Why is Blackstone's Ratio taught as a "principle"?

[Francesca R]

I'd appreciate a sceptical consideration of this.

Blackstone's ratio, after William Blackstone (http://en.wikipedia.org/wiki/William_Blackstone), is the commonly heard statement that "It is better that ten guilty persons escape than that one innocent suffer", and it has been taught and learned by scholars of criminal law in most civilised societies for two hundred and fifty years. Accordingly it shapes the design of criminal justice systems (statutes, rules of evidence)

So the question is "Why?" What are the costs and benefits (the trade-offs between freeing the guilty and convicting the innocent)? And does it really mean "ten" or "any number"? (I suspect "any number" renders the idea absurd)

My own suspicion is that this represents society grabbing hold of a number, that they can cite a legacy authority for thinking up (except that it was not really "thought" up), and thereby avoid difficult moral questions of the value (price) of liberty and the cost of unconvicted criminal activity.

Thoughts?

[Wowbagger]

I don't think the specific number is important. Just the concept. Though, the calculation costs vs. benefits might be a valid argument against it.

It might depend on the level of crime. 10 mass murders going free would only lead to killing of many more innocent people, than if we killed one innocent person while killing the other 10.

But, if the choice was sending 1 innocent person to jail (along with 10 guilty) or allowing 10 petty-cash theives going free (along with the 1 innocent person), I think justice would usually favor the 10 going free.

But, you can replace that number "10" with anything you want, to test what is acceptable, in a given set of circumstances.

[Giggywig]

I'd appreciate a sceptical consideration of this.

Blackstone's ratio, after William Blackstone (http://en.wikipedia.org/wiki/William_Blackstone), is the commonly heard statement that "It is better that ten guilty persons escape than that one innocent suffer", and it has been taught and learned by scholars of criminal law in most civilised societies for two hundred and fifty years. Accordingly it shapes the design of criminal justice systems (statutes, rules of evidence)

So the question is "Why?" What are the costs and benefits (the trade-offs between freeing the guilty and convicting the innocent)? And does it really mean "ten" or "any number"? (I suspect "any number" renders the idea absurd)

My own suspicion is that this represents society grabbing hold of a number, that they can cite a legacy authority for thinking up (except that it was not really "thought" up), and thereby avoid difficult moral questions of the value (price) of liberty and the cost of unconvicted criminal activity.

Thoughts?

I have no expertise whatsoever in the area, but if I had to guess it's more in line with trying to not give the government free hand in incarcerating innocent citizens, so you place certain safeguards in that regard.

[Beerina]

It was born in an era of very poor scientific forensics, combined with an era of jailing being used very heavily for political oppression.

"The People" don't hire a government to jail innocent civilians. They hire a government to jail criminals to laws they determine.

Ironically, while it would be scientific to do a study that shows you get the best bang for the buck by jailing N innocent people for every M guilty ones that go free, that's only fine insofar as you're not one of the innocent ones.


I once encountered an anarcho-capitalist (a libertarian on steroids) who said, in a similar question about the death penalty, that the very occasional execution of an innocent was worth it to society for the benefit of the execution of murderers.

I accused him of being socialist for justifying a massive wrong by its benefit to the masses. Chastened, he returned to his hole in a big tree, nailed the door shut, and hasn't been hear from since, though his electricity and Internet bills somehow get paid regularly.

[Francesca R]

I don't think the specific number is important. Just the concept.Yes but the concept without a number implies it also is better for 100 or 1000 guilty to be acquitted than for one innocent to be convicted. In the limit that means acquit everybody, but then you have an adverse consequence that the equilibrium number of criminals is apt to rise because the net benefit of crime increases.

Though, the calculation costs vs. benefits might be a valid argument against it.That's what I thought, not that it is easy to do.

It might depend on the level of crime.Probably. So we could restrict it to just one crime--say, burglary of £50,000 of someone's belongings--and one penalty for the convicted, which is, say, three years in prison. Then what?

[roger]

I read it as a pretty simple "the ends don't justify the means" argument. It doesn't seem to be making any kind of utility argument. "hey, let's round up all the Japanese and put them in a camp until the war is over, just in case". Uh, no, let's not.

[Francesca R]

"The People" don't hire a government to jail innocent civilians. They hire a government to jail criminals to laws they determine.Yes. So this is about what the ex-ante tolerance for error might be. And that must be larger than zero because otherwise the government would not be given the authority in the first place.

Ironically, while it would be scientific to do a study that shows you get the best bang for the buck by jailing N innocent people for every M guilty ones that go free, that's only fine insofar as you're not one of the innocent ones.So to do an objective study, you should assume you will not be an unlucky loser (or a lucky winner--getting away with robbery) of the system thus conceived.

I once encountered an anarcho-capitalist (a libertarian on steroids) who said, in a similar question about the death penalty, that the very occasional execution of an innocent was worth it to society for the benefit of the execution of murderers.Implicitly it must be (for states that have the death penalty), or else they would abolish it.

[Francesca R]

I read it as a pretty simple "the ends don't justify the means" argument.As in "convicting ten guilty people [=good] does not justify convicting one innocent [=bad] if that is going to be what it takes"? Yes that's a way of stating the same thing, which leaves the "Why?" question still there.

[fuelair]

So the question is "Why?" What are the costs and benefits (the trade-offs between freeing the guilty and convicting the innocent)? And does it really mean "ten" or "any number"? (I suspect "any number" renders the idea absurd)

My own suspicion is that this represents society grabbing hold of a number, that they can cite a legacy authority for thinking up (except that it was not really "thought" up), and thereby avoid difficult moral questions of the value (price) of liberty and the cost of unconvicted criminal activity.

Thoughts?[/QUOTE]
One reason I can think of (and it would go under the costs part): if I am ever convicted of a crime I did not do then when I get out of jail(whatever) there will be crimes committed including jury members, investigators, DA, judge etc. If I get caught, well I just go back in.

[Francesca R]

One reason I can think of (and it would go under the costs part): if I am ever convicted of a crime I did not do then when I get out of jail(whatever) there will be crimes committed including jury members, investigators, DA, judge etc. If I get caught, well I just go back in.Right, as in: "Convicted innocents are likely to become turned into future criminals" (go on a rampage of vendetta or something).

It is a bit of a stretch to imagine that such a concern would outweigh the likelihood that ten free already-criminals would inflict societal costs with more crime.

[jimbob]

Insn't it in part because of the difference between an active wrong and a passive wrong?

The state is committing an active wrong in incarceating an innocent person, whilst letting a guilty person off, is a sin of omission.

You might think that this doesn't matter, but *emotionally* it does seem to. I have heard this stated in an extreme hypothetical case:

Suppose you could save some people, but by doing so would doom other innocent people, would you do it?

Scenerio 1:
A train with ten people and a train with one person are both rushing towards a bridge that has collapsed, you can only get to the signal to stop one train, which do you chose?

Scenerio 2:
A train with ten people is rushing towards a bridge that is collapsed, and you could stop this train by killing someone in some enormously implausible and hypothetical manner. Do you kill this person, knowing it wil save more lives?

Most people do the simple sum of the greater good in the first case, and not in the second case.


Maybe we could make this a bit less hypothetical:

Would you condone shooting a Brazilian electrician, to prevent a suicide attack, that turned out to be nonexistant in retrospect?

[qwints]

I'm so glad I only took one class in philosophy - trolley problems are hard.

I've always read the Blackstone quote as setting forth the justification for the legal standard of "beyond a reasonable doubt." Read literally, it means that you should be more than 90% sure of a person's guilt before you convict them. With all due respect to Beerina, it really has nothing to do with forensics. Forensics are only as good as the labs that administer them. Fingerprints, for example, seem to only be about 80% accurate http://www.brooklaw.edu/students/journals/bjlp/jlp13i_zabell.pdf.

The basic reason for Blackstone's ratio is that most Anglo-American legal theorists are negative retributivists. That is, they believe that punishment schemes should do the greatest good for the greatest number while not harming an innocent. Note that avoiding harm to an innocent is not a consequence of a utilitarian philosophy at all. It can probably be justified by appeals to compassion or perhaps by a Rawl's style veil of ignorance.

[Earthborn]

It might depend on the level of crime.I think it also depends -- and perhaps more importantly -- on the severity of the punishment. In a society where the role of the justice system is not viewed to be to retaliate, but by protecting society by isolating dangerous individuals, and in which prison is designed not to cause suffering, the equation will look very different. If an innocent person does not suffer by being in prison and can be freed when new evidence comes to light, then it may be considered worth it to lock up an innocent person instead of letting ten guilty person's go free.

[jimbob]

Further to my previous points: one is a specific harm to a particular inndividual, any cost-benefit would be a general reduction in crime-levels (if it worked).

I doubt that there is any evidince that banging-up innocent people reduces the crime rate.

I also thought that legal punishments are generally supposed to be punishmnets for crimes that have been committed, and not "protecting the public" by keeping criminals off the streets. I thought that the protection was supposed to be in the deterrent effect, not the administration of justice.

This argument becomes far more important when you are discussing capital punishment.

Which is where I have heard the statment actually applied:

"Better a hundred guilty people go free than one innocent person is hanged"

[Careyp74]

it has been taught and learned by scholars of criminal law in most civilised societies for two hundred and fifty years. Accordingly it shapes the design of criminal justice systems (statutes, rules of evidence)

Thoughts?

What exactly is being taught and learned? The quote, perhaps, but there is no actual ratio, they just call it that. The idea is that the system is set up to keep the innocent from being punished, so it is ok that it isn't tough enough to get every criminal. 10 is merely an arbitrary number thrown in to make it a better sounding quote. There is no rule stating that for every 10 convictions, there should be one aquittal.

This stems from the bible, in many different passages, which is probably where he got the number 10 from.

[Uncayimmy]

So the question is "Why?" What are the costs and benefits (the trade-offs between freeing the guilty and convicting the innocent)?


If you're talking purely a monetary cost/benefit ratio, then really the most cost effective method would have extreme and violent punishments for people who just *seemed* likely to have committed the crime...and their families. Administrative costs would be minimal and the deterrent would be very strong (public floggings of an alleged thief and the immediate family).

I bring up this extreme example because your most likely gut reaction is that this is wrong. Thus there is some inherent moral value to not punishing the innocent.

In my mind Blackstone's statement is simply a reminder that as a society our goal is to protect the innocent. This means not only protecting the innocent victims of crimes but also the innocent people who had nothing to do with it. We consider kidnapping to be a heinous crime. As a practical matter locking up an innocent person is no different than kidnapping to the person being locked up and to those who would want to have contact with that person.

Thus the judicial system should take steps to ensure that convicting the innocent is an extremely unlikely event just like the first rule of medicine is to do no harm.

Another practical aspect of this is that the more likely it is for an innocent person to be convicted, the more likely it is that the government will abuse its power. Making accidental wrongful convictions more difficult has the added benefit to making intentional wrongful convictions less likely to be attempted and less likely to be successful.

To me the numbers don't have a specific meaning. He could have just said that it is far worse for society to convict the innocent than it is to let the guilty go free.

As a practical matter the numbers are irrelevant. If you *know* that 1 in 20 convictions is a wrongful conviction, then you can overturn those convictions and eliminate the problem.

His statement is just a guideline for forming laws.

[Professor Yaffle]

I just see it as stating that because the system is not perfect, the evidence needs to weigh very heavily towards a person's guilt before we convict. Its why a conviction needs to be beyond reasonable doubt, rather than based on the balance of probabilities, or on mere suspicion or rumour.

That may just be restating the principle, but if you are asking me why this is the case, I can only ask you whether you think it is reasonable to lock up everyone who is ever suspected of a crime? If you don't think it is reasonable, then you can ask yourself the question of why you believe this. If you do think it is reasonable.... I'm worried!

I think pretty much every person would draw the line somewhere, otherwise pretty much everyone could be locked up on some premise or other. We just differ in where we draw the line - in how many innocent people jailed (or killed etc) makes it worth it in order to rightly imprison (or kill) a guilty person.

[Wolfman]

I've commented on similar discussions before -- in China, the perception is pretty much the opposite, and not just from the gov't's perspective. The vast majority of Chinese feel that one of the main responsibilities of the gov't is to protect them, and if it is necessary to imprison a few innocent people in order to be sure that one can provide greater protection, that is a necessary sacrifice.

Several years ago, China instituted the concept of "innocent until proven guilty" to all non-capital crimes (still doesn't exist for capital crimes), to much praise from the West. But unreported in most Western media, the vast majority of Chinese were opposed to this change.

Much of it comes down to where one draws the line of "reasonable doubt"...at what point does one determine that there is enough evidence to find someone guilty?

In the West, we tend to focus more on avoiding by all possible means imprisoning someone who is not guilty; in China, the majority of people focus rather on avoiding by all possible means letting a criminal go free. In the West, we view the imprisonment of an innocent person as damaging to society; in China, they tend to view the release of a criminal as damaging to society.

There are arguments on both sides. If there is a rapist who is almost certain to rape again, and whom we have evidence to demonstrate to around 80% certainty that he is guilty...but we let him go free because of the 20% "reasonable doubt"...does that not do significant damage to society? Particularly to the subsequent rape victims, and their family/friends, who suffer at his hands?

But then, with a lower standard of proof, it is easier for those in power -- government, police, etc. -- to abuse that lower standard in order to target people they don't like, regardless of guilt or innocence (and this certainly happens in China).

I had a discussion about this with a group of Law post-graduates that I was teaching at Beijing University many years ago, and one of them suggested what I found to be a fairly novel approach...kind of bringing both systems together.

Have a lower standard of "reasonable doubt", allowing the imprisonment of more guilty people, removing them from the streets, and protecting the general populace. But in cases where it is later discovered that an innocent person was imprisoned, those responsible for imprisoning that person must face trial with the same standard of "reasonable doubt" as to whether or not the imprisonment was justified. And if it is found -- within that loose standard of "reasonable doubt" -- that they engaged in unfair practices (falsified evidence, withheld evidence, etc.) in gaining that conviction, they will face severe consequences.

He suggested that in cases where the innocence or guilt of a suspect was really in question; or in cases where authorities were simply trying to railroad someone they didn't like; that such a system would provide a significant disincentive to abusing the system, since they could later face prosecution with the same low standard of reasonable doubt regarding their actions.

I'm not advocating for this idea, just throwing it out there...I have my doubts how well it would work. But thought it would add new fodder to the discussion.

[69dodge]

If you're talking purely a monetary cost/benefit ratio, then really the most cost effective method would have extreme and violent punishments for people who just *seemed* likely to have committed the crime...and their families. Administrative costs would be minimal and the deterrent would be very strong (public floggings of an alleged thief and the immediate family).

I bring up this extreme example because your most likely gut reaction is that this is wrong. Thus there is some inherent moral value to not punishing the innocent.

Yes, of course harming the innocent is bad. (Otherwise, why bother even trying to punish the guilty? All they're guilty of is harming the innocent.) I assume Francesca intended that any punishing of innocents be counted as a cost.

[Uncayimmy]

Yes, of course harming the innocent is bad. (Otherwise, why bother even trying to punish the guilty? All they're guilty of is harming the innocent.) I assume Francesca intended that any punishing of innocents be counted as a cost.

You said "harming" the innocent while I said "punishing" the innocent. There's a distinction. Not everyone views these two ideas the same way. Group punishments are common in school. A teacher may find it unacceptable to keep the class inside writing an essay during recess for no good reason. That's harming the innocent. However, it might be acceptable to punish the entire class because a few students misbehaved. That's harming *and* punishing the innocent along with the guilty. Likewise, punishing just the wrong students is also harming and punishing the innocent, but at least it was probably done in good faith.

I included the family being punished to further push the issue. To use the example above that would be like punishing students from another class that didn't misbehave.

It's all a rather long winded way of saying that Blackstone is trying to establish that punishing the innocent under any circumstances is much worse than letting the guilty go free. I figured anybody who didn't get that didn't see much of a problem with punishing the innocent. Maybe I misunderstood.

[Nyarlathotep]

No one has ever come up with a 100% accuarate and fair system of justice that always punishes the truly guilty and always exonerates the truly innocent. A society CAN, however, decide on whether it wants to be more likely to err in favor of punishing the guilty at the expense of unjustly punishing the truly innocent or in favor of exonorating the innocent in favor of occasionally letting the guilty go free.

'Blackstones Ratio" is just a way of stating a preference for the latter. the exact numbers don't mean a thing, IMO.

[Uncayimmy]

No one has ever come up with a 100% accuarate and fair system of justice that always punishes the truly guilty and always exonerates the truly innocent.

I'm working on it.

'Blackstones Ratio" is just a way of stating a preference for the latter. the exact numbers don't mean a thing, IMO.

I wouldn't say they don't mean a thing. Clearly he's trying to establish much more than just a preference. Granted, it's not quantifiable numerically, but the expression "ten times worse" usually conveys a severe difference.

[Ivor the Engineer]

This issue is formally defined for medical screening tests. The relevant terms are:

Sensitivity, Specificity, Positive Predictive Value (PPV), Negative Predictive Value (NPV) and Prevalence.

The sensitivity of a test is the proportion of true positives which are identified by the test. I.e. sensitivity = TP / (TP + FN).

The specificity of a test is the proportion of true negatives which are identified by the test. I.e. specificity = TN / (TN + FP).

The PPV is the proportion of positives identified by the test which are true positives. I.e. PPV = TP / (TP + FP). Also PPV = (sensitivity).(prevalence) / [(sensitivity).(prevalence) + (1-specificity).(1-prevalence)].

The NPV is the proportion of negatives identified by the test which are true negatives. I.e. NPV = TN / (TN + FN). Also NPV = (specificity).(1-prevalence) / [(specificity).(1-prevalence) + (1-sensitivity).(prevalence)].

Prevalence is the proportion of the population known or believed to have the disease.

For tests used for mass screening of disease, such as those used for bowel cancer or HIV, it is generally preferred for the test to have a high negative predictive value (i.e. a negative result on the test makes it very unlikely a person has the disease), at the cost of a low positive predictive value (i.e. a positive result on the test does not give much confidence a person has the disease).

What is not obvious initially is the prevalence of the disease affects both the PPV and NPV. When the prevalence is very low, even very sensitive and specific tests can result in low PPVs. When the prevalence is very high the situation is reversed and the NPV of a test suffers.

This can be related to the trial process by considering the known or estimated prevalence of guilty defendants when setting the threshold for the burden of proof to keep the PPV constant. In trials for crimes where the prevalence of guilty defendants is known or believed to be high, the burden of proof can be set lower than it is in trials where the prevalence of innocent defendants is known or believed to be high.

The estimated prevalence of guilty defendants for any particular crime is what people seem to disagree on. E.g., male on female rape vs. female on male rape. Buried in this is also the estimate of prevalence of guilty defendants in various sub-groups. E.g., defendants accused of raping an acquaintance after consumption of alcohol/drugs and heavy petting compared to defendants accused of raping a stranger at knife-point. It all gets very subjective very quickly.


Note: TP = True Positive. FP = False Positive. TN = True Negative. FN = False Negative.

[CFLarsen]

It isn't a question of 10, 100 or any number of criminals walking.

The question is: Can a society live with just one innocent person being punished for a crime he didn't commit?

If you say "yes", then state:

1. The highest ratio of innocent citizens that is acceptable to be unjustly punished.

2. The reasoning behind why this should be so.

3. How you will implement it.

4. Why it is "justice".

5. What you would feel like if you were the innocent being unjustly punished.

Those promoting the idea that "it is better to punish more, if only we can get all the criminals" always assume that they themselves will never be among those unjustly punished. It is always the other guy.

Kinda like those who think they will never be victim of a crime or a traffic accident.

Ultimately, it is an argument of ignorance. Only this one paves the way for dictatorships: It will in the end be up to your rulers to decide who among the innocent population are punished.

And there ain't no way you can prevent it - because you said it was OK.

[Francesca R]

Can a society live with just one innocent person being punished for a crime he didn't commit?What is the cost to society of this happening? Do you think the cost is "infinite"?

[CFLarsen]

The cost is that there is no justice - at all.

Anyone can be punished for a crime he didn't commit. At any given moment.

[Francesca R]

One innocent person punished = no justice at all?

I don't agree with that.

What would be the solution if it was true? Abolish punishment seems to be the only viable one. Is there another?

[Wudang]

But the law is not just supposed to protect society - it is also supposed to protect the rights of the individual.
If we were guided purely by costs (financial and to society) then shooting the few people with AIDS rather than treating them would be a good idea (one of my better strawmen I think).

[CFLarsen]

If one innocent person punished means that there is still justice, what is the highest ratio of innocent citizens that is acceptable to be unjustly punished?

[Mashuna]

It isn't a question of 10, 100 or any number of criminals walking.

The question is: Can a society live with just one innocent person being punished for a crime he didn't commit?



Surely all societies have had innocent people punished for crimes they did not commit? Not as a deliberate choice (we know the murderer was someone in this room, so we're going to lock you all up), but certainly for other reasons.

Society seems to be coping ok with it.

[CFLarsen]

It isn't a question of whether or not societies have had innocent people punished for crimes they did not commit.

It is a question of how many innocent people is acceptable.

[Francesca R]

If one innocent person punished means that there is still justice, what is the highest ratio of innocent citizens that is acceptable to be unjustly punished?Your question does not follow from the premise, and is a non-sequitur.

If one innocent person is punished and ten guilty ones are punished too, there is not "no justice". Is there?

If there was "no justice", in that situation, then it is indistinguishable (justice-wise) from one where one innocent person is punished and ten guilty ones go free. Can you distinguish between those two?

[Ivor the Engineer]

Surely all societies have had innocent people punished for crimes they did not commit? Not as a deliberate choice (we know the murderer was someone in this room, so we're going to lock you all up), but certainly for other reasons.

Society seems to be coping ok with it.

Only because we have people who are today still wanted by the government, surviving as soldiers of fortune. If you had a problem, if no one else can help, and if you can find them, maybe you can hire...

[CFLarsen]

The question is not a non sequitur. It is crystal clear logic:

If one innocent person punished, there is still justice.

OK.

One.

The line is drawn at least at one. So, where is the line drawn, exactly?

What about two?

What about three?

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?

I can understand why this is an uncomfortable question. But if one is OK, then we have to know how many in all.

It is an unavoidable question.

[Francesca R]

If one innocent person punished, there is still justice.

OK.You said there is "no justice". We are still on that before you move on.

If there was only ever one crime, and an innocent person was punished for it, there would have been no justice. You would be right. But there are more crimes, and guilty people are punished.

Please explain why regardless of how many guilty are punished, one punished innocent means "no justice" because that is what I think you said.

If you do not mean that--please clarify.

Thanks

[D'rok]

The way that principle was explained to me is that it is an expression of the need to protect the individual from the state. It is an important liberal ideal. There is a massive criminal justice apparatus that is wielded against an accused. All of the inertia of the system is oriented towards conviction, even if that isn't the explicit intent - once the battleship gets under way, it is hard to steer or stop.

In a just liberal society, Blackstone's ratio reminds us that, when the full weight of the state is bearing down upon one of us, we bloody well better be careful not to let it crush that person, even if it means we miss a few that "should" be crushed. The "why" question is really questioning our fundamental notions of liberal justice, not any particular number or ratio.

Doing a utilitarian cost-benefit analysis misses the point, as it usually does. Justice cannot be calculated.

[CFLarsen]

We are discussing if it is acceptable to punish innocents, and if so, just how many innocents are acceptable.

If we say "Yes" to the first, we have to be specific about just how many.

We can't just leave it hanging on a principle: This is real life, with real life implications.

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?

[Ivor the Engineer]

<snip>

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?

That's the wrong question. The correct one is:

What is the highest ratio of innocent [to guilty] citizens that is acceptable to be unjustly punished in error?

If your answer to the above question is zero, forget about having any justice at all in the real world.

[CFLarsen]

How can people be unjustly punished not in error?

[Francesca R]

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?
I understand that you have tossed that question out there. It is a good one. I decline to answer that--at least not yet.

However I asked you a question about what you posted, which you answered. Specifically, you think that the "cost" of one innocent being punished is the rather extreme "no justice at all". I am asking you about why that is the case. Do you decline to answer any further?

Should I conclude that you believe "it is better that every criminal goes free than one innocent is punished"? Scrap the justice system?

[CFLarsen]

Declining questions makes a bad case for demanding answers to one's own questions.

[Francesca R]

Doing a utilitarian cost-benefit analysis misses the point, as it usually does. Justice cannot be calculated.If the costs and benefits of justice are incalculable, why would a society seek to achieve justice?

[Ivor the Engineer]

How can people be unjustly punished not in error?

The "in error" refers to intent. Many innocent people around the world have been and are punished on purpose.

[D'rok]

If the costs and benefits of justice are incalculable, why would a society seek to achieve justice?
Justice as a notion is incalculable. Blackstone's ratio is addressing the notion of liberal justice, not the benefits that a society may or may not enjoy from actualizing that particular notion of justice. It is a philosophical principle, not a law and economics formula.

[Francesca R]

Declining questions makes a bad case for demanding answers to one's own questions.That's as may be but yours was general (and not based on a position I have taken--at least not yet) and mine was asked of your partly-stated position. And it is not a demand--though you are free to point out whether I would be correct or incorrect in the stated conclusion I would draw if you deline to answer it. Or you can just leave that to me. It doesn't really matter, except that your position will perhaps remain muddy, and will be left to others to infer.

Declining questions makes an even poorer case for taking positions than it does for seeking answers from others based on what they've said. That's why politicians don't often get to say "Well what's your answer?" to journalists ;)

[Francesca R]

Justice as a notion is incalculable. Blackstone's ratio is addressing the notion of liberal justice, not the benefits that a society may or may not enjoy from actualizing that particular notion of justice. It is a philosophical principle, not a law and economics formula.I must admit I see Blackstone's ratio as certainly applying to the real-world implementation of the notion of justice. Otherwise he might as well have said "No innocent should be punished ever, and every guilty should be punished always". After all--that seems to summarise the notion rather well.

[CFLarsen]

If "one innocent person punished means no justice at all" is disagreed with, then a position has been taken.

If one innocent person punished can mean justice, then:

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?

[D'rok]

I must admit I see Blackstone's ratio as certainly applying to the real-world implementation of the notion of justice. Otherwise he might as well have said "No innocent should be punished ever, and every guilty should be punished always". After all--that seems to summarise the notion rather well.
He was using a clever turn of phrase. Otherwise known as rhetoric. The notion of liberal justice is that "no guilty should be punished by the state at the expense of the innocent". To do otherwise would be incompatible with liberalism.

[Francesca R]

If "one innocent person punished means no justice at all" is disagreed with, then a position has been taken.That's true. I disagree with it because if guilty people are also punished then there is some justice at all.

[Francesca R]

He was using a clever turn of phrase. Otherwise known as rhetoric. The notion of liberal justice is that "no guilty should be punished by the state at the expense of the innocent". To do otherwise would be incompatible with liberalism.What do you mean by punishing the guilty "at the expense of the innocent"?

[CFLarsen]

If a position has been taken on the issue of innocent persons being punished wrt justice being done, then:

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?

[Travis]

Seeing a loved one being prosecuted in a court for a heinous crime they didn't commit was so hard to bear for me that I would never wish on it anyone else. Thankfully Judge duTemple pointed out the absurdity of charging someone with conspiring to murder unspecified people based purely on a psychological profile or else my brother would still be serving his prison sentence. Maybe you are fine living someplace where numerous innocents are imprisoned in the name of not letting anyone who is guilty go free but I would not be.

[Francesca R]

If a position has been taken on the issue of innocent persons being punished wrt justice being done, then:

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?The position taken is not "Punishing the innocent does justice". It doesn't. The position is, rather, "Punishing one innocent does not render any possibility of any justice at all in the system to zero".

And you disagree with that, don't you?

[D'rok]

What do you mean by punishing the guilty "at the expense of the innocent"?Ah. The disingenuous "pretend not to understand your meaning even though I'm obviously an intelligent person" strategy.

Not interested.

Cheers.

[linusrichard]

I'm not sure about CFLarsen's "no justice" formulation, but he's right in principle. It is not acceptable for innocent people to be punished.

It is better for n guilty people to go free than for 1 innocent person to be punished, where n is an arbitrarily large number. This is not a cost-benefit analysis. This is not economics. This is fundamental to our system of criminal justice.

[Ivor the Engineer]

Those interested in the implementation of a feasible justice system in the real-world should look at receiver operating characteristics.

http://en.wikipedia.org/wiki/Receiver_operating_characteristic

[Francesca R]

Seeing a loved one being prosecuted in a court for a heinous crime they didn't commit was so hard to bear for me that I would never wish on it anyone else.Yes. Seeing a loved one wronged by heinous crime for which the perpetrator walks free is also hard to bear and few would wish it on anyone else.

The costs to individuals close to those wronged by the state (leaving aside the perpetrator for a moment) is immense. The costs to society due to the state's failures in each case are also "non zero".

Is one cost (the former--your example) immeasurably higher to society than the other though?

[Francesca R]

Ah. The disingenuous "pretend not to understand your meaning even though I'm obviously an intelligent person" strategy.No. With respect, what do you mean?

[Francesca R]

I'm not sure about CFLarsen's "no justice" formulation, but he's right in principle. It is not acceptable for innocent people to be punished.It is not. It is also not acceptable for guilty people to not be punished.

It is better for n guilty people to go free than for 1 innocent person to be punished, where n is an arbitrarily large number. This is not a cost-benefit analysis. This is not economics. This is fundamental to our system of criminal justice.Can you define what it is as well as what it is not? The OP seeks a "sceptical consideration of this", not a doctrinaire proclamation.

[Ivor the Engineer]

I'm not sure about CFLarsen's "no justice" formulation, but he's right in principle. It is not acceptable for innocent people to be punished.

It is better for n guilty people to go free than for 1 innocent person to be punished, where n is an arbitrarily large number. This is not a cost-benefit analysis. This is not economics. This is fundamental to our system of criminal justice.

Yes, it is. You accept it if you believe a system for justice is required in the real world.

[Nyarlathotep]

If a position has been taken on the issue of innocent persons being punished wrt justice being done, then:

What is the highest ratio of innocent citizens that is acceptable to be unjustly punished?


Because there is no magic number. No one with any understanding of the world is out there saying "Well, One in a thousand is okay, but if it went up to one in nine-hundred-ninety-nine, now that would just be going to far!"

Most people simply look at the system and decide if they think adequate safeguards are in place to keep the numbers of innocent people punished as low as possible. And before you even ask, there is no magic number there either.

[Ivor the Engineer]

A perfect justice system is impossible over the long run, but in theory it is possible for a practical justice system to approach (but never reach) as close to perfect as you wish.

http://en.wikipedia.org/wiki/Noisy-channel_coding_theorem

The Shannon theorem states that given a noisy channel with channel capacity C and information transmitted at a rate R, then if R < C there exist codes that allow the probability of error at the receiver to be made arbitrarily small. This means that theoretically, it is possible to transmit information nearly without error at any rate below a limiting rate, C.

[linusrichard]

It is not. It is also not acceptable for guilty people to not be punished.
Yes it is. I think it's perfectly acceptable for, for example, someone who is guilty of drug possession to be acquitted because the evidence was unlawfully seized and is thus inadmissible. Or, to take the general case, it is perfectly acceptable for a guilty person to be acquitted because the prosecution failed to prove its case to the jury. Unfortunate, maybe, but absolutely acceptable. It is also absolutely acceptable for a judge to give probation or a suspended sentence, or for a prosecutor to offer a dismissal, whether in exchange for something or not. All of these are totally acceptable ways for guilty people to go unpunished.
Yes, it is. You accept it if you believe a system for justice is required in the real world.
I think the only disagreement here may be what the word "acceptable" means.

[CFLarsen]

It shouldn't make any difference between seeing a loved one wronged by a heinous crime, than seeing a total stranger being submitted to the same.

But it does make the point clearer. Because this is about what can happen to you and me.

"Adequate" safeguards, without actually quantifying what those safeguards are? That's just a cowardly approach: Claim the moral high ground, just be sure not get your own balls caught in the cotton gin.

Until they do get caught, when you are punished unjustly.

The question stands: How can there be justice, if innocent people are punished?

[Francesca R]

Yes it is. I think it's perfectly acceptable for, for example, someone who is guilty of drug possession to be acquitted because the evidence was unlawfully seized and is thus inadmissible. Or, to take the general case, it is perfectly acceptable for a guilty person to be acquitted because the prosecution failed to prove its case to the jury. Unfortunate, maybe, but absolutely acceptable.In your examples the state failed. Why is that perfectly acceptable? Why does the state get a free pass to unlawfully seize evidence, or fail to prove a case that's true?

[Francesca R]

The question stands: How can there be justice, if innocent people are punished?The question stands: If one punished innocent means that there can be no justice, does the holder of that position think that dismantling the justice system is the best feasible action? (In another word--anarchy)

[Nyarlathotep]

The question stands: How can there be justice, if innocent people are punished?


Because imperfect justice <> no justice. The fact that some few may unfortunately be imprisoned unjustly doesn't make the punishment meted out to the truly guilty any less just. And if you accept only "No innocent person will ever be punished unjustly" as your standard, you may as will give up on the whole idea of justice altogether and live in anarchy, because that is the only way you will ever acheive that standard, until someone invents an infallible human. Good luck with that. Until then you live with the fact that justice is not and never wil be perfect and do the best you can with what you have.

[Francesca R]

A perfect justice system is impossible over the long run, but in theory it is possible for a practical justice system to approach (but never reach) as close to perfect as you wish. [ . . . ]I applaud the effort to present the issue in a technical framework by the way. It is apt to be eschewed about as much as presenting the same in economic terms, but also valid.

[RecoveringYuppy]

A perfect justice system is impossible over the long run, but in theory it is possible for a practical justice system to approach (but never reach) as close to perfect as you wish.

http://en.wikipedia.org/wiki/Noisy-channel_coding_theorem

Important to point out that the "noisy channel" at issue here would be the information reaching the jury. The cost of lessening the noise of that channel would be a loss of privacy. Giving a jury perfect information would require monitoring and recording and revealing every bit of the accused's life to them. And, your definition of "perfect justice" includes catching everyone guilty of something, then everyone's life has to be reviewed, not just the accused.

[Francesca R]

"Adequate" safeguards, without actually quantifying what those safeguards are? That's just a cowardly approach: Claim the moral high ground, just be sure not get your own balls caught in the cotton gin.

Until they do get caught, when you are punished unjustly."You would not like it if it happened to you" is a cowardly approach. It elevates individual interest in a public good (justice) above collective (public) interest in the same. It is well established mathematically (see "public goods theory", or "logic of collective action" or "tragedy of the commons" or "CC-PP game") that individual self-interest acting alone will result in the failure to deliver the public good that is justice.

Do you want to dismantle the justice system simply because you don't want it to misfire and come after you?

[RecoveringYuppy]

It isn't a question of 10, 100 or any number of criminals walking.

The question is: Can a society live with just one innocent person being punished for a crime he didn't commit?

If you say "yes", then state:
1. The highest ratio of innocent citizens that is acceptable to be unjustly punished.

2. The reasoning behind why this should be so.

3. How you will implement it.

4. Why it is "justice".

5. What you would feel like if you were the innocent being unjustly punished.

In order:

Yes, inevitably there will be mistakes and innocents will be punished.

1. Like all other issues of value, this is chosen by negotiation. Everyone has a different subjective opinion about how much safety they want versus how much liberty and privacy they would be willing to give up to obtain that level of safety.

2. It's a subjective question of value that feeds in to a cost benefit analysis.

3 and 4 are a bit too broad. And my opinion is that they are tougher questions to answer if you think the justice system needs to avoid punishing even a single innocent person, so I'll let you go first on these.

5. Horrible. I would also feel horrible if someone guilty went free or if I had to live my life under a microscope to make sure that all the possible evidence of my innocence would be available if I were ever to be unjustly charged.

[Ivor the Engineer]

I applaud the effort to present the issue in a technical framework by the way.

Thank you.

It is apt to be eschewed about as much as presenting the same in economic terms, but also valid.

I don't mind; my posts in this thread are more for my own enjoyment and understanding. If others find them interesting then that's a bonus.:)

[linusrichard]

In your examples the state failed.
In the two examples you chose to quote, yes.
Why is that perfectly acceptable? Why does the state get a free pass to unlawfully seize evidence, or fail to prove a case that's true?
I don't understand the question the way you've phrased it.

Why is it not acceptable? We can say it's unacceptable for the prosecution to do a bad job, but how can we say it's unacceptable for them, having tried their best, to lose? Some things just can't be proven. Some juries just can't be convinced.

Maybe you're concerned with the "right" of the crime victim to have the criminal prosecuted. A criminal prosecution is not, generally, the way to vindicate a private right. If Smith assaults Jones, the state can prosecute Smith -- the case is State (or People or Commonwealth or Regina) v. Smith. Jones is not a party. If Jones wants her rights vindicated in court, the place for that is Jones v. Smith. Even if the state loses its case, Jones can win hers. The most famous example being the successful wrongful death suit against O.J. Simpson, after he was acquitted of murder.

[Francesca R]

I don't understand the question the way you've phrased it.It is not acceptable for the state to submit inadmissable evidence in a trial. It might be the correct least bad consequence to declare a mistrial, but it is not acceptable in the sense of "No improvement needed here". Similarly for the state failing to procure justice for any crime. And similarly for the discovery of a wrong conviction. None of these are acceptable, except that we can't turn back the clock and do it again correctly this time. There are costs.

The cost to society (not to individuals here) of an unpunished guilty person is the increased risk of others being vicitimised by future crimes. The cost of a wrong conviction is the increased risk to society that others will be falsely accused and convicted.

[linusrichard]

It is not acceptable for the state to submit inadmissable evidence in a trial. It might be the correct least bad consequence to declare a mistrial, but it is not acceptable in the sense of "No improvement needed here".
Okay, yes, I agree with what you're saying, but it's not what I was saying. You misunderstood my example. I was not saying it's perfectly acceptable for the state to submit inadmissible evidence. I was saying it's perfectly acceptable for the state to lose because it was unable to submit unlawfully seized would-be evidence.
Similarly for the state failing to procure justice for any crime.
And here we disagree. (EDIT: Maybe it depends on what justice means. Is it just to convict someone of an utterly unprovable crime that they did in fact commit? I'll leave that to better jurisprudes.)
And similarly for the discovery of a wrong conviction. None of these are acceptable, except that we can't turn back the clock and do it again correctly this time. There are costs.

The cost to society (not to individuals here) of an unpunished guilty person is the increased risk of others being vicitimised by future crimes.
Yes.
The cost of a wrong conviction is the increased risk to society that others will be falsely accused and convicted.
That's one cost.

But this isn't about costs and benefits. If you're thinking in those terms, then maybe you're right. Maybe you can come up with a ratio, and figure out your costs and benefits according to the ratio, and decide how many free guilty persons cost as much as one punished innocent, and work it out. But what I'm saying is that it is unacceptable as a matter of principle to punish innocent people, and that the cost is irrelevant; it is always too high.

[quarky]

The ratio that exists must be the acceptable ratio, because it is accepted.

(I have no idea what it is, but someone might)

[Francesca R]

Okay, yes, I agree with what you're saying, but it's not what I was saying. You misunderstood my example. I was not saying it's perfectly acceptable for the state to submit inadmissible evidence. I was saying it's perfectly acceptable for the state to lose because it was unable to submit unlawfully seized would-be evidence.OK

But this isn't about costs and benefits. If you're thinking in those terms, then maybe you're right.That is a common remark. . .[ . . . ] it is unacceptable as a matter of principle to punish innocent people, and that the cost is irrelevant; it is always too high. . . . but this does not get us anywhere. Society can aspire to such principles but it cannot not consistently embed them all in its workings. If it did, the logic would be (i) the cost of an innocent being punished is always too high, so (ii) it must be avoided at all (other) costs, so (iii) nobody must ever be punished. This leads to suspension of the justice system.

The problem is that principles conflict with each other. The principle that the guilty must be punished is in conflict with the one that no innocent may be punished. This brings us to a cost-benefit trade-off unless you think that one principle must fully give way to the other, in which you have to abandon justice. The cost of unpunished guilt and wrongfully punished innocence, and the probabilities of these things happening is the "uncomfotable question" that is frequently avoided.

[CFLarsen]

Again, it isn't a question of people being unknowingly punished.

All societies are made up of individuals. Those societies we call civilized, democratic and free are comprised of people acting together for a common good.

For that to happen, there can't be groups of people being treated differently than other groups. E.g., one group cannot be punished unjustly while another group is.

To argue otherwise is to accepting things like apartheid. Segregation. State-sanctioned racism.

So, what is the highest ratio of innocent citizens that is acceptable to be unjustly punished?

Yes, we keep returning to this. Inevitably.

[RecoveringYuppy]

@prior post

If there is one group where "mistakes" happen at a higher or lower rate than others then you have a problem (and that problem is known to exist in most judicial systems).

But that is different than having an error rate that applies to all people.

[Francesca R]

If "one innocent punished" means "no justice at all" does that mean scrap the justice system?

The answer seems to be "yes", but CFLarsen seems loth to confirm this. Yet saying "no" would be to retract the statement so CFLarsen appears loth to answer either way. The conclusion I draw is that the statement: "one innocent punished means no justice at all" is in a spot of trouble.

[Madalch]

Chastened, he returned to his hole in a big tree, nailed the door shut, and hasn't been hear from since, though his electricity and Internet bills somehow get paid regularly.

So that's who the hermit really is.....

[jimbob]

The recent news about the conviction for the killing of Rachel Nickell (http://news.bbc.co.uk/1/hi/england/london/7783796.stm) highlights another point, from a purely pragmatic point of view:

"Better 10 guilty people go free than one innocent is hanged, whilst the guilty person gets off scott free because the case is prematurely closed."

From what I remember of the case there was a lot of unhappiness when Colin Stag was released "because he obviously was guilty".

From an observer article (http://www.guardian.co.uk/commentisfree/2006/jun/25/comment.homeaffairs)



The senior officer looked at us as if we were fools. He could assure us that Stagg was as guilty as Crippen. He was so self-confident that we might have believed him if we had not read up on the case. Other editors were less sceptical and the anonymous mutterings convinced them that a liberal judge had let a terrible killer walk free on a technicality. I talked to some of them recently and they seemed to be genuinely shocked that DNA evidence now shows that Rachel Nickell's killer may be a man held in Broadmoor. Shocked and a little ashamed, because what all those off-the-record briefings produced was a police-approved Stagg hunt led by a pack of C-list celebs and thoughtless hacks.

The worst of it was that the police and media persuaded the family of Rachel Nickell that the crucial difference between Stagg and Hindley was that Stagg had got away with murder. The News of the World ran lipsmacking pieces on how the 'weirdo' demanded 'bizarre sex' with his 'terrified' girlfriend yards from where Rachel Nickell was murdered. The Daily Mail quoted Andre Hanscombe, father of her son, saying he was '99 per cent certain' that Stagg was guilty and the government should remove the double jeopardy law so he could be tried again. It also ran a serialisation of a self-justificatory book by the officer in charge of the case, Detective Inspector Keith Pedder, headlined 'How British Justice Betrayed Rachel's Son'.

Anyway this is aside from the main point, which is the difference between sins of commission and sins of omission. The state failing to "hang" guilty person is *omission*, the state "hanging" an innocent person is *comission*. In the first case a wrong has not been "avenged", in the second, a wrong has been committed by the state.

[Foolmewunz]

I don't know that I have any difficulty with the ratio or princple or whatever you want to call it.

Blackstone is referring to the underlying philosophy by which we should guide our justice system. And in that respect I agree with LinusRichard. The target has to be to protect the innocent. That may be an unreachable target, but in a just society it must be the goal. And any revisions to the law to make it more convenient to law enforcement by arresting and convicting more innocent people in order to "get to" even more guilty ones automatically assumes, by anyone arguing that point, "well, except me, of course".

The minute you can do as Travis mentioned and personalize this,(and I personally believe you have to in order to have this sort of moral discussion) and realize the impact on a human life that is caused by erroneous prosecution, conviction and incarceration, then all theoretical pragmatic discussion needs to be chucked aside.

If I'm the great designer of a society, my target is zero erroneous prosecutions and incarcerations of innocent persons. Any tolerance of anything less is an invitation to extrapolate that relaxation into a police state.

We're not and Blackstone's not discussing law enforcement, safety rails in showers, building levies to hold back flooding rivers, etc.... He's talking about jurisprudence and the dispensation of "justice".

The assumption that there has to be some magical number at which the punishment of an innocent is acceptable weighed against the benefits of X number of bad guys getting their just rewards, is not tenable to me.

And we're talking about guilt or innocence of laws that we're assuming we all agree with, I assume. How about laws we don't agree with? Bismark's corollary (that it's better that ten innocents should suffer in order that one guilty be apprehended and punished) had some noted proponents in Joe Stalin, Mao Zedong, Pol Pot, and others*. They were arresting, prosecuting, convicting, and punishing (sometimes skipping over the prosecuting and convicting part) criminals, after all... according to their laws.

*Avoiding Godwinning by only the narrowest of margins, I realize.

[jimbob]

Agreed entirely Foolmewunz,

Further to that, I'd argue that most justice systems are not based on "reparations", but on "punishment": i.e. of the justice system comitting an appropriate "wrong" to the guilty person, and not on the guilty person "correcting" the wrong to the victim. You can't un-murder someone.

If you view Blackstone's ratio in this light, then it is saying that justice is better served by the justice system erring by failing to commit wrongs against guilty people rather than committing wrongs against the innocent.

If you can't undo wrongs, and the best you can do is punishment, then it is less bad to fail to punish (thus not committing a wrong against someone who deserved it), rather than committing another wrong against someone who doesn't "deserve" it.

[Euromutt]

While I agree entirely with the sentiment of what you're saying, Foolmewunz, that does keep leading back the question Francesca's been asking all along, to wit:

If it is unacceptable for even a single innocent to be wrongfully convicted, does this not imply we should dismantle the criminal justice system entirely, given that this is the only way to guarantee with absolute certainty that nobody will ever be wrongfully convicted? In other words, is it better that a potentially infinite number of guilty men go free, than that one innocent be punished?

[jimbob]

Euromutt,

I think it is recognised that perfection is unobtainable, so we (collectively) set the burden of proof at: "beyond reasonable doubt" but not "balance of probabilities".

[Francesca R]

Anyway this is aside from the main point, which is the difference between sins of commission and sins of omission. The state failing to "hang" guilty person is *omission*, the state "hanging" an innocent person is *comission*. In the first case a wrong has not been "avenged", in the second, a wrong has been committed by the state.So what though? The state is a human made thing--it is supposed to do some things and it is supposed to refrain from doing others. The effects on society of failure on either count can be big or small. Just by saying they are different in the same way as "not enough" is different from "too much" doesn't suggest that one is worse than the other, never mind by how much.

[Francesca R]

The minute you can do as Travis mentioned and personalize this,(and I personally believe you have to in order to have this sort of moral discussion) and realize the impact on a human life that is caused by erroneous prosecution, conviction and incarceration, then all theoretical pragmatic discussion needs to be chucked aside.This sounds like you are saying "This can not and should not be thought about objectively". ,Now that can't be true, can it?

Any tolerance of anything less is an invitation to extrapolate that relaxation into a police state.This sounds like you have no alternative but to abolish state prosecution altogether, as pointed out to others before. Unless you do not, in fact, mean "tolerate".

The assumption that there has to be some magical number at which the punishment of an innocent is acceptable weighed against the benefits of X number of bad guys getting their just rewards, is not tenable to me.So does this. So I don't think you are actually saying what you believe.

Society may aspire to no wrongful convictions, but it can and does tolerate some. And that is even after considering "What if it was me".

[Foolmewunz]

This sounds like you are saying "This can not and should not be thought about objectively". ,Now that can't be true, can it?

This sounds like you have no alternative but to abolish state prosecution altogether, as pointed out to others before. Unless you do not, in fact, mean "tolerate".

So does this. So I don't think you are actually saying what you believe.

Society may aspire to no wrongful convictions, but it can and does tolerate some. And that is even after considering "What if it was me".

This is why philosophical discussions ultimately go no where.
Everything that I wrote was towards the underlying philosophical approach to establishing a system of justice. That was how Blackstone was looking at it IMHO, and that's how I'm responding.

Everything close to valid in the objections to my post have been based on creating finite mathematical formulas to apply the underlying philosophy. Sorry, but it doesn't work that way.

You can do this with anything in the realm of "principles" when those are discussions of morality.

I say, "One man, one vote". Has to be. Any fair society will by its very definition of "fair" have to strive for this.

You can respond, "Oh, so if we can see that we haven't achieved that end, we should dismantle the U.S. Constitution or UK Parliamentary system?" And we can go 'round and 'round for hours playing this game.



Not everything can be put onto a Gantt chart.


"I love you."
"How much?"
"3.72 Kilos."

ETA: I think you're just arguing a point. You realize that saying that you have to understand the personal/human element in any discussion of fairness in justice is NOT saying that you can't bring some objectivity into the discussion. It's saying that you have to also balance that with subjectivity.

[Foolmewunz]

While I agree entirely with the sentiment of what you're saying, Foolmewunz, that does keep leading back the question Francesca's been asking all along, to wit:

If it is unacceptable for even a single innocent to be wrongfully convicted, does this not imply we should dismantle the criminal justice system entirely, given that this is the only way to guarantee with absolute certainty that nobody will ever be wrongfully convicted? In other words, is it better that a potentially infinite number of guilty men go free, than that one innocent be punished?

Yes, because the other side of that coin is that by not taking this stand it would mean that there would also be a potentially infinite number of innocents who are punished, n'est ce pas?

I understand the pragmatic side of the discussion. I'd love to have you people in the DA's office. I would not like you writing my Constitution or making my laws, however.

ETA: That's the editorial, you.... not you, personally.

[Euromutt]

I don't think you can simply dismiss this by saying "that's just not quantifiable." You can point out why it's not quantifiable.

Perhaps the problem with Francesca's question is that there's an implicit assumption that we know how many innocent people the criminal justice system actually wrongly convicts, and that we can readily identify instances of this occurring. We conjure up all the ratios we like, but at the end of the day it's all academic because we have no way of knowing when any such ratio is being exceeded. If we did, this conversation would be redundant, because we'd know how many miscarriages of justice had occurred, at which point we could set about identifying and rectifying them.

[Richard Masters]

If you're talking purely a monetary cost/benefit ratio, then really the most cost effective method would have extreme and violent punishments for people who just *seemed* likely to have committed the crime...and their families. Administrative costs would be minimal and the deterrent would be very strong (public floggings of an alleged thief and the immediate family).

I bring up this extreme example because your most likely gut reaction is that this is wrong. Thus there is some inherent moral value to not punishing the innocent.

In my mind Blackstone's statement is simply a reminder that as a society our goal is to protect the innocent. This means not only protecting the innocent victims of crimes but also the innocent people who had nothing to do with it. We consider kidnapping to be a heinous crime. As a practical matter locking up an innocent person is no different than kidnapping to the person being locked up and to those who would want to have contact with that person.

Thus the judicial system should take steps to ensure that convicting the innocent is an extremely unlikely event just like the first rule of medicine is to do no harm.

Another practical aspect of this is that the more likely it is for an innocent person to be convicted, the more likely it is that the government will abuse its power. Making accidental wrongful convictions more difficult has the added benefit to making intentional wrongful convictions less likely to be attempted and less likely to be successful.

To me the numbers don't have a specific meaning. He could have just said that it is far worse for society to convict the innocent than it is to let the guilty go free.

As a practical matter the numbers are irrelevant. If you *know* that 1 in 20 convictions is a wrongful conviction, then you can overturn those convictions and eliminate the problem.

His statement is just a guideline for forming laws.

Cost does not only mean money.

[Richard Masters]

I don't think you can simply dismiss this by saying "that's just not quantifiable." You can point out why it's not quantifiable.

Perhaps the problem with Francesca's question is that there's an implicit assumption that we know how many innocent people the criminal justice system actually wrongly convicts, and that we can readily identify instances of this occurring. We conjure up all the ratios we like, but at the end of the day it's all academic because we have no way of knowing when any such ratio is being exceeded. If we did, this conversation would be redundant, because we'd know how many miscarriages of justice had occurred, at which point we could set about identifying and rectifying them.

It is quantifiable... How many wrongful sentences have been overturned from new evidence?

[Euromutt]

It is quantifiable... How many wrongful sentences have been overturned from new evidence?That's a fair point, but we can only know that these sentences were wrongful after the fact, on the basis of new evidence that was (presumably) not available to the court at the time the conviction was imposed. There could be any number of people wrongfully incarcerated at this very moment, but we have no way of knowing at present that their convictions are wrongful because the evidence that may lead to the overturning of that conviction has not yet come to light (if it ever does).

So we know how many mistakes the criminal justice system has made in the past, but we have no way of knowing whether past performance reflects on the present, especially since the new evidence leading to those exonerations are frequently the product of new techniques in evidence gathering, which will (hopefully) also result in fewer people being wrongfully convicted at present and in the future.

[jimbob]

Originally Posted by jimbob
Anyway this is aside from the main point, which is the difference between sins of commission and sins of omission. The state failing to "hang" guilty person is *omission*, the state "hanging" an innocent person is *comission*. In the first case a wrong has not been "avenged", in the second, a wrong has been committed by the state.
So what though? The state is a human made thing--it is supposed to do some things and it is supposed to refrain from doing others. The effects on society of failure on either count can be big or small. Just by saying they are different in the same way as "not enough" is different from "too much" doesn't suggest that one is worse than the other, never mind by how much.


But there *is* a difference.

Whilst you can argue (and I'd agree) that there is a harm to society as a whole from the guilty going free, it is not possible to say what individuals will be harmed. Letting a guiltly person go free is not condemming any particular individual to being wronged. Hanging/imprisioning an innocent person *is*.

I would class harming an innocent individual for the greater good, as wrong. Society doesn't choose to "punish" the individual, it is an idividual (in the form of a judge, acting as the representitive of society) who actually chooses to "punish" the individual. If you accept that all individuals have equal worth, then what right has one individual to choose to harm an innocent person "for the greater good"?

[Francesca R]

But there *is* a difference.I know that, but I don't think it amounts to much.

Whilst you can argue (and I'd agree) that there is a harm to society as a whole from the guilty going free, it is not possible to say what individuals will be harmed.Correct, but so what? Ex-ante there is harm, that society would rather avoid, originating from failure to correct, deter, and protect society from criminal behaviour (BTW I disagree with your observation that the purpose of punishment is to "commit a wrong")

Letting a guiltly person go free is not condemming any particular individual to being wronged. Hanging/imprisioning an innocent person *is*.I have to say this is semantic. One could cue up moral dilemmas such as thowing a lever that will divert a runaway train from killing one person you can see, onto another track where, 2km away, it will kill ten people you cannot see instead.

I would class harming an innocent individual for the greater good, as wrong. Society doesn't choose to "punish" the individual, it is an idividual (in the form of a judge, acting as the representitive of society) who actually chooses to "punish" the individual. If you accept that all individuals have equal worth, then what right has one individual to choose to harm an innocent person "for the greater good"?That drives a coach and horses through justice though. You've joined those who say that any conviction of innocents is to be ruled out. Yet unless you are agitating for dismantling courts and dismissing police, you can't actually believe it. So why write it?

This is what I mean in the OP by "a sceptical consideration". I think we have a tendency to say things that we do not really believe.

[Francesca R]

This is why philosophical discussions ultimately go no where.This isn't, or need not be, a philosophical discussion. It's not in that forum and the OP does not ask for philosophy.

Everything close to valid in the objections to my post have been based on creating finite mathematical formulas to apply the underlying philosophy. Sorry, but it doesn't work that way.Sorry, but it works that way. Currently a certain ratio exists of walking guilty to punished innocents. I don't know if it is 10:1. Certainly it varies according to crime, and I assert that, for example, for murder it is low and for rape it is high. Different ratios are de-facto accepted by society depending on what the crime is. A question is what is acceptable. And that depends on costs and benefits. On another thread the consensus acknowledged, for example, that a greater ratio of walking guilty to convicted innocents was probably "inevitable" (read--acceptable) for sex crimes because of the greater difficulty of proving the crime. Some people (not everybody) thought that false conviction was a greater danger for that crime than for others. On that basis there would be a Blackstone-esque argument for not prosecuting such crimes very often.

More generally, wrongful conviction and wrong aquittal are real issues that in real life inform the drafting of criminal statutes, rules of evidence, court procedures and judicial rulings. So it is rather necessary that this does not remain in the realm of principle. (And it doesn't of course--or else every crime would be tried identically)

ETA: I think you're just arguing a point.Eh, yes . . . and you're not arguing one. :)

You realize that saying that you have to understand the personal/human element in any discussion of fairness in justice is NOT saying that you can't bring some objectivity into the discussion. It's saying that you have to also balance that with subjectivity.Fine, though it seemed to me that you used the subjectivity part as a way to say "this can't be discussed". And that still looks like your position.

[jimbob]

But there *is* a difference.I know that, but I don't think it amounts to much.

Whilst you can argue (and I'd agree) that there is a harm to society as a whole from the guilty going free, it is not possible to say what individuals will be harmed.Correct, but so what? Ex-ante there is harm, that society would rather avoid, originating from failure to correct, deter, and protect society from criminal behaviour (BTW I disagree with your observation that the purpose of punishment is to "commit a wrong")

Maybe my choice of words was poor:

Retribution not restitution.


Letting a guiltly person go free is not condemming any particular individual to being wronged. Hanging/imprisioning an innocent person *is*.I have to say this is semantic. One could cue up moral dilemmas such as thowing a lever that will divert a runaway train from killing one person you can see, onto another track where, 2km away, it will kill ten people you cannot see instead.

That is a slightly different situation, admittedly one which I had been thinking about.

It would be akin to throwing a lever that *will* kill one person, but which *might* save an unspecified number other people (as yet unknown). It might (by letting the *real* guilty person get away) actually cause the deaths of a larger (unspecified) number of people.

In such a situation you'd have to be pretty bloody certain that the deaths would be otherwise inevitable.


I would class harming an innocent individual for the greater good, as wrong. Society doesn't choose to "punish" the individual, it is an idividual (in the form of a judge, acting as the representitive of society) who actually chooses to "punish" the individual. If you accept that all individuals have equal worth, then what right has one individual to choose to harm an innocent person "for the greater good"?That drives a coach and horses through justice though. You've joined those who say that any conviction of innocents is to be ruled out. Yet unless you are agitating for dismantling courts and dismissing police, you can't actually believe it. So why write it?


I said that was an ideal to aim for; in reality* I'd settle for "beyond reasonable doubt" but not "balance of probabilities".



*Because there is harm to society by letting guilty go free, but Id say that it is a lesser evil than wrongly punishing theinnocent