Hacker News Reader

Comment by ozr | Hacker News Reader

> That’s right, I think that inflicting suffering on vicious wrong-doers is morally praiseworthy even in the absence of any deterrent effect.

Aside from the morals, it's essentially for a functioning society. The intrinsic urge to harm someone that has harmed you can't be allowed to happen. The state has to have a monopoly on violence to function.

If you are not allowed to inflict suffering on people that have wronged you, the state has to.

If the state doesn't, a large enough portion of people will do it themselves, and the state loses legitimacy.

View on HN

How does the (in democratic societies) generally accepted principle in dubio pro reo fit into this? Your assumption would predict that societies that adhere to that standard can not function, because even if everyone is convinced of the guilt of an accused murderer, he can not be punished without formal proof and that might not be available.

I think this idea of required punishment to retain social stability is stupid and is not relevant in praxis. Prevention of crime and support for victims, as well as establishing the truth and encouraging sincere remorse from the perpetrator are much more powerful forces for social peace. In terms of practical social functioning, rehabilitation of criminals is much more valuable than punishment, which is why most democracies at least aspire to this ideal.

Rehabilitation basically doesn't happen, in the sense that interventions don't seem to reduce recidivism rates. Now there are a group of people who basically made one bad mistake due to negligence or other circumstances who don't go back, but this isn't generally what one would consider "reform" and it kinda ignores that we can't seem to do much for the people who continually return to prison after committing new crimes. That said, there may be some limited cases where we can do some good, such as identifying and treating people with pyschosis.

> I think this idea of required punishment to retain social stability is stupid and is not relevant in praxis.

You've never seen mom's brain matter staining the basement walls, or heard that her killer had grandma on same hit list that mom was crossed off of. If the state had not intervened, what do you think would have happened there?

This is not a hypothetical situation.

> Rehabilitation basically doesn't happen, in the sense that interventions don't seem to reduce recidivism rates.

Sources? Here are some that confirm the effectiveness of various interventions/approaches:

- https://link.springer.com/article/10.1007/s11292-020-09420-3

- https://pubmed.ncbi.nlm.nih.gov/31548680/

- https://www.researchgate.net/publication/315907207_Reducing_...

> You've never seen mom's brain matter staining the basement walls […]

What? First, contract killing (which is what you seem to describe here) is exceptionally rare in most societies and second, rehabilitation does not preclude police from interfering with crime. Of course you stop a killing if you have prior knowledge of it, if necessary by restraining/incarcerating the alleged killer. Rehabilitation is implemented once guilt has been established to allow a convicted criminal to re-enter society as a productive individual as soon as possible while minimizing the risk to other members of society.

That was not a contract killing, rather it was a far more common type of murder. The list was something like an enemies list. I have no idea why he kept one, mind you, but the police showed it at trial and it freaked out grandma and some other family who were on there. I'd imagine his former boss wasn't any too happy to realize how easily it could've been him, either, for that matter.

I think you underestimate the revenge cycle that might occur if society wanted to be more lenient than life in prison, which was the result here. The state can't reasonably be more lenient than the average person, or the average person is going to decide that the cost of a few years is worth the revenge. You're already looking at a lifetime of pain for losing someone dear to you, what's a couple of years if you admit your guilt and surrender quietly? And that's before we get into "hey, this guy made a hit list with my name on it, and we have to take that seriously because the person at the top of the list was already murdered."

For the rest, one of your studies says "There was no evidence of significant differences on rearrest or reconviction. Moderator analysis showed no significant moderating effect for risk score, age, or sex." and "the results would be considered mixed, at best." Which... yeah, is pretty much what I've seen for the most part. Maybe I'm not communicating what I think accurately enough, but this sort of result pretty much exactly what I have in mind. You find studies with promising numbers that still turn out to be meh and don't leave you convinced that the intervention is actually changing what the people are capable of.

Importantly, I do not say that nothing shows some promise, or that we shouldn't keep trying, I'm just saying we don't have anything that looks like it can be scaled up to have big impact. And I get that it's hard to measure when plenty of people will do maybe one crime due to some big mistake and had no intention of doing more. I have some small hope that maybe intervening for psychosis could help with violent offenses, but that might assume too much about people taking their meds, unless they're willing and able to use long-lasting injections.

> I think you underestimate the revenge cycle that might occur if society wanted to be more lenient than life in prison, which was the result here. […]

My guess is you are writing from a U.S. point of view. You are severely underestimating how much of an outlier the U.S. is in terms of its justice system when compared to other democracies. Germany, not exactly known for its historical social cuddliness, for example doesn't even allow "life in prison" sentences. Our constitution guarantees the possibility for every convicted, no matter the crime, to regain their freedom. Even people convicted of multiple murders will get their first parole hearing after 15 years and while a court can order their detention beyond this, this is technically no longer a prison sentence and only allowable if there is a high likelihood that the person in question will cause additional harm if released. This assessment is checked every few years with the goal that no one will spend more time in prison than necessary.

This approach to criminal justice enjoys broad support in Germany and most western European societies. All of these societies are also considerably more safe than the U.S. and many other places that put an explicit emphasis on retributive justice. I'm not aware of notable cases of revenge violence due to this.

Retributive justice is an outdated philosophy that only sounds reasonable when you assume the worst in people and ignore the effects of the intergenerational trauma that it produces.

The problem with that is that when you have a situation where the punishment is known to be lenient and there's someone threatening your life, you might just be willing to pay the price for revenge.

Yes, I'm aware that some places don't do life in prison or such. I'm pointing out the harms of that.

What you are describing is the plot of a Liam Neeson action flick, and not how real life works. Revenge killings are not a thing or you’d read about them in societies with “lenient” punishments.

And if someone threatens your life in the moment you are entitled to reasonable measures of self defense, including killing the attacker if you can not avoid harm to yourself or others in another way, that is a completely separate issue.

It's weird to see you flip things like that, when you're the one talking with no actual experience with these things. Anyhow, I certainly have read about them and they have been common throughout history, with justice systems being created to displace such revenge cycles.

I can't help if you're not aware of even thing like the Hatfields & the McCoys, but I can push back when people want to head us back to the bad old days.

Jesus, I was talking about societies with modern justice systems focused on rehabilitation, not civil-war era wild west. Yes, of course absent a functioning justice system revenge cycles will occur. But the whole argument in this thread is about the relative merits of rehabilitation vs. punishment in modern society. And on that front the evidence on the outcome is pretty solid: Punishment-focused justice systems (like in the U.S.) result in more people in prison, are more costly and correlate with higher crime rates (not sure about the current evidence on causality, but I think it is pretty obvious that the trauma and stigma associated with a punishment-leaning justice system would tend to lead to more crime).

Studies quite consistently show that increasing the punishment for a given crime does not reduce the rate it is committed at, all other things being equal. Because guess what, criminals are either unaware that they are committing a crime/are committing a crime in a moment of passion (in which case the punishment is of no consequence) or they think that they won't be caught. Which is one of the many reasons the death penalty is so stupid. Catching criminals and bringing the to trial quickly is much more relevant than the severity of the punishment and the concept of rehabilitation is an extension of that: You find criminals quickly, you sentence them quickly and you invest in them leaving the justice system as quickly as possible to become productive members of society again. Why would I want to spend hundreds of thousands of Dollars on keeping someone behind bars for the rest of their life when I can also try spending a fraction of that on cognitive therapy, apprenticeships and social work to get them out of prison and into a stable life situation in which they have a decent chance of not committing a crime again?

It wouldn't be a problem if people would just stop advocating for going back to the bad old days. You don't know how it feels, but absolutely, if you lower the cost too much people are going to pay it.

The more we reduce this to some sort of transactional justice where everyone is supposed to pretend that nothing happened after a few years in a cage and everything is perfectly okay because people have "reformed" (never mind the people who keep going back...) the more you go back to the bad old days by making the justice system dysfunctional.

Also we can't very well make police more efficient because we don't really agree on the laws. Some people want all drugs to be legal (never mind the addicts who fall into a spiral where they rob others), some people want approximately no drugs to be legal (never mind the other costs). But there's yet to be any social consensus on whether we end up with the Tenderloin, Singapore or somewhere between.

Problem is that I think increasing leniency looks more or less like a phase change at some point. And nobody wants to acknowledge that historically, justice systems made society more peaceful via literal selection effects. Jail was not used so much in the past, there was more corporal punishment or death. Part of the problem is that jail only works while they're inside and imposes other costs, including moral ones.

> How does the (in democratic societies) generally accepted principle in dubio pro reo fit into this?

By creating a society that, on balance, trusts the state and it's institutions to enact justice (apply suffering, in context).

It will fail to be perfect, same as any other system. But it just needs to work well enough that a sufficient mass of citizens don't lose faith in it and try to enact their own forms of justice.

We can all point to cases where murderers got away with it, or someone who was ruled guilty received a sentence we think is wrong, but _overall_, we're ok with the state managing violence on our behalf.

> In terms of practical social functioning, rehabilitation of criminals is much more valuable than punishment, which is why most democracies at least aspire to this ideal.

There is little evidence that any rehabilitation schemes are particularly successful, with the gaps between the most- and least- effective not being particularly meaningful.

You prevent crimes by removing people that commit them from society.

> You prevent crimes by removing people that commit them from society.

By that logic the U.S. should be among the least crime-ridden societies in the world, having an incarceration rate of 531 people per 100,000 compared to 67 per 100,000 in i.e. Germany: https://en.wikipedia.org/wiki/List_of_countries_by_incarcera...

But in reality, Germany experiences significantly less crime than the U.S.: https://worldpopulationreview.com/country-rankings/crime-rat...

In fact, only about a third of convicted criminals in Germany will be convicted again to a prison term after serving their sentence/paying their fines: https://www.spiegel.de/panorama/justiz/kriminalitaetsstudie-... This is similar in the U.S.: https://worldpopulationreview.com/country-rankings/recidivis... A "removal from society" of people that engage in crime previously thus does not seem to be a prerequisite for preventing crime.

Especially murderers seem to experience quite low recidivism rates, likely owing to the fact that murders tend to be highly personal affairs and/or highly dependent on the the availability of a significant motive and opportunity.

The lesson here seems to be that you prevent crime by limiting the situations that would give people the opportunity or require them to commit it. I.e. you provide a social safety net, limit access to firearms, require transparency in financial transactions, etc. And if someone commits a crime, you invest specifically in that person's ability and commitment to not engage in further crime in the future or put differently, you rehabilitate them. That will of course vary from person to person and for some a release back into society might not be an option. But putting people into prison without investing in rehabilitation will just create the conditions to put more (or the same) people into prison (again).

Good luck with generation-spanning vendettas created by such approach.

The point is to avoid exactly that. If we can (mostly) agree that the state will punish someone sufficiently and appropriately, there's no need for vendettas.

Pardon my ignorance, but if utility is the test for determining what is good and bad, wouldn't having those labels provided as input cause a sort-of recursive formula where one needs to bootstrap the whole thing on some other ethical system?

The state has no legitimacy to me because it uses suffering for retribution. It's inherently oppressive.

The alternative is for individuals to inflict suffering on each other, which is just vigilantism and mob violence.

No, the alternative is to not apply retribution. To me anyone who applies retribution is nothing more than a thug, whether doing at the behest of the state or their own accord.

Ok. Now convince everyone else of that.

Plenty of people are convinced of that - it's pretty much the position the linked post is arguing against, for a reason.

>> [...] even in the absence of any deterrent effect.

i think your points fall into that deterrence bucket though.

I don't think so. I'm not suggesting the use of violence as a deterrent[0]. My point is that the human urge to enact violence in response to wrongdoing has to be satisfied. If it's not satisfied by the state, it'll lead to mob justice and vigilantism.

[0] I think in many contexts it is, but that's distinct from my earlier post.

And we go back to revenge killings and mob justice, which have been determined to be worse than a reasonably functioning state (North Korea and Nazis Germany not being considered reasonable).

Reply on news.ycombinator.com