Hacker News Reader

Comment by splitrocket | Hacker News Reader

Humans consistently make the same mistake about understand the motivations, the inhibiting and promoting pressures of other people. We assume that everyone else is exactly like us. Therefore, when we build systems and institutions that attempt to alter behavior, we encode our own motivations and desires as the impetus.

I.E. why we have UX researchers, do user testing, A/B testing etc when we develop products. Our instincts about what other folks will pay for/want/value/use are rarely totally accurate.

“I don’t want to go to jail, therefore, jail will be a globally effective crime deterrent” is demonstrably false for this subset of the population.

From a purely behavioral science perspective, we know that humans deeply discount uncertain future downsides compared to short term certain upsides.

If you are interested in this sort of stuff, I strongly suggest reading Mark Kleinman, particularly his work on “swift, certain, fair” as an effective means to address this population and challenge.

View on HN

Prison is a deterrent, certainly, but it also exists to keep violent folks away from the rest of society so they can't do more harm.

This study indicates that incarceration, at least as it is currently implemented, is emphatically not a deterrent for repeat violent offenders.

As for the public safety element, yes, but these people mostly weren’t locked up for life. What this shows, at least, in my opinion, is that much more intensive psychological and therapeutic interventions are needed, because if anything, incarceration as it currently exists would seem to exacerbate the problem.

Or, more simply, you could incarcerate repeat offenders for longer periods of time.

What is the concrete benefit of simplicity in this context, and how do you quantify whatever that benefit is?

Becaue the cost is both clear and intensive: Depriving a human being of liberty for additional years or decades. This should be done only when absolutely neccessary, should it not? We disapprove of imprisonable offenses intensely, but these are still human beings, with rights. It seems to me a humane system does not inflict thousands of cumulative additional incarceration years merely to keep the implementation of justice “simple.”

Simple in the sense that if you can identify this 1%, keeping them locked up or otherwise away from the rest of society longer will very effectively reduce crime. There is no way that therapy is going to be as effective. We are talking about repeated violent offenders here, really dangerous individuals. Of course they are still people but their happiness is secondary to keeping them from repeatedly inflicting harm on innocents.

>keeping them locked up or otherwise away from the rest of society longer will very effectively reduce crime.

The problem with this is it only looks at the problem from a maintenance perspective rather than a preventative root cause analysis. Like adressing the relationship between cyclical poverty and cyclical incarceration

I don't think people liked how Three Strikes laws have turned out.

I guess you could lock up ~1% of the population indefinitely, but by the time you've gotten to that point something else has clearly gone wrong, since most affluent countries are able to achieve relatively low violent crime rates without such widespread use of incarceration. If we were talking about 0.01% hard-core violent criminals then sure, just incarcerate them, but 3m+ people?

As far as I'm able to find, only two large countries in history have ever locked up more than even 0.5% of their population for extended periods of time: the USSR for approximately 15 years during the height of the Stalinist period (late 1930s to early 1950s), and the USA for about 30 years, from the late 1980s to present. So I wouldn't consider that a "normal" approach to managing crime.

I don't think comparisons to "other affluent countries" are apt; the socioeconomic and cultural makeup of the United States is so starkly different from pretty much any other country (even our immediate neighbors to the north and south).

In particular, the institution of slavery and perhaps even more so, the culture of segregation, condescension, and disdain that followed from The Reconstruction onwards resulted in such drastic dehumanization, disenfranchisement, and demoralization of a huge percentage of the population that has not been seen anywhere else in the world.

No solution can be drafted that does not address this ugly stain on our history (continuing to the present) and still ever hope to work, and whitewashing history books and tiptoeing around this elephant in the room in the name of political correctness does us all a disservice.

It's been seen plenty of places in the world. Probably most, by area anyway. There are still places where a significant portion of the population are not only disdained and disenfranchised, but actively killed without much consequence: the Rohingya for a current example, or the lowest castes in India

Absolutely. But few would make comparisons or draw parallels to them, whereas they'll be quick to do so for the Nordic nations or similar.

That would work but incarceration is an unscalable solution. Some CA stats:

  - $71k/person/year
  - 130,000 inmates
  - 135% of capacity 
  - highest rate of incarceration in the world

We're maxed out - and all this doesn't even solve the root issues of mental illness and poverty. We need alternative ideas with better ROI.

Sure. But what are these alternatives? 1% of the population are psychopaths and no one knows how to make them not be so. What else can we do but put them away?

Either why aren't 1% of people psychopaths in other developed nations or alternatively why are these 1% behaving in other nations.

Maybe we should figure out why they are "psychopaths", rather than locking them up and throwing away the key?

Hahah, apparently this is a controversial suggestion. Never realized HN is so right-wing.

The easy solution is prison but its also crazy expensive both in terms of raw cost and societal cost (broken up families, addiction, developing gang ties in jail, recidivism, opportunity cost of taking a working age person out of the job pool).

IMO there is no politically easy solution or we would have already done it. Society has to change how it views criminals and invest resources in the things that increase the likelihood of crime. For example, rich psychopaths commit a lot less violent crime than poor psychopaths. Social services, addiction counseling, education, etc help. But its extremely hard to convince people more resources should go to helping criminals vs punishing them.

Well, Gulag was profitable AFAIK

Studies show that things like three strike laws do reduce crime rates. But, for example in California, it costs $70k per year to jail them. Intensive outpatient therapy costs half as much. If we moved to two strikes and then mandatory high expensive therapy it could be more humans and cost effective.

We've taken that approach for quite a while in the US and the results are bad, to put it mildly.

"Our study suggests that 3 months of imprisonment in an impoverished environment may lead to reduced self-control, measured as increased risk taking and reduced attentional performance. This is a significant and societally relevant finding, as released prisoners may be less capable of living a lawful life than they were prior to their imprisonment, and may be more prone to impulsive risk-taking behavior. In other words, the impoverished environment may contribute to an enhanced risk of reoffending."

https://www.frontiersin.org/articles/10.3389/fpsyg.2018.0006...

You're missing the other half of the picture. How many were deterred because of prison? Would it have been 2 or 5 or 10% were it not for the deterrents? You can't say.

You're assuming people use logical, rational thought which has ample amounts of evidence against it. Jails used to be horrific, yet crime rates were higher. The presence of a jail sentence or even a death sentence as a deterrent is widely discredited. The only uses for prisons are to prevent future crimes by removing the ability to perform crimes and intensive rehabilitation. Most former convicts do not want to reoffend but the American social and legal system of punishing people for their entire lives for offenses committed once prevents effective reintegration into society.

All the science points to preventing initial conditions for conviction and significant improvements needed post release.

http://www.slate.com/articles/business/the_dismal_science/20...

https://nij.gov/five-things/Pages/deterrence.aspx

https://www.nij.gov/topics/corrections/recidivism/Pages/welc...

The problem is that they become worse in prison. I’d be for a policy where once you went to jail you were never let out into the general population — you’d have to go live on an island with ex-prisoners.

Sort of like transportation?

Most criminal cases don't get to a conviction. Ask a police officer. Save for very few countries (e.g. Japan)

Let me put it this way: the current “standard” criminal justice system is exceedingly effective at dissuading the sort of people who designed and operate the system from committing crime, but those folks are not the population that in fact commit crimes.

> exceedingly effective at dissuading the sort of people who designed and operate the system from committing crime

More like the people who designed and operate the system consider everything they do legal and only criminalize the activity of people not like them.

> More like the people who designed and operate the system consider everything they do legal

This is true by definition. Anyway, what sort of violent crime do you propose to legalize?

Here is one that is fraught where I live - disciplining children with a smack. Many people think hitting children in a disciplinary context should be legal.

Hitting or smacking? Smacks are a very limited subset of hits.

How do you define a “smack” versus a “hit”?

You probably also need to consider things like how often is physical discipline being used, and is it actual discipline or just an excuse for abuse.

What is more wrong -

a parent physically abusing their child when they feel like it (bear in mind the parent is also taking care of said child)?

a parent bound by society reduced to moaning at their child (think kids under 10) hoping to control them?

I know which is more perverse.

You don't.

All forms of hitting children are abusive.

Shouting at, or ignoring, children is also (emotionally) abusive. Should we lock up parents who ever act in one of those ways towards their children?

Also, what about when children hit their siblings, or their parents, or their class mates? I don't think that locking these children up is conducive to their welfare either.

If one does neither it doesn’t matter.

Bar brawls between consenting adults.

It'd be interesting to see how often the state presses assault charges without much participation from one of the parties involved.

Not often unless they already have a reason to target the person (repeat offender, wanted for another crime they can't prove, etc).

I remember being in Texas and seeing on the news how a guy shot another guy in a bar and they were not prosecuting because the person who was shot was not pressing charges, but that's Texas.

To play devil's advocate, maybe the system is very, very good as evidenced by all of the people who aren't running around committing crimes. In other words, survivorship bias.

To verify if this is the case, you should compare incarceration rate and crime rate with otherwise similar countries.

Tautology - People who are effectively deterred from committing crimes do not commit crimes.

The system may be very effective at deterring many would-be criminals. Any system will be ineffective against some subset and so this statement would still apply.

Under any system, most crime will be committed by those whom that system does not effectively deter.

Yes, true.

So, in order to improve the system and reduce crime, you need to understand those who are currently not effectively deterred.

More of the same is not going to work if you hope to improve outcomes.

But you don't want people who currently aren't committing crimes to begin doing so. Who's to say that the current deterred subset of the population won't begin to commit crimes if our style of punishment is changed?

That seems to start from the assumption that everyone would commit crime if they could.

They specifically referred to the "deterred subset" ... which is much smaller then "everyone".

In any case, the parent's point seemed pretty empty.

> Who's to say that the current deterred subset of the population won't begin to commit crimes if our style of punishment is changed?

This is what research is for. We may never know what is optimal in an absolute sense, but we can figure out some relative costs/benefits of different approaches to problems and choose our tradeoffs appropriately.

Beautifully parsed and simplified.

But this only applies to premeditated crimes, where the system is a factor that goes into deciding whether to commit the crime.

Crimes of passion (i.e. many violent crimes, especially) don't take the system and its deterrents into account.

In the minds of many people the purpose of the system is not entirely to deter crime; some criminals will not be deterred and the secondary purpose of the system is to segregate such people from society by placing them into little boxes until the system decides they are no longer a danger to the rest of us.

So are you saying that the solution is to lock up these people for life?

Because I hope that’s not what you are saying.

You hope that’s not what the commenter is saying, but what is your proposal instead?

We can put significant effort into understanding criminals, for the purposes of both preemptive treatment and reactive rehabilitation, but this is not likely to eradicate crime. What do you suppose we should do with criminals who continue to commit violent or disruptive crimes in perpetuity?

To clarify, I think the system could improve: extremely long sentences may not need to be issued for first time offenders, even if they are very serious. But I don’t think the system can remove permanent sentences if criminals exhibit a clear predilection that is resistent to all known forms of treatment.

http://www.swiftcertainfair.com

Been studied, replicated, and thoroughly tested.

It’s just one option as well.

There are many many more.

Do you mind summarizing what that is a bit? I’d like to read about it, but either the link isn’t working for me (I’m on mobile) or it doesn’t contain much information aside from the homepage.

Copy pasting from their website:

About one-third of probationers and parolees fail the terms of their supervision. Over three-quarters of parolees are re-arrested within five years, and over half return to prison. High failure rates for people on probation and parole—whether for new offenses, revocations, or individuals who abscond from supervision—result in increased crime, crowded prisons and jails, and strained public budgets. Indeed, every single one of the seventeen states that have undertaken justice reinvestment since 2010 has found probation or parole revocations to be an important driver of their growing prison populations. And despite a variety of local, state, and Federal initiatives over the last two decades, failure rates for supervised individuals have not meaningfully fallen.

Why Swift Certain and Fair? Because—if the conditions are right—a SCF program can take high-risk individuals in your jurisdiction and substantially reduce their drug use, revocation and re-arrest rates, and the subsequent reliance on incarceration.

Swift: SCF programs deliver a sanction immediately upon detection of a violation, which is critical in changing behavior. Responding swiftly also supports perceptions of the fairness of the program, reinforcing that element of SCF programs.

Certain: The consistency and predictability of punishments make the consequences of bad behavior clear to the offender, reinforcing the need to make better decisions and change behavior. An essential role is played by the clearly-defined behavioral contract with the offender—a tool that has repeatedly been proven to increase perceptions about the certainty of punishment.

Fair: The clear demarcation of the offenders’ new supervision conditions and the opportunity for a fresh start allow the offender to regain their sense of self-control. The contract enhances the perceived fairness of any subsequent punishment; this perception of fairness is further supplemented by the sparing rather than harsh level of punishment upon violation, which additionally limits disruption of employment and other non-criminal routines and relationships.

> and substantially reduce their drug use

Drug use is a symptom, the result of another problem. Consider, for example, an undiagnosed personality disorder, or an untreated childhood trauma. Furthermore, what I'm going to mention next isn't helping that either.

> the opportunity for a fresh start

That's just marketing shenanigans. In reality this is rather relative; if one has become a convicted felon, they remain a convicted felon for life. This haunts them till the end of their time. The stigma hampers functionality in society in multiple ways: friendships, and employment first and foremost, but also hypocritically the 2nd amendment (of which I as a European am no proponent of, but what's sauce for the goose is sauce for the gander), and the loss of other rights [1]. This society inhibited perfectionism inhibits a defeatist attitude which lures a convicted felon back on the dark path of criminal behaviour on top of things like personality traits or already financial issues. Prisoner series like Orange Is The New Black (though certainly not the first!) beautifully articulate this specific problem, and a myriad other.

The problem is that we see criminals as individuals who cause an issue in society. Instead, we should regard them as individuals who are unable to participate in society and look at it from that angle, by trying to make them participate in society. Getting rid of the defeatist attitude is one way to increase participation in society, looking for medical treatment. Some examples of problem solving procedures (vary on individual basis) are dental procedure (for e.g. former meth addicts), basic insurance, medicine/drugs (methadone for e.g. former heroin addicts), psychological therapy.

I don't want to be negative about the program, though:

> Fair: The clear demarcation of the offenders’ new supervision conditions and the opportunity for a fresh start allow the offender to regain their sense of self-control. The contract enhances the perceived fairness of any subsequent punishment; this perception of fairness is further supplemented by the sparing rather than harsh level of punishment upon violation, which additionally limits disruption of employment and other non-criminal routines and relationships.

This puts the other parts into perspective, making it indeed more fair for the convicted felon (or less unfair, depending on your perspective). That doesn't mean the sentence itself isn't still out of proportion; it already is put out of proportion because there's societal punishments on top of the sentence. Some of which are harmful for rehabilitation.

Incidentally, fixing this might also make things like arrests generally less violent.

[1] https://thelawdictionary.org/article/what-rights-do-convicte...

Do any among the "many more" studies address the recidivist, violent 1% who are the issue here? This one does not, and what this discussion needs is more relevant information.

The way society treats recidivism, with a series of escalating time-outs, is very similar to how we treat questionable behavior in computer security with rate-limiting. It's not a binary choice where you either drop all pretense of security or have an oracle to always know exactly who should be kept locked away. Rather, there are benefits and trade-offs balanced in the design of each system, including the understanding that there is limited knowledge to go on at each enforcement point.

I think it is interesting to find such parallels in the design features of such a wide range of systems as courts, prisons, network firewalls, and authentication systems. One can imagine various interpretations in response to these parallels: on one extreme, the designs could demonstrate common cognitive biases of the humans designing them; on the other, the functioning systems could be illustrating natural truths that have been discovered through the evolution of these systems...

Lots of people (in the US at least, not, say, Sweden) care more about revenge and suffering inflicted on the criminals than rehabilitating them (or even preventing them from doing more harm -- they are e.g. totally ok if they do more harm, but inside, to other inmates).

What I hope is that those people get to experience prison for inside, and even better, for some BS reason, like the reasons tons of people are in jail for (possession of marijuana, breathing while black, and so on).

Isn't there a bit of irony there? In that you want to send the send-people-to-prison-to-suffer people to prison to suffer?

Well, as long as they're promoting the idea they may as well get a taste of their own medicine. It's not like they don't have access to alternative arguments based on empathy or moral principles or empirical data showing that their highly punitive approach works poorly.

Its totally in keeping with the revenge is more important than results mindset.

Let's call it poetic justice, or taste of their own medicine.

You sound just like the people you condemn.

And yet only one side (between me and the people I condemn) actually controls what the law is like and helps inflict pain and harsh conditions upon millions of people serving time.

So what I "sound like" is quite irrelevant. Except if you believe that people who cherish revenge and punishment of inmates will one day replace the people population based on my wish.

> until the system decides they are no longer a danger to the rest of us.

Which kind of implies that one of the goals of prison is/should be to move to a state where they are no longer a danger.

>we encode our own motivations and desires as the impetus

I don’t see how that’s true. Wouldn’t it be more accurate to say, the system was encoded based on what the designers knew about motivation and desires at the time?

The fact that what they knew was based on their own motivations seems incidental. I doubt if they had the perfect formula evidenced and understandable to them at the time, that they would have intentionally ignored it to still encode it in their own image.

So we can really boil it down to, people didn’t understand human nature as well when these systems were created. Then, even as better understanding evolved and continues to improve, the problem of inertia and entrenched thinking works against change.

From that we can conclude one of the biggest problems with the criminal justice system is, simply that it’s impeded in the exact same way so many other systems are.

edit after comment: Point is the description of the problem seems to imply pretentious motivations, when the simplest explanation is they just didn’t know how things worked at the time.

> The fact that what they knew was based on their own motivations seems incidental. I doubt if they had the perfect formula evidenced and understandable to them at the time, that they would have intentionally ignored it to still encode it in their own image.

My understanding of the comment you’re responding to is that the nature of human experience precludes that ideal understanding, not that people would purposely ignore it. The point appears to be that humans encounter intrinsic difficulty in understanding the myriad idiosyncratic motivations of others in the species, not that they deliberately don’t want to understand it.

In other words: I don’t think you and the parent commenter are disagreeing.

I certainly disagree the nature of human experience precludes ideal understanding.

My point is the comment suggests almost an egocentric process, i.e. modeling a system in our own image, when actually it’s explained by simple ignorance.

Right, I follow what you’re saying. My followup to that is that I think (personally, and what the commenter is saying) that ignorance in the modeling of human behavior both causes and is caused by egocentrism in a feedback loop.

In any case, you disagree with that, so I suppose you are disagreeing with the parent; my interpretation of the parent commenter’s point was that this ignorance you speak of is pretty tightly coupled with innate egocentrism. I don’t think they’d consider that necessarily malicious if they consider it from that framework, whereas you might consider it malicious if the process is egocentric because you disagree with the same relationship; vis-a-vis ignorance and egocentrism.

Okay, so they didn’t understand.

Does that preclude us from trying to fix the problem?

On the contrary, it should add to our motivation. It’s just usually most productive to target outrage as accurately as possible. Not just because it’s fair, but because outrage is a scarce resource. To fight institutional injustice requires wasting as little energy as possible on the wrong targets.

From a purely scientific point of view (since that is what you are espousing), you can't make the comment you just did without proving that without the threat of jail time this subset of the population would not be committing even more crimes even more often.

Additionally, speaking as someone that agrees with the scientific premise of your argument, any alternative would be a trade-off, i.e. the percentage of the population that are not committing crimes at all would need to remain sufficiently deterred, which may change when the deterrent does.

Reply on news.ycombinator.com