There's no law section on 2p2, and I think the closest category for this thread is philosophy.....

I watched a documentary last night called One Killer Punch. It covered three cases where a person landed a single punch on another person and the victim ended up dying. Two of the men were convicted and sentenced to manslaughter, albeit with fairly lenient prison sentences. This started a discussion of whether the variance of the outcome of your actions should determine the punishment that you receive.

The two guys that were convicted certainly committed a crime, in that they punched someone else in the face, but their convictions were based on an extremely likely outcome of their action. People get in fights and punch people all the time, yet most either totally get away with it or get a minor punishment. These two men did exactly the same thing, yet due to a completely random outcome they ended up with much harsher sentences.

Let's take four parallel universe scenarios where person A throws a punch at person B. A's action is exactly the same in each scenario, he throws the same punch in the same with with identical power. One of four things happens:

- B slips out of the way, the punch misses and he runs away
- The punch glances off B's face, causing minor bruising
- The punch lands square on the chin, breaking B's jaw and causing him to lose several teeth
- B is knocked unconscious by the punch, he drops to the ground, hits his head on the floor and subsequently dies.

Despite A taking exactly the same action in each scenario, the punishment will be more severe for each subsequent outcome, which is determined purely by variance.

Is it fair that he suffers a harsher punishment if the extremely unlikely scenario of B dying happens? Is it fair that if the punch misses, he will get away with any punishment at all?

For those interested, a review of the documentary is here. It's just been released on C4 in the UK so probably not available anywhere else yet.

In each of the 4 scenarios you outline the crime is different
1- attempted assault
2- assault
3- grievous bodily harm
4- manslaughter

Most importantly, the outcome for victim B is radically different.

So why should the outcome for the assailant A not vary proportionally ?

There's a thing called The eggshell skull (or 'you take your victim as you find him') rule that might apply. IOW, it's not a defense that the victim was extraordinarily susceptible to being punched even though the offender didn't know it.

I understand that, hence stating in my OP that A would be punished progressively more harshly for each scenario.

What I'm struggling with is should the punishment/crime be more serious based on the outcome if the initial action is identical?

Shouldn't A's action in itself be the thing that is punished, and not the outcome?

Is punishing more harshly dependent on the outcome not just results orientated thinking?

Is it fair that two people who commit the exact same crime are punished differently just because one happened to have a much more serious, yet far more unlikely outcome?

Thanks, I knew there'd be some sort of legal precedent for this. I'm sure if I search around with that term I'll find some deeper explanation.

Suppose criminal punishment is justified because it deters future crime. It does this by placing a penalty on the criminal for committing the crime. This penalty affects the EV calculation of the criminal, hopefully making committing the crime sufficiently bad as to motivate them to not do it.

However, the distribution of outcomes (i.e. variance) doesn't affect the criminal's EV calculation. Suppose the sentencing judge flips a coin and doubles the EV loss for half of convicted criminals and lets the other half go free. This should have, from an EV perspective alone, the same deterrence effect on the criminal and so is equally justified.

By analogy, the fact that river outcomes in poker don't match EV is not results-oriented thinking, it is just an artifact of the rules of poker. A better way of putting your question then is whether and how much criminal punishment should attempt to minimize variance.

What would a system look like that punished offenders based on the statistical likelihood of the outcome of a perpetrator's actions? First question would be how would you determine what belongs in list of potential outcomes to consider for punishment? How do you assign a potential likelihood to their outcomes that is fair? How would this be better for other crimes that are more serious like murder? I don't have answers to these questions I just think they're interesting for discussion.

Sent from my SM-G900R4 using Tapatalk

It is to outsource retribution to the state, rather than have the local yokels exact revenge on their own. This reduces the likelihood of escalation.

The important thing to understand is that the whole point of a legal system is, or should be, to address the harm caused to the victim. The point isn't really the criminal's actions, but the consequences they have on the person they hurt.

Disagree. The point of criminal punishment is to prevent future crime.

Should probably read the thread title before I spout off next time.

EDIT: So yeah, it isn't results-oriented thinking if you think the goal of criminal punishment is some social effect rather than trying to match up the crime and the punishment on some kind of moral scale.

Is it fair that someone other than me won the Lotto last Saturday?

The way we punish outcomes is bizarre when you think about it. Agreed.

Think of aggressive acts having a lottery of outcomes though. If the victim loses the lottery, the thinking is the perpetrator should too. It seems reasonable if you look at it that way.

This is basically how life works, by the way. If you do a wrong risky thing repeatedly, and get away with it, life gives you no consequences. If you get unlucky, you get disproportionate consequences to your average EV. Human justice merely follows natural justice in many ways.

But if deterrence is all you care about then it would be perfectly proper to execute an innocent man, since this will deter crime to the same degree so long as the fraud is not discovered by the public. There is another problem with this position, which is that for maximum deterrence you should have maximum punishment, for example capital punishment for petty theft, which is obviously unjust.

Restitution for the victim of crime should be the overriding priority, not rehabilitation, deterrence, denunciation or punishment.

Restitution just makes crime a freeroll (you can see how well that's working out on the corporate scale). And if you get raped by a poor, what does focusing on restitution even mean?

Punishing an innocent person is usually a mistake from a deterrence perspective. What are you deterring?

As for capital punishment: two points. First, my admittedly limited look at the evidence seemed to indicate that the certainty and quickness of the punishment rather than its severity were most important in deterring crime.

Second, even if it is a more effective deterrent, I think we shouldn't regard this as a reductio. Say we decided to punish drug dealing with death and this cut this crime in half. Here we have to balance the harm of killing the convicted dealers vs. the reduced social harm from less drugs and crime and the people who now don't commit this crime and so don't go to prison. It is not immediately obvious to me that one side or the other here has the right (waving aside the practical and political problems that would make that unsustainable).

Why? Isn't it better to stop a crime from happening in the first place rather than it happening even if the criminal pays restitution to the victim?

It's certainly a very interesting question (and kudos for that).

That said, I think the principle in question is a very tricky legal grounds. Consider that someone somewhere must determine what the possible consequences of actions are, which leads to legal justification for classifying acts as harmful regardless of consequence.

In addition you have to debate exactly what constitutes the initial action. Is it the punch, the intent to punch, the emotion that leads to the intent and so forth.

Fwiw, this legal debate already exists (kind of) in the form of legislation regarding "hate speech".

The possible purposes of punishment are deterrence, incapacitation, restitution, retribution and rehabilitation.

Or because it turns you on sexually. Right, Brian?

There is also that.

I think you forgot tradition; when we punish because we have to, always accompanied by people touting "well, it IS the law".

I think the process has two main goals

1) Make sure crime doesnt pay
2) Somehow lift up the victim of crime. But also here it shouldnt pay to be a victim.

The first point is important for general prevention, the ppl doing crime needs to understand that it might have a short term profit but sooner or later you will pay a bigger price.

Among people that are in the middle of these processes the "revenge" part is not present. The moment you actually have to make these big decisions on other peoples future somehow makes you this way. Lawyers, judges and the common man play judges (or whatever word you use for this) all experience it this way.

Punishment doesnt work, this is proven over and over. And from what i understand on others, the more that is learned about why people chose crime the more it becomes clear that punishment is to a large degree a failed idea. What punishment does do very well is to label people and put them outside of society in such a way that it gets really hard to get back in. The reason punishment persists is probably the common man in the street that needs his dose of revenge to get up in the morning.

I have written about these things often. Its OK in my opinion if luck plays a small part in the severity of the punishment. But not too much. There shouldn't be that much disparity between breaking someone's nose with a punch and unluckily killing him. But personally I think the disparity should be rectified mainly by increasing the lessor punishment. If someone kills somebody while driving drunk, I think his sentence should be a bit less than it is now. But plain drunk driving, especially when very drunk, should be jail time first offense.

It depends if the victim's death affected a lot of people. If the victim had no family or friends, then you give the defendant a "fair" (lenient) sentencing.

If the defendant inadvertently killed someone who had a lot of friends and stature in society, then you repair the secondary victims (the people who mourn the loss of actual victim) by giving the perpetrator an unfair, harsh sentence.

Not trying to rain on your parade, but ... (better just not spill big secrets, lol)

That, in general, is how laws work. Even the ones that have nothing to do with punishment.

You can, very easily.