Morality and infinity

The simplest moral distinction may be the binary distinction between wrong and right.

But there are important moral distinctions that can be made between different wrong acts. Some acts are worse than others. One distinction that can be made is the punishment merited by the act - no longer a binary distinction.

For each wrong act, then, we might state the punishment that it merits. Having done this, the statement that it is a wrong act is, on the face of it, superfluous, since the wrongness of the act would seem to consist in its meriting a certain punishment. How can an act be wrong if it does not merit any punishment? So instead of classifying acts according to whether they are wrong, we might classify them according to what punishment they merit (and of course most acts merit no punishment at all).

But now we are relying on the idea that a certain punishment is "merited". But what does this mean? It seems to mean something like this: normally it would be wrong to put someone in a cell for a year, but because they committed a certain act (e.g. theft), it is not wrong to do that. So "merit" seems, itself, to be a binary statement about wrong and right - one which, as explained above, fails to make important non-binary distinctions. And since it is a statement about right and wrong, we can perform the same operation on the punishment that we performed on the original act, replacing the binary right/wrong distinction with a non-binary distinction among punishments merited by the act.

Rather than say:

Theft merits the punishment of one year in prison,

we say:

Putting a thief in prison for one year merits no punishment.

We can perform this operation recursively. For example:

Putting someone in prison for putting a thief in prison for one year merits one year in prison.

Putting someone in prison for one year for putting someone in prison for putting a thief in prison for one year merits no punishment.

And so on.

This recursion is, of course, more than the human mind can handle on a daily basis. Nevertheless, it's really there, lurking "behind" moral statements. Events can happen which act out the chain at a deep level. A corrupt official might be imprisoned for a list of crimes, one of which might be to have placed a judge in prison for having placed one of the official's cronies in prison. There is no question that the corrupt official deserves to be in prison and that his jailers do not, for that just act, themselves deserve punishment. But to recognize this is to recognize that:

It would be unjust to jail someone for jailing someone for jailing someone for jailing a thief.

In the specific hypothetical:

It would be unjust to jail the official's jailers for jailing the official for jailing the judge for jailing a thief.

This recursion (I'm not sure it's a recursion - I can't think of a better word) can be carried out arbitrarily far. But not to infinity, because both ends are needed - i.e., there must be an end that contains the punishment that is or is not merited, and there must be an end that contains the initial act which hypothetically gives rise to the contemplated chain of punishments. And since this cannot be carried out to infinity, ultimately we are left with a binary judgment that such-and-such act is wrong or right.

Share this

I completely agree. As a

I completely agree. As a libertarian, I often say that I'm roughly confident in the first level of wrong, the initial agression in an agression free world, but not much beyond this. I do not know a convincing theory of libertarian punishment, or punishment of unfair punishment, etc.

One of my concerns

I don't know how well it ties in, but one of my concerns motivating this entry is the distinction between (natural) law and morality generally - I was thinking that the distinction might be based on the punishment merited (robbery merits imprisonment or execution; betraying a friendly trust merits ostracism or disgrace before mutual friends).

Those who distinguish between the legal and the moral see no problem. Morality is this, legality is that.

But I do not distinguish between (natural) law and morality - I consider it immoral to rob someone, and I consider it moral to respond to an attempted robbery with force. And yet, there are so many other things that are called "immoral" - such as having sex before marriage. Are these two acts (robbery and sex before marriage) located on the same continuum, or is the word "morality" here being used in very different ways?

In some cases claims of immorality are nothing more than the product of busybodies who have nothing better to do than to make sure that other people are as miserable as they themselves are. But, clearly, in many cases there is merit to claims of morality, and we are well served to heed them. Gluttony and lust are not merely against the teachings of some religion. Lack of restraint in eating and sex can land us in serious trouble that has nothing to do with other people and everything to do with an early grave.

So I think that morality that goes beyond the ethical principles of libertarianism isn't junk, that it should be taken seriously and probably merits being considered morality. And yet I also think there is a distinction between the immorality of promiscuity and the immorality of robbery.

One possible way to distinguish them is in terms of the merited punishment. Gluttony and lust are self-punishing and also punished by our friends drawing away from us (ostracism). (For example, if you are a married man and your male friend is known to feel no restraint in whom he sleeps with, then it is reasonable for you to keep him away from your wife, which probably requires keeping him out of your life entirely. Double if it's your daughter you're worried about.) Whether it merits these punishments I am not sure.

And yet I also think there

And yet I also think there is a distinction between the immorality of promiscuity and the immorality of robbery.

Perhaps the distinction you are looking for is Spooner's distinction between crimes and vices?

But I don't see how punishment tracks the discussion at all. There are all sorts of theories and justifications for punishment: retribution, restorative compensation, rehabilitation, incapacitation, deterrence, etc. These justifications are not always consistent with each other and often lead to contradictory results.

How can an act be wrong if it does not merit any punishment?

As you mentioned, vices like gluttony are wrong but do not merit punishment. But even interpersonal crimes like theft do not always merit punishment - sometimes the best solution is just to forgive and forget, perhaps because the costs of enforcement are too high.

The best reason I have found for why something like gluttony is wrong is also the best reason I have found for understanding why something like theft is wrong: the hypothetical imperative (which, incidentally, helps bridge the is-ought gap). Given the nature of the world in which we live, if we want to live the good life and flourish, then we must act in a certain way, abiding by certain rules. Gluttony is wrong because it is inconsistent with the good life. Theft is wrong for the same reason.

We might think of exceptional cases where it is not at all clear how the act of theft will eventually come back and bite the thief in the ass, and I'm willing to grant that these sorts of cases do occur frequently. But in order to take advantage of these sorts of situations, the rational thief needs to be a very good actor, and most of us are not very liars.

Not vices

Perhaps the distinction you are looking for is Spooner's distinction between crimes and vices?

No, because it's possible to offend someone aside from yourself without committing a crime. If you break a promise to someone you haven't necessarily thereby committed a crime, but you have offended someone aside from yourself. Anybody who proves themselves to be a bad credit risk is not thereby a criminal, but they do deservedly have their reputation blackened (their credit rating lowered). It's wrong to break promises, and to make promises that you can't keep, but not criminal.

So no, I do not mean the distinction between interpersonal offenses and offenses against oneself (sometimes referred to as victimless crimes).

But I don't see how punishment tracks the discussion at all.

Well, on a first pass it tracks the distinction between crime and non-crime extremely well. Criminals land in jail or dangle from the end of a rope. If something is not considered a crime you don't go to jail for it and don't typically get strung up for it by vigilantes (though, of course, it is not impossible for the vigilantes themselves to be in the wrong - just as it is not impossible for the city police to be in the wrong). If a certain sort of act would not get you sent to jail or strung up, it's evidently not considered a crime (note that I say considered a crime - dealing drugs is evidently considered a crime by those who advocate such treatment even if libertarians do not consider it a crime). Crimes are offenses that merit a certain severity of response.

There are other offenses that do not merit such a severe response even if they do merit some sort of softer response, such as ostracism from your circle of friends. These are not considered crimes, but they are often considered wrong. They break some rule of behavior. Crimes are always wrong but wrong things are not always crimes.

There are all sorts of theories and justifications for punishment: retribution, restorative compensation, rehabilitation, incapacitation, deterrence, etc. These justifications are not always consistent with each other and often lead to contradictory results.

That is not important if the theories are wrong. Suppose that I proposed the inverse square law for gravity and you responded that there are many conflicting theories of gravity. How would such a response in any way count against my proposal?

If, on the other hand, certain of these theories are right, then they are pertinent to my point, but then at the same time they could not contradict each other.

But even interpersonal crimes like theft do not always merit punishment - sometimes the best solution is just to forgive and forget, perhaps because the costs of enforcement are too high.

I think I translated "merit" as "it would not be wrong to do that to him", which means "it would not violate the person's rights to do that to him." What is the best solution is not the same thing as what I am within my rights to do. For example, when I forgive someone I am within my rights not to forgive him but I have judged it the best solution to forgive him.

The distinction between rights and best solution is important to capitalism. A market participant is within his rights to trade with any willing partner. Within this set of possibilities, all of which lie within his rights, he perceives and selects a best solution.

Forgiveness matters, is valuable to the recipient of forgiveness, only when the act merits punishment of some sort. In social settings, the punishment is some degree of social exclusion. Forgiveness matters to the person forgiven because it takes back the threat of social exclusion. The punishment doesn't have to be something formal. The mere presence of anger spoils a relationship, effecting a degree of ostracism. (Anger is itself already punishment, because it spoils social interaction.)

The best reason I have found for why something like gluttony is wrong is also the best reason I have found for understanding why something like theft is wrong

Actually, the reason you give is for why both these things are ill-advised (unwise). But being ill-advised is not the same thing as being immoral. Immoral things are usually ill-advised, but not everything that is ill-advised is immoral.

We might think of exceptional cases where it is not at all clear how the act of theft will eventually come back and bite the thief in the ass, and I'm willing to grant that these sorts of cases do occur frequently. But in order to take advantage of these sorts of situations, the rational thief needs to be a very good actor, and most of us are not very liars.

The problem arises because you focus on what is advisable for the would-be criminal to do. This is not the only thing that you can focus on. You can also focus on what is advisable for the potential or actual victim to do, what is advisable for the neighbors to do, and so on - which, as a matter of fact, is what I've been focusing on (i.e., not just the act itself, but the response to the act - e.g. the punishment). For example, it is advisable to guard against thieves even if the thieves are perfect liars. It might be advisable to buy a gun and to install theft deterrent devices.

Facts about the advisability of guarding against Act X can be used as evidence of the immorality of Act X, even Act X is advisable to the perpetrator.

I am largely persuaded that a game-theory approach to morality most closely adheres to the reality. A game-theory approach concerns best strategies - i.e., it concerns what is advisable. But it is not reducible in any very straightforward way to what is advisable to a particular player of the game.

If we assume that the other players are following such-and-such strategies, then there is some optimal strategy for a particular player. But the optimal strategy for a particular player is relative to the actual strategies of all the other players, so absent an assumption about the strategies of the other players the optimal strategy for a particular player is undefined - i.e., there is no such thing as that which is well- or ill-advised. And so it does not make sense to say, "assume all the players are following optimal strategies" - at least, not unless you refine the idea into the idea of a Nash equilibrium (say).

One might define a crime as that for which, assuming that society is in a certain game-theoretic equilibrium (something like a Nash equilibrium), it is part of the optimal (and therefore actual) strategy to respond to violently.

Crimes are offenses that

Crimes are offenses that merit a certain severity of response.

I see this statement as true tautologically, and thus not very informative. The "correct" response to a crime depends on our underlying theory of punishment, on what we are trying to do with the law. If we are trying to make the victim whole again, then the appropriate "severity" of the response will look very different than if what we are trying to do is something like retribution. Unless we can agree on what the law is trying to accomplish, we have no basis for concluding anything on grounds of the response we happen to choose.

Tautologies sometimes need to be said

I see this statement as true tautologically, and thus not very informative.

Thank goodness, then, that that one sentence was not the whole of my commentary. Sometimes the blazingly obvious needs to be said

a) because sometimes people will surprise you by contradicting it, showing that it was not, after all, blazingly obvious to everyone, and

b) in order to help tie together the rest of the argument. If you chop a mathematical proof into careful steps, each step is liable to seem blazingly obvious and not very helpful, but they add up to a proof of a statement that is far from obvious. In fact, technically, all proved theorems are tautologies, but mathematics is far from useless.

Generally speaking tautologies are at worst harmless.

The "correct" response to a crime depends on our underlying theory of punishment, on what we are trying to do with the law.

I don't agree with that formula. Without going into detail, I think it is wrong for much the same reason the following would be wrong:

The "correct" price depends on our underlying theory of prices, on what we are trying to do with prices.

To which I would respond: while you can doubtless come up with an idea about the "correct" price, whatever that might mean, I am interested in the market price, and the market price is what it is as an unintended side-effect of thousands or millions of people engaged in specific trades. Nobody is in a position to decide the price of a good, and therefore it does not even make sense to talk as if someone is choosing a price in accordance with an underlying theory of prices.

Now, if you are talking price controls, instead of market prices, then it makes a lot more sense to talk about underlying theories of prices and what one is trying to do with prices. A state can impose a price in accordance with an underlying theory of prices, in accordance with what it is trying to do with prices.

But it does not make sense to talk about market prices this way. Market prices, while they are an effect of human activity, are outside of control by a human agency in much the way that a volcanic eruption or earthquake is outside of its control. So, natural law, market prices, volcanic eruptions, and earthquakes are what they are regardless of what theories you or I or everyone together might hold. And while someone in an insane asylum who thinks he controls market prices, natural law, volcanoes and earthquakes might be "trying to do" something with them, that really has nothing to do with them. That's just the lunatic's delusion.

There is a market price, and there is a natural law (or, in David Friedman's a-c framework, there is, I am fairly sure, a common law that the various courts would eventually find themselves moving toward in order not to go out of business), and neither one is a function of what either you or I are trying to accomplish.

Prices do, of course, change in response to supply and demand, and these can change. In principle, then, natural law might change, and I would argue that it is at the very least dependent on the species we are (rather than being universal over all specie), but in practice, while certain aspects of law do change (e.g. whether one drives English style on the left side or American style on the right side of the road), the basics do not change. Murder, robbery, theft - in most cases there aren't culture clashes about these. E.g., secretly poisoning a rival is considered murder everywhere.

As for the proximate activity. Buying and selling is the proximate activity that, in the aggregate, determines price. And resolving specific conflicts is the proximate activity that, in the aggregate, determines law. But just as the main interest of a buyer is to obtain the good at the lowest price possible, and not to shape the market price, similarly, the main interest of a party to a conflict is to bring about a resolution as favorable to himself as possible, and not to shape the law. The law, like a price, is an unintended consequence.

To be sure, people have been writing down the law since Hammurabi. But I think it unlikely that Hammurabi invented the law. Rather, he expressed his understanding of the law, and that law of which he was aware had been shaped over the generations preceding him. If he invented part of the law, that part is suspect, and would in any rate be tested over the future generations. The test would be similar to testing a price in the market - a given price will survive if it happens to be the market price.

Nice

Interesting way to break this down, but help me out here.

Can this:

Putting someone in prison for putting a thief in prison for one year merits one year in prison.

Logically follow from this:

Putting a thief in prison for one year merits no punishment.

Where did the "merits one year in prison" come from in the first example? And how did that become merited all of a sudden. That kind of thing hasn't been accounted for yet. Did you mean it merits no year in prison?

I'm not understanding you here in that specific example.

Very interesting hole you've found.

Can this: Putting someone in

Can this:

Putting someone in prison for putting a thief in prison for one year merits one year in prison.

Logically follow from this:

Putting a thief in prison for one year merits no punishment.

No, and it's not intended to. In fact, if it logically followed then my argument would be undermined, since I rely on it not logically following. My point is that merely saying that something is wrong is to leave important bits out. You are saying more if you also say what punishment the wrong deed merits - that is stepping back one level. And every time you stepped back one more level, you would be saying more. In the bits you mention, I am stepping back a level, so I am saying more. But if the farther level logically followed from the nearer level then I would not be saying more, undermining my point.

Moreover I wasn't offering either one as an actual concrete claim about a specific moral question, but as a hypothetical example of what such a claim would look like.

Also, you seem to think that the more nearly correct statement is the opposite of one of the statements I made (you suggested "no year in prison" instead of "one year"). I may have got it wrong, but if so then you can simply reverse the example to get a valid one.

Yes!

Moreover I wasn't offering either one as an actual concrete claim about a specific moral question, but as a hypothetical example of what such a claim would look like.

Right. I understand. Trust me, I got the point of your blog the first go around.

No, and it's not intended to. In fact, if it logically followed then my argument would be undermined, since I rely on it not logically following.

Okay. I didn't think so. Those recursive illogical hypotheticals...

My point is that merely saying that something is wrong is to leave important bits out.

The recursive examples tend to SHOW in a systematic way, how the "bits" can stand out.

But if the farther level logically followed from the nearer level then I would not be saying more, undermining my point.

Yes, I saw that. I just wanted to be clear that's what you were doing, and I wanted to be point it out to others. In my opinion it's not at all obvious if someone is just casually reading away to spot little simple things like that.

And by the way, I wasn't being sarcastic when I said this:

Very interesting hole you've found.

I quite liked the systematic approach you used.

But I was genuinely confused as to the "1 year in prison" remark.