Q&A: What to Be Concerned About
What to Be Concerned About
Question
Hello Rabbi!
I wanted to know your opinion on two matters.
1. In severe crimes such as rape and murder, what is more concerning: a wrongful conviction, or refraining from conviction and thereby letting wrongdoing go unpunished?
2. Is there room for a society of learners, and if so how much and in what way?
Thank you very much!
Answer
1. It is generally accepted (and this is also the view of Maimonides) that it is preferable to let offenders go free than to convict an innocent person. That is why criminal conviction requires proof beyond a reasonable doubt. The reasoning for this is not simple, and it may depend on the situation in question. For example, if the punishment serves only to deter others, then punishment itself is less important, since deterrence is achieved through other offenders, and here this person is not punished because it is not certain that he was an offender. But if the danger comes from the person himself, then there is more reason to be concerned about letting him go free. Sometimes administrative detention is used in such a situation. Moreover, I wrote here in the past in a column (47) that in my opinion the whole conception of punishment in the world is distorted. Punishment is set in proportion to the severity of the offense, and therefore dangerous people are released after serving their sentence. In my opinion, they should remain in prison as long as they are dangerous, regardless of the severity of their offense.
2. Definitely not. There is room for learners, but not for a society of learners (that is, an entire society in which everyone studies). I also wrote about this in the past in column (34).
Discussion on Answer
You can’t determine which is more severe unless you first decide what the null hypothesis is. In doing that, you’ve already smuggled in your assumptions. The assumption that a person is innocent until proven guilty says exactly that, so there’s nothing to derive from it.
That’s not an assumption but a presumption, because it is known that most people do not commit serious crimes like rape or murder. Something like a statistical majority that is not directly before us.
Aside from that, the logic of reasonable doubt is primum non nocere. If you haven’t managed to show that you are bringing benefit, then refrain from acting.
That’s a common mistake. There is no evidentiary presumption here based on a majority. Most people in general are not murderers, but among those who are brought to trial there is no such majority. After all, they don’t put on trial a person they grab randomly off the street. A prosecution comes after screening—for motive, opportunity, possibility and ability, testimony, police investigations, and so on. Among such people there is certainly no majority of innocents.
This is like the presumption of ownership, which is not based on a majority. True, usually what is in a person’s possession belongs to him, but among those involved in a legal dispute there is no majority in favor of the current possessor.
Therefore both of these presumptions are legal presumptions, not a statistical majority.
And regarding benefit as well, you definitely do bring benefit by putting the defendant in prison: certain deterrence, and possibly also just deserts.
Regarding the presumption, I understand. You are right.
Regarding the benefit, if he is not guilty then there is no deterrence, but rather a loss of public trust in the justice system, and also the addition of another person to the criminal world. Aside from that, the meaning of criminal law is a monopoly on violence and its regulation. Before acting violently toward a person (by imprisoning him), it must be proven that there is indeed a reason for it. Otherwise, the justice system will be no different from an underworld gang settling scores.
I disagree. The public does not know that he is innocent. On the contrary, we are talking about a case where there is good evidence against him, but some doubt still remains. So why not convict? You are probably punishing a guilty person, deterrence is created for potential offenders (and they cannot even count on acquittal because of some small doubt). Set against that is a small concern about convicting an innocent person. It is really not trivial not to convict.
And of course this is very far from gangs that use violence without inquiry and without justice. The two are nothing alike.
In statistics, a Type I error is usually considered more serious. That is, rejecting the null hypothesis and accepting the alternative hypothesis. I remember that in a course the lecturer gave this very example.
But there his assumption was that a person is innocent until proven guilty. It’s interesting whether the example is still relevant if one does not accept that assumption.