next to "collective punishment" in terms of awfulness. And some of these proposals blur the line between the two. Technically, if we trace the inheritance it is inherited debt, if we do it based on race it is collective punishment.

The judicial system has moved well away from either of these, horrible slipups and backsliding aside, and this is very good.

I offer no defense of this. If necessary, I could come up with some pretty weak mitigating factors, but I'm not even going to bother here.

This is an abomination in almost if not literally every case that has ever happened in history, and the proposal that this would be an exception is one that would require such massive amount of evidence as I would struggle to describe.

I do however want to say this abomination does not invalidate the numbers themselves, or motivate me to not discuss them.

If slavery did do 11 quadrillion dollars of damages (it probably didn't), I wish to believe and say slavery did 11 quadrillion dollars of damages, even if some people go "therefore we need to do race based inherited slavery again" in response. Which is itself further than even a lot of the crazy proposals go.

To conclude, my questions for you are these:

Will you stop unfairly linking me to 11 quadrillion dollars, a number I have now explicitly expressed skepticism of, and limit your claims to numbers in the trillions (or high billions), which I find of varying but much higher plausibility?

Did I answer your question adequately? To repeat and summarize my answer: If you mean "the total of all those harms" the answer is "probably a cosmic amount" and this number seems quite sensical to me. You should care about this number because, even if some people want to do crazy things about it, it is true and tragic. You should probably care about it less than the slavery number more though, since avoiding mistakes of that magnitude is probably a lot more viable a goal compared to the (admittedly much smaller) payoff than trying to avoid all of such a broad class of mistakes.

How does your 'it must be nonsense because atoms"' reasoning not apply equally well to the "someone bombed the world" case, or if it is equally well, why does someone doing bombing to other people make their amount owed to the first victim less?

Have I sufficiently disavowed collective racial punishment for you to stop yelling at me for it?

Stop telling me that the internal logic of the math is correct. I don't care and have never cared and have told you in every post that I do not care and have told you in every post that the internal logic of the math is valueless and irrelevant to the discussion and the entirety of the claim that i made and that you responded to is that the internal logic of the math is valueless and irrelevant.

If the amount of reparations owed is "a cosmic amount" your math is incorrect. It is wrong. It is incorrect. The answers are not correct. The numbers are not true. They are valueless. They are not true. They do not model reality. They do not need to be suppressed because they are dangerous, but I want the people saying them to shut up because they are making valueless noises and making demands based on them. Any use of those numbers is "misuse" because they have no value or utility or accuracy because they are wrong. The numbers are wrong. They aren't correct.

Stop telling me the internal logic of the numbers works. Stop doing that. I don't care. There's absolutely nothing you could possibly say about the method that you used to arrive at those numbers that would afford them any value. None. Stop telling me the reasoning is sound. Stop doing that. I don't care, because I know the reasoning is not sound, because I can see the result is wrong.

Profits and poverty: The economics of forced labour

As a reductio, suppose rather than slavery a group of people kidnapped all the cancer and other medical researchers in the United States, and also stole 90% of the real wealth everyone else in the United States, repurposing it into torture equipment to torture the researchers, and after doing this for fifty years they smashed the equipment into scrap and threw it into the deepest part of the ocean also killed the researchers.

Inherited debt is a complicated thing, and sins of the father aren't sins of the son, so I have objections about saddling some random child they had who played no part and got no benefit due to having no emotional investment in this crazy scheme.

More directly though, the reparations/debt here should clearly be massive. It should be all the stuff they stole as wealth, weregild or such for the murders, and lots of people with, say, cancer are probably also owed massive amounts for, say, setting back cancer research in a way that will or has killed them.

I think the slavery case is less intense than that, but the core moral complication is this: suppose someone steals a billion dollars by stealing ten thousand dollars from many people, invests it very poorly, winds up with only a million dollars, and the uses that to buy a house for their kid before they die, and then the theft is discovered, and those billion dollars would have been inherited by others since at least some of the original owners died before the discovery. I don't believe sins of the father to sins of the son, and it is especially tricky here because a lot of the stolen wealth was lost and so it can't be just taking away a billion dollar house that never should have been given to the kid, the father ended up heavily in debt beyond his ability to pay. It does seem plausible that the million dollar house should be paid back as reparations, but the kid doing that and winding up still nine hundred and ninety nine million dollars in debt to the descendants definitely doesn't seem ok.

Yet still, if we honestly add up the amount that the people are owed, it is a billion dollars. Plus interest. The thief being "judgement proof" by being dead doesn't change that.

And consider further: the kid of the researcher killers may be one owed reparations! Suppose that kid has some chronic illness, and the treatment for that has now been set back thirty or more years!

I find it entirely plausible that even very wealthy descendants of very wealthy slave owners are in fact worse off due to the destructive effects of slavery.

This is one of the reasons I dislike framings like "men benefit from misogyny". I find it quite plausible that literally no one alive on the planet net benefits from misogyny. How many technological advantages were missed out on? In the 1500s, would any man be made better on net by any slight benefits granted via misogyny compared to, say, access to 21st century technology? If misogyny managed to slow down technological advancement by even 1%, which is probably an underestimate, for all the time it has existed, that easily adds up to 400 years of technological progress.

Slavery had far less time to do damage, and influenced less of the population, though it probably was a bigger productivity loss on those it hit on average. Still, it adding up to a massive damages bill seems quite easy. More than the total GDP of the United States... that is tricky. Reinvestment, exponential curves, I think at least by conventional economics (no singularity or x-risk type stuff) probably still wouldn't, but I wouldn't be too badly shocked if it managed to. Especially if you count the civil war.

How to deal with that debt is tricky, as we cannot resurrect the slave owners and make them pay it back themselves. Also even if we could it would probably take them a very long time, as throwing away 70% or more of a lifetime of productivity with each slave quickly adds up to a very hard to pay debt, easily many lifetimes of labor with a mere handful of slaves.

None the less, this is what is owed. More to the slaves, and any that inherited the debt owed to them, but also all others who would have benefited from inventions, infrastructure, or even "my grandfather would have spent a penny less on entertainment if the brilliant comic Sam Priceson had been a comic rather than a slave, which might have eventually lead to me inheriting an extra penny". All such were robbed, though some by dimes and some by lives, quite different scales.

If not, then why is it outright impossible, rather than implausible, that some mass economically disastrous policy followed for generations even as economists scream could half the GDP of a nation? How much wealth of Russia has been destroyed by their idiotic policies? Are you certain the GDP of Russia since it started is enough to repay all the damage dealt by their policies. Once again I think it probably is, but perhaps it is not.

Failure to listen to the "dismal science" can destroy vast amounts of wealth. While infinite seems quite dubious here, a few hundred trillion going up in smoke seems far from impossible. A few quadrillion less likely, but "nonsensical"?

Wealth may be measured in dollars, but the wealth of the world is not in dollars. It is in bridges, books, inventions, diseases cured, and so many other things. The total wealth of the world is far more than the number of dollars in it. A dollar is just a promise that you can have some of that wealth to go to you.

How many breakthroughs could those slaves have made, were they free? How much wealth would those breakthroughs generate? Most slaves would not have made breakthroughs. Breakthroughs are flashy, finding a cure for a disease memorable, but that is only part of the value lost. Most would have been ordinary. But how many roads weren't made? Books unwritten? Little restaurant unopened? Shoes uncobbled? Vegetables unfarmed? Little bits of scientific progress undiscovered? These things add up. Much of our wealth is from the accumulation of millions of ordinary people contributing. If you cut down a massive group of them to a third or less of a contribution, that adds up.

One of the famous characteristics theft often has is that it destroys wealth on net.

But if I steal your house, only to burn most of it down leaving only fifty dollars worth of remains, it may be that the money left between both of us isn't enough to buy a new house.

Atoms don't forbid this from happening. Nor do they forbid a million people doing that at once.

Does my destructive behavior mean I owe you less than a house? Does your poverty mean that? Does our combined poverty mean that?

You arrived at a nonsensical conclusion that cannot possibly be true and I don't need to do anything other than point to the fact you arrived at a nonsensical conclusion that cannot possibly be true.

That's not true at all.

Lets forget all this moral stuff and think in terms of really straightforward contract law with gold currency:

If I lend you 1kg of gold and you promise to pay me back 2kg of gold next year then you owe me 2kg of gold. There's no law of reality that says the extra 1kg must exist at all, let alone that you're able to get it. Maybe you've got a brilliant business idea and you can get that 1kg easily, maybe you don't and I'm a moron for lending it to you. But either way the debt is equally real, equally valid, and equally unrestricted by what's actually possible to repay.

Taking one step further to very well established and concrete notions of common law damages:

Suppose I smash your window. In any sane notion of justice I owe you at least the price of replacing the windows. There's no law of justice that says the amount I owe you needs to exist. It needs to *have existed*, (ignoring stuff like exemplarary damages for the moment), there has to have been something worth that in the past which I broke. But the value of your window is not conserved in the shards, I didn't suddenly become wealthy by smashing it.

This is why liability insurance exists and is mandatory for so many cases. Because it's very well established that you can hurt someone more than the amount you benefit from your offense.

It seems to me utterly trivial than slavers owe their slaves *vastly* more than it's possible for them to repay. The same way that murderers or arsonists do. If we were to establish today a totally ordinary common-law court and allow a random contemporary slaveholder to be sued by his slaves for what they were justly owed then the slaveholder would die a bankrupt with the vast marjority of his debt unpaid.

The calculations that result in these figures are not calculations like judging the wrongs done by a person to a person and assigning a dollar amount. We know that because we can look at them, they're in the OP of this thread. They are made out of so many layers of abstraction and assumptions, and are so biased in choosing what to count and what not to count, that the only thing the numbers mean is "the person calculating this number wanted it to be this number." It is a calculation of economic damages that is untethered from the concept of economy, designed only to create the largest number it can in order to be outraged by it and make infinite demands while pointing at it.

My observation is only that there just isn't a global bound on "the amount justly owed by a particiular crimimal in just restitution for a particular crime". That's 100% determined by what the actual crime was and how much the actual victim was deprived of. Obviously OP is making up some bullshit number rather than actually calculating damages like common law courts have been doing justly and reasonably for centuries. But people bullshitting doesn't mean we need to impose an artifical global limit on a totally ordinary process of any sensible legal system.

What is the total economic loss of murder?

Vast, murder is really really bad.

What are "we" owed as damages from all that we lost from murder?

"We", what kind of commie bullshit is this, the muder victim (which means his family) is owed a huge amount. Wareguild is a perfectly reasonable system and obviously just. It didn't stop being obviously just because modern states have decided to use criminal punishments instead.

Who the fuck owes it to us?

The muderer. .... Obviously? Who else?

Because the total economic loss from murder, calculated the way this calculation is done, would outstrip the total value of goods and services ever produced by humanity by a lot.

It would! Murder is really bad, people who do it ought to die bankrupt.

Does it make the slightest bit of sense to say that someone "owes us" the value destroyed by murder?

Yes, because murder is bad and if you do a bad thing the person hurt by it ought to take as much as they can from you until they're justly compensated.

Why should we care the slightest bit about that number, much less act on it?

Because we care about that number for smaller crimes where we can recover the entire amount of a just compensation, and it's unreasonable to just stop doing that when you cross the boundary from a broken arm to an amputated leg or whever you arbirtarily decided that criminals shouldn't compensate their victims anymore.

This whole thing would make a lot more sense if at the top of this thread there was a post saying "The amount of damages is at least a quadrillion, and this must be paid by the white race which is definitely a reified thing with concrete interests, to the black race which is a thing that must be protected apart from any number of actual black people. Race war is great. And if you want to agree with any part of this post to any degree, either get your own post or you are commiting to agreeing with all of it".

Or something weaker. But I read the thing multiple times I thought and I cannot figure out what you are responding to except some vague "people who bring up this number usually think race war nonsense"-y thing. Which I guess probably exists? I don't see why we can't calculate the number and see if there is any way we can squeeze some actual justice out of it in the absence of resurrection magic though.

I didn't read the linked paper super deeply, and the calculations looked maybe iffy, but I didn't see anything about races owing other races there.

Did I miss it? Did I miss something in some post before shieldfoss?

You certainly seem to be acting like I did, but I failed to find it.

Profits and poverty: The economics of forced labour

As a reductio, suppose rather than slavery a group of people kidnapped all the cancer and other medical researchers in the United States, and also stole 90% of the real wealth everyone else in the United States, repurposing it into torture equipment to torture the researchers, and after doing this for fifty years they smashed the equipment into scrap and threw it into the deepest part of the ocean also killed the researchers.

Inherited debt is a complicated thing, and sins of the father aren't sins of the son, so I have objections about saddling some random child they had who played no part and got no benefit due to having no emotional investment in this crazy scheme.

More directly though, the reparations/debt here should clearly be massive. It should be all the stuff they stole as wealth, weregild or such for the murders, and lots of people with, say, cancer are probably also owed massive amounts for, say, setting back cancer research in a way that will or has killed them.

I think the slavery case is less intense than that, but the core moral complication is this: suppose someone steals a billion dollars by stealing ten thousand dollars from many people, invests it very poorly, winds up with only a million dollars, and the uses that to buy a house for their kid before they die, and then the theft is discovered, and those billion dollars would have been inherited by others since at least some of the original owners died before the discovery. I don't believe sins of the father to sins of the son, and it is especially tricky here because a lot of the stolen wealth was lost and so it can't be just taking away a billion dollar house that never should have been given to the kid, the father ended up heavily in debt beyond his ability to pay. It does seem plausible that the million dollar house should be paid back as reparations, but the kid doing that and winding up still nine hundred and ninety nine million dollars in debt to the descendants definitely doesn't seem ok.

Yet still, if we honestly add up the amount that the people are owed, it is a billion dollars. Plus interest. The thief being "judgement proof" by being dead doesn't change that.

And consider further: the kid of the researcher killers may be one owed reparations! Suppose that kid has some chronic illness, and the treatment for that has now been set back thirty or more years!

I find it entirely plausible that even very wealthy descendants of very wealthy slave owners are in fact worse off due to the destructive effects of slavery.

This is one of the reasons I dislike framings like "men benefit from misogyny". I find it quite plausible that literally no one alive on the planet net benefits from misogyny. How many technological advantages were missed out on? In the 1500s, would any man be made better on net by any slight benefits granted via misogyny compared to, say, access to 21st century technology? If misogyny managed to slow down technological advancement by even 1%, which is probably an underestimate, for all the time it has existed, that easily adds up to 400 years of technological progress.

Slavery had far less time to do damage, and influenced less of the population, though it probably was a bigger productivity loss on those it hit on average. Still, it adding up to a massive damages bill seems quite easy. More than the total GDP of the United States... that is tricky. Reinvestment, exponential curves, I think at least by conventional economics (no singularity or x-risk type stuff) probably still wouldn't, but I wouldn't be too badly shocked if it managed to. Especially if you count the civil war.

How to deal with that debt is tricky, as we cannot resurrect the slave owners and make them pay it back themselves. Also even if we could it would probably take them a very long time, as throwing away 70% or more of a lifetime of productivity with each slave quickly adds up to a very hard to pay debt, easily many lifetimes of labor with a mere handful of slaves.

None the less, this is what is owed. More to the slaves, and any that inherited the debt owed to them, but also all others who would have benefited from inventions, infrastructure, or even "my grandfather would have spent a penny less on entertainment if the brilliant comic Sam Priceson had been a comic rather than a slave, which might have eventually lead to me inheriting an extra penny". All such were robbed, though some by dimes and some by lives, quite different scales.

If not, then why is it outright impossible, rather than implausible, that some mass economically disastrous policy followed for generations even as economists scream could half the GDP of a nation? How much wealth of Russia has been destroyed by their idiotic policies? Are you certain the GDP of Russia since it started is enough to repay all the damage dealt by their policies. Once again I think it probably is, but perhaps it is not.

Failure to listen to the "dismal science" can destroy vast amounts of wealth. While infinite seems quite dubious here, a few hundred trillion going up in smoke seems far from impossible. A few quadrillion less likely, but "nonsensical"?

Wealth may be measured in dollars, but the wealth of the world is not in dollars. It is in bridges, books, inventions, diseases cured, and so many other things. The total wealth of the world is far more than the number of dollars in it. A dollar is just a promise that you can have some of that wealth to go to you.

How many breakthroughs could those slaves have made, were they free? How much wealth would those breakthroughs generate? Most slaves would not have made breakthroughs. Breakthroughs are flashy, finding a cure for a disease memorable, but that is only part of the value lost. Most would have been ordinary. But how many roads weren't made? Books unwritten? Little restaurant unopened? Shoes uncobbled? Vegetables unfarmed? Little bits of scientific progress undiscovered? These things add up. Much of our wealth is from the accumulation of millions of ordinary people contributing. If you cut down a massive group of them to a third or less of a contribution, that adds up.

One of the famous characteristics theft often has is that it destroys wealth on net.

But if I steal your house, only to burn most of it down leaving only fifty dollars worth of remains, it may be that the money left between both of us isn't enough to buy a new house.

Atoms don't forbid this from happening. Nor do they forbid a million people doing that at once.

Does my destructive behavior mean I owe you less than a house? Does your poverty mean that? Does our combined poverty mean that?

You arrived at a nonsensical conclusion that cannot possibly be true and I don't need to do anything other than point to the fact you arrived at a nonsensical conclusion that cannot possibly be true.

That's not true at all.

Lets forget all this moral stuff and think in terms of really straightforward contract law with gold currency:

If I lend you 1kg of gold and you promise to pay me back 2kg of gold next year then you owe me 2kg of gold. There's no law of reality that says the extra 1kg must exist at all, let alone that you're able to get it. Maybe you've got a brilliant business idea and you can get that 1kg easily, maybe you don't and I'm a moron for lending it to you. But either way the debt is equally real, equally valid, and equally unrestricted by what's actually possible to repay.

Taking one step further to very well established and concrete notions of common law damages:

Suppose I smash your window. In any sane notion of justice I owe you at least the price of replacing the windows. There's no law of justice that says the amount I owe you needs to exist. It needs to *have existed*, (ignoring stuff like exemplarary damages for the moment), there has to have been something worth that in the past which I broke. But the value of your window is not conserved in the shards, I didn't suddenly become wealthy by smashing it.

This is why liability insurance exists and is mandatory for so many cases. Because it's very well established that you can hurt someone more than the amount you benefit from your offense.

It seems to me utterly trivial than slavers owe their slaves *vastly* more than it's possible for them to repay. The same way that murderers or arsonists do. If we were to establish today a totally ordinary common-law court and allow a random contemporary slaveholder to be sued by his slaves for what they were justly owed then the slaveholder would die a bankrupt with the vast marjority of his debt unpaid.

The calculations that result in these figures are not calculations like judging the wrongs done by a person to a person and assigning a dollar amount. We know that because we can look at them, they're in the OP of this thread. They are made out of so many layers of abstraction and assumptions, and are so biased in choosing what to count and what not to count, that the only thing the numbers mean is "the person calculating this number wanted it to be this number." It is a calculation of economic damages that is untethered from the concept of economy, designed only to create the largest number it can in order to be outraged by it and make infinite demands while pointing at it.

My observation is only that there just isn't a global bound on "the amount justly owed by a particiular crimimal in just restitution for a particular crime". That's 100% determined by what the actual crime was and how much the actual victim was deprived of. Obviously OP is making up some bullshit number rather than actually calculating damages like common law courts have been doing justly and reasonably for centuries. But people bullshitting doesn't mean we need to impose an artifical global limit on a totally ordinary process of any sensible legal system.

What is the total economic loss of murder?

Vast, murder is really really bad.

What are "we" owed as damages from all that we lost from murder?

"We", what kind of commie bullshit is this, the muder victim (which means his family) is owed a huge amount. Wareguild is a perfectly reasonable system and obviously just. It didn't stop being obviously just because modern states have decided to use criminal punishments instead.

Who the fuck owes it to us?

The muderer. .... Obviously? Who else?

Because the total economic loss from murder, calculated the way this calculation is done, would outstrip the total value of goods and services ever produced by humanity by a lot.

It would! Murder is really bad, people who do it ought to die bankrupt.

Does it make the slightest bit of sense to say that someone "owes us" the value destroyed by murder?

Yes, because murder is bad and if you do a bad thing the person hurt by it ought to take as much as they can from you until they're justly compensated.

Why should we care the slightest bit about that number, much less act on it?

Because we care about that number for smaller crimes where we can recover the entire amount of a just compensation, and it's unreasonable to just stop doing that when you cross the boundary from a broken arm to an amputated leg or whever you arbirtarily decided that criminals shouldn't compensate their victims anymore.

This whole thing would make a lot more sense if at the top of this thread there was a post saying "The amount of damages is at least a quadrillion, and this must be paid by the white race which is definitely a reified thing with concrete interests, to the black race which is a thing that must be protected apart from any number of actual black people. Race war is great. And if you want to agree with any part of this post to any degree, either get your own post or you are commiting to agreeing with all of it".

Or something weaker. But I read the thing multiple times I thought and I cannot figure out what you are responding to except some vague "people who bring up this number usually think race war nonsense"-y thing. Which I guess probably exists? I don't see why we can't calculate the number and see if there is any way we can squeeze some actual justice out of it in the absence of resurrection magic though.

I didn't read the linked paper super deeply, and the calculations looked maybe iffy, but I didn't see anything about races owing other races there.

Did I miss it? Did I miss something in some post before shieldfoss?

You certainly seem to be acting like I did, but I failed to find it.

Profits and poverty: The economics of forced labour

As a reductio, suppose rather than slavery a group of people kidnapped all the cancer and other medical researchers in the United States, and also stole 90% of the real wealth everyone else in the United States, repurposing it into torture equipment to torture the researchers, and after doing this for fifty years they smashed the equipment into scrap and threw it into the deepest part of the ocean also killed the researchers.

Inherited debt is a complicated thing, and sins of the father aren't sins of the son, so I have objections about saddling some random child they had who played no part and got no benefit due to having no emotional investment in this crazy scheme.

More directly though, the reparations/debt here should clearly be massive. It should be all the stuff they stole as wealth, weregild or such for the murders, and lots of people with, say, cancer are probably also owed massive amounts for, say, setting back cancer research in a way that will or has killed them.

I think the slavery case is less intense than that, but the core moral complication is this: suppose someone steals a billion dollars by stealing ten thousand dollars from many people, invests it very poorly, winds up with only a million dollars, and the uses that to buy a house for their kid before they die, and then the theft is discovered, and those billion dollars would have been inherited by others since at least some of the original owners died before the discovery. I don't believe sins of the father to sins of the son, and it is especially tricky here because a lot of the stolen wealth was lost and so it can't be just taking away a billion dollar house that never should have been given to the kid, the father ended up heavily in debt beyond his ability to pay. It does seem plausible that the million dollar house should be paid back as reparations, but the kid doing that and winding up still nine hundred and ninety nine million dollars in debt to the descendants definitely doesn't seem ok.

Yet still, if we honestly add up the amount that the people are owed, it is a billion dollars. Plus interest. The thief being "judgement proof" by being dead doesn't change that.

And consider further: the kid of the researcher killers may be one owed reparations! Suppose that kid has some chronic illness, and the treatment for that has now been set back thirty or more years!

I find it entirely plausible that even very wealthy descendants of very wealthy slave owners are in fact worse off due to the destructive effects of slavery.

This is one of the reasons I dislike framings like "men benefit from misogyny". I find it quite plausible that literally no one alive on the planet net benefits from misogyny. How many technological advantages were missed out on? In the 1500s, would any man be made better on net by any slight benefits granted via misogyny compared to, say, access to 21st century technology? If misogyny managed to slow down technological advancement by even 1%, which is probably an underestimate, for all the time it has existed, that easily adds up to 400 years of technological progress.

Slavery had far less time to do damage, and influenced less of the population, though it probably was a bigger productivity loss on those it hit on average. Still, it adding up to a massive damages bill seems quite easy. More than the total GDP of the United States... that is tricky. Reinvestment, exponential curves, I think at least by conventional economics (no singularity or x-risk type stuff) probably still wouldn't, but I wouldn't be too badly shocked if it managed to. Especially if you count the civil war.

How to deal with that debt is tricky, as we cannot resurrect the slave owners and make them pay it back themselves. Also even if we could it would probably take them a very long time, as throwing away 70% or more of a lifetime of productivity with each slave quickly adds up to a very hard to pay debt, easily many lifetimes of labor with a mere handful of slaves.

None the less, this is what is owed. More to the slaves, and any that inherited the debt owed to them, but also all others who would have benefited from inventions, infrastructure, or even "my grandfather would have spent a penny less on entertainment if the brilliant comic Sam Priceson had been a comic rather than a slave, which might have eventually lead to me inheriting an extra penny". All such were robbed, though some by dimes and some by lives, quite different scales.

If not, then why is it outright impossible, rather than implausible, that some mass economically disastrous policy followed for generations even as economists scream could half the GDP of a nation? How much wealth of Russia has been destroyed by their idiotic policies? Are you certain the GDP of Russia since it started is enough to repay all the damage dealt by their policies. Once again I think it probably is, but perhaps it is not.

Failure to listen to the "dismal science" can destroy vast amounts of wealth. While infinite seems quite dubious here, a few hundred trillion going up in smoke seems far from impossible. A few quadrillion less likely, but "nonsensical"?

Wealth may be measured in dollars, but the wealth of the world is not in dollars. It is in bridges, books, inventions, diseases cured, and so many other things. The total wealth of the world is far more than the number of dollars in it. A dollar is just a promise that you can have some of that wealth to go to you.

How many breakthroughs could those slaves have made, were they free? How much wealth would those breakthroughs generate? Most slaves would not have made breakthroughs. Breakthroughs are flashy, finding a cure for a disease memorable, but that is only part of the value lost. Most would have been ordinary. But how many roads weren't made? Books unwritten? Little restaurant unopened? Shoes uncobbled? Vegetables unfarmed? Little bits of scientific progress undiscovered? These things add up. Much of our wealth is from the accumulation of millions of ordinary people contributing. If you cut down a massive group of them to a third or less of a contribution, that adds up.

One of the famous characteristics theft often has is that it destroys wealth on net.

But if I steal your house, only to burn most of it down leaving only fifty dollars worth of remains, it may be that the money left between both of us isn't enough to buy a new house.

Atoms don't forbid this from happening. Nor do they forbid a million people doing that at once.

Does my destructive behavior mean I owe you less than a house? Does your poverty mean that? Does our combined poverty mean that?

You arrived at a nonsensical conclusion that cannot possibly be true and I don't need to do anything other than point to the fact you arrived at a nonsensical conclusion that cannot possibly be true.

That's not true at all.

Lets forget all this moral stuff and think in terms of really straightforward contract law with gold currency:

If I lend you 1kg of gold and you promise to pay me back 2kg of gold next year then you owe me 2kg of gold. There's no law of reality that says the extra 1kg must exist at all, let alone that you're able to get it. Maybe you've got a brilliant business idea and you can get that 1kg easily, maybe you don't and I'm a moron for lending it to you. But either way the debt is equally real, equally valid, and equally unrestricted by what's actually possible to repay.

Taking one step further to very well established and concrete notions of common law damages:

Suppose I smash your window. In any sane notion of justice I owe you at least the price of replacing the windows. There's no law of justice that says the amount I owe you needs to exist. It needs to *have existed*, (ignoring stuff like exemplarary damages for the moment), there has to have been something worth that in the past which I broke. But the value of your window is not conserved in the shards, I didn't suddenly become wealthy by smashing it.

This is why liability insurance exists and is mandatory for so many cases. Because it's very well established that you can hurt someone more than the amount you benefit from your offense.

It seems to me utterly trivial than slavers owe their slaves *vastly* more than it's possible for them to repay. The same way that murderers or arsonists do. If we were to establish today a totally ordinary common-law court and allow a random contemporary slaveholder to be sued by his slaves for what they were justly owed then the slaveholder would die a bankrupt with the vast marjority of his debt unpaid.

The calculations that result in these figures are not calculations like judging the wrongs done by a person to a person and assigning a dollar amount. We know that because we can look at them, they're in the OP of this thread. They are made out of so many layers of abstraction and assumptions, and are so biased in choosing what to count and what not to count, that the only thing the numbers mean is "the person calculating this number wanted it to be this number." It is a calculation of economic damages that is untethered from the concept of economy, designed only to create the largest number it can in order to be outraged by it and make infinite demands while pointing at it.

My observation is only that there just isn't a global bound on "the amount justly owed by a particiular crimimal in just restitution for a particular crime". That's 100% determined by what the actual crime was and how much the actual victim was deprived of. Obviously OP is making up some bullshit number rather than actually calculating damages like common law courts have been doing justly and reasonably for centuries. But people bullshitting doesn't mean we need to impose an artifical global limit on a totally ordinary process of any sensible legal system.

What is the total economic loss of murder?

Vast, murder is really really bad.

What are "we" owed as damages from all that we lost from murder?

"We", what kind of commie bullshit is this, the muder victim (which means his family) is owed a huge amount. Wareguild is a perfectly reasonable system and obviously just. It didn't stop being obviously just because modern states have decided to use criminal punishments instead.

Who the fuck owes it to us?

The muderer. .... Obviously? Who else?

Because the total economic loss from murder, calculated the way this calculation is done, would outstrip the total value of goods and services ever produced by humanity by a lot.

It would! Murder is really bad, people who do it ought to die bankrupt.

Does it make the slightest bit of sense to say that someone "owes us" the value destroyed by murder?

Yes, because murder is bad and if you do a bad thing the person hurt by it ought to take as much as they can from you until they're justly compensated.

Why should we care the slightest bit about that number, much less act on it?

Because we care about that number for smaller crimes where we can recover the entire amount of a just compensation, and it's unreasonable to just stop doing that when you cross the boundary from a broken arm to an amputated leg or whever you arbirtarily decided that criminals shouldn't compensate their victims anymore.

And so should journalists and news reporters who report uncritically on their claims. Fire them. Never let them work in the profession again. EVER.

The best one I have right now is that "cops who claimed or backed up claims" was the original post and you swapped in "all cops".

I don't have the numbers on hand to know if the gap between those two is enough to reach "very effective loophole".