PHILOSOPHY 33

Hobbes on punishment

I tried to answer four questions today.

  1. What is the problem with the right to punish in 28.2?
  2. What does it mean for the subjects to authorize the sovereign to punish them?
  3. Why does Hobbes repeatedly try to distinguish punishment from what he calls hostility?
  4. Can the sovereign punish the innocent or not?

What is the right to punish?

Before getting to those questions, though, we should answer another one: what features does the right to punish have? Two seem pretty clear.

  1. Permission: whoever has the right to punish is at liberty to use force against those who are being punished.
  2. Exclusivity: the right to punish can only be held by one party, the sovereign.

I also believe that Hobbes thought the right to punish has a third feature. (It is debatable whether I am right.)

  1. Status: only the sovereign is capable of punishing. What I mean is that only the sovereign’s acts could count as punishment much as only the acts of a legally competent adult can count as making a valid contract.

What is the problem?

In the first half of 28.2, Hobbes says there is a problem with deriving the sovereign’s right to punish from the social contract. Then, in the second half of the same paragraph, he offers a solution: the sovereign’s right to punish is comes from the sovereign’s right of nature and not the social contract.

I said that there is a mismatch between the problem and the solution. As Aaron noted, the problem assumes that the right to punish is a claim right. The problem, as Hobbes described it, is that the subjects cannot have obligations to submit to punishment. This, Hobbes said, means that the sovereign’s right to punish cannot be granted by the subjects. That follows only if the right to punish is a claim right. What it means for A to have a claim right is that someone else, B, owes an obligation to A. If B does not owe an obligation to A, A does not have a claim right against B. That fits Hobbes’s point in the first half of the paragraph: the subjects do not owe an obligation to submit to the sovereign’s punishment, therefore, they cannot give the sovereign a (claim) right to punish them.

However, the solution in the second half of the paragraph describes the right to punish as a liberty, meaning that the person with the right lacks obligations not to do something. In this case, the sovereign is said to have no obligations not to use force against the subjects. Whether the subjects are obliged to submit to punishment is completely irrelevant to the sovereign’s liberty to punish them. All that matters is whether the sovereign has any obligations not to punish.

So when Hobbes said there was a problem for the right to punish, he was using the term “right” to mean one thing and when he said that he had a solution to the problem, he was using a different meaning for the term “right.” This is not a serious problem for him. But it does help to make sense of what is going on in that paragraph.

If successful, Hobbes’s solution would cover the first two features of the right to punish: it explains why the sovereign is permitted to use force and why the sovereign’s right is exclusive (everyone else surrenders their rights to use force).

However it is an awkward solution at best. The right of nature is the right to save your own life.

The RIGHT OF NATURE … is the liberty each man hath, to use his own power, as he will himself, for the preservation of his own nature; that is to say, of his own life; and consequently, of doing any thing, which in his own judgment, and reason, he shall conceive to be the aptest means thereunto. (Leviathan 14.1)

But most crime has nothing to do with the sovereign’s life. So that is awkward. And Hobbes’s definition holds that punishment can only be inflicted “by public authority” (28.1). We don’t yet know exactly what that means, but the right of nature does not make anyone a public authority.

So it seems pretty clear that we need more than just the right of nature. The subjects really do have to give the sovereign at least part of the right to punish in the social contract. If you read ahead in the chapter, you will see that Hobbes frequently holds that punishment must be authorized or come from an authority. So it seems pretty clear that what the subjects do in the social contract is authorize the sovereign to punish. That brings us to our next question.

What does it mean for the subjects to authorize the sovereign to punish them?

If you look through chapter 28, you will see Hobbes saying that the sovereign’s ability to punish logically depends on being authorized to punish by the subjects. This is clearest in the paragraph on usurpers, people who claim the right of sovereignty without actually having it. They cannot punish because the subjects have not authorized them to do so.

the evil inflicted by usurped power, and judges without authority from the sovereign, is not punishment; but an act of hostility; because the acts of power usurped, have not for author, the person condemned; and therefore are not acts of public authority. (28.6, emphasis added)

Presumably, then, when the sovereign punishes, it is by acts that do have “the person condemned” as an author. But what could it mean for the subjects to have authorized their own punishment? There are two possibilities.

  1. The subjects have given the sovereign permission to punish them.
  2. The subjects have taken ownership of the sovereign’s efforts to punish them.

Both are possible, but I think it’s most likely to be the second. (Why? Because the sovereign does not need to be given permission to punish. Sovereigns have all the permission to use force that they need from the right of nature; that’s what Hobbes said in 28.2.)

But Helena, and then Patrick, hit me with a tough question. Why doesn’t the following argument show it’s impossible for the subjects to authorize their own punishments?

  1. If the subjects authorized the sovereign to punish them, then they would own the sovereign’s acts of punishment.

  2. If the subjects owned the sovereign’s acts of punishment, then it would be as if the subjects were punishing themselves.

  3. But the law of nature forbids anyone from harming themselves.

    A LAW OF NATURE, … is a precept, or general rule, found out by reason, by which a man is forbidden to do that, which is destructive of his life, or taketh away the means of preserving the same; and to omit that, by which he thinketh it may be best preserved. (Leviathan, 14.3)

  4. Therefore, the subjects cannot authorize the sovereign to punish them (assuming that punishments are harmful, of course).

I said that I did not know how to answer this. I added that it struck me as a relatively small problem for Hobbes because he could have dropped the third premise without giving up anything important for his theory. While he says that people cannot be obliged to put their lives at risk many times throughout the book, he never again even suggests that they are not permitted to do so after that paragraph in 14.3. For example, he never suggested that soldiers do anything wrong by putting their lives at risk; his only point is that they have to do something special in order to be obliged to put their lives at risk (see 28.16). When we read Locke we will see what it is like for a thinker to have a real commitment to the idea that no one is allowed to put their lives at risk. Hobbes, by contrast, just doesn’t do much with this idea. So if he withdrew the suggestion that people are forbidden from putting their lives at risk from 14.3, nothing significant would change for his broader theory. That is what I mean in saying it is a small problem: it’s a problem because two things he said appear to be in tension with one another, but it’s a small problem because he could give one up without damaging his theory.

That is what I said and I stand by most of it. But I have a confession to make.

I lied. I actually do know how to answer this problem.

Why did I do that? Well, we had so much detail on the table already that I didn’t think we could add this without overloading everyone’s brains. The answer is long and not essential to your understanding of the text. So I’m going to post it at the bottom of this page. In the meantime, I’m going to finish my summary of what we did talk about. If you’re still interested when you get to the end, you can read about it there.

Why is punishment different from hostility?

If you look at paragraphs 28.3-28.13 and 28.22-28.23, you will see that there are two distinctions that consistently line up with one another:

  1. Punishment vs. hostility
  2. Subjects vs. enemies

Hobbes’s idea is that a sovereign’s use of force against its subjects (that is, punishment) is qualitatively different than the use of force against enemies (hostility).

Punishment is governed by rules and hostility is not. For instance, only the guilty can be punished, whereas enemies can be attacked without regard to whether they are nocent or innocent.

There are obvious advantages to being a sovereign’s subject rather than an enemy. If you know that the sovereign’s use of force against you is governed by rules, you do not have to fear surprise attack. Rather, your liability to violence at the sovereign’s hand is largely in your control. Follow the laws and you greatly reduce the chances of being hurt. And because you are much less afraid of the sovereign, you have no incentive to try to eliminate the sovereign before the sovereign eliminates you. This means that while sovereigns use force against their subjects, it does not produce the same spiral that violence in the state of nature causes.

The subjects’ authorization of the sovereign’s punishment is supposed to play a role in this, although I have to confess that I find it difficult to explain. This is the best I can do. Authorization is the way that the relationship between sovereign and subject is established. It is a way of signaling that the sovereign’s use of force against the subjects will be seen as limited as opposed to a war against the subjects. Here is how each side sees it.

For subjects, authorization is a way of distinguishing themselves from the sovereign’s enemies. I want my local sovereign to think of me as a subject such that the sovereign will follow the rules of punishment rather than using violence against me at will.

For sovereigns, authorization signals that the subjects will not regard the use of force in punishment as a war. Those being punished will resist, of course; that is their right. But everyone else will see the sovereign as using force in accord with the rules of punishment. By contrast, an unauthorized power, like a usurper, cannot punish, by Hobbes’s definition of “punishment.” So subjects cannot trust that the usurper will respect the other limits on force that follow from the definition of punishment.

So that is what the subjects give to the sovereign in the social contract: the status to punish under Hobbes’s definition of punishment. With that status, sovereigns can use force against their subjects without automatically starting the cycle of violence like the one in the state of nature.

It is natural to think that the sovereign remains in the state of nature with the subjects, as Julian suggested. The sovereign’s permission to punish comes from the right of nature that is left “as entire, as in the condition of mere nature” (28.2) and the subjects are, notoriously, at liberty to resist punishment. But the important point concerns the restrictions on punishment: the innocent are fair game in war but not for punishment. A sovereign that sticks to the rules that define punishment will not be at war with its subjects, in Hobbes’s opinion. Violence between sovereign and subject is regulated, especially where the innocent are concerned, while violence in war is not regulated. That is tremendously important. If the sovereign were at war with its subjects, the subjects would have the familiar incentives to kill the sovereign before it kills them (see chapter 13). Hobbes very much wanted to avoid that result.

Can the sovereign punish the innocent?

But didn’t Hobbes say that the sovereign can harm the innocent without injustice (see 21.7, e.g.)? Yes he did! So which is it? Can the sovereign punish the innocent or not?

There is no single answer here because our author wanted to have things both ways. So the answers are “no,” “yes,” and “mostly not.”

“No” because punishing the innocent is logically impossible. Punishment, by definition, requires a conviction for violating the law. So force used against an innocent person can only count as hostility and not punishment.

“Yes” because the sovereign can do anything to a subject without injustice. It would not be punishment according to Hobbes’s definition, but that is just a point about words. The things done to the subject could be identical to punishments: imprisonment, exile, execution, and so on. To put it another way, a sovereign could, in theory, decide not to punish subjects but instead to treat them with hostility. Hobbes’s position is that while that would be a violation of the laws of nature (28.22), it would not be unjust (21.7).

“Mostly not” because sovereigns have strong reasons for confining their use of force to the rules of punishment. Sovereigns that punish the innocent will undermine the deterrent force of the law. There is no reason to obey the law if you will be punished whether you do so or not. And subjects who fear arbitrary violence from the sovereign have a strong incentive to kill the sovereign first. Since sovereigns can safely use violence if they keep to the rules that define punishment, it almost always makes sense for them to do so.

In the end, I think Hobbes wanted to have things both ways. He wanted to gain the advantages of rules governing the sovereign’s use of force and also to keep the sovereign free from rules. As he wanted to have things both ways, he never fully resolved this tension.

Key concepts

  1. The problem Hobbes sets for himself in 28.2
  2. Do the subjects give the sovereign any part of the right to punish in the social contract?
  3. The distinction between punishment and hostility
  4. Hobbes’s conflicted views on punishing (or simply harming) the innocent.

The Helena Problem

Are you still reading? Boy, you are tough! Now I’m coming back to the problem that I know how to solve even though I said I did not. Let me repeat what the problem was. Hobbes clearly believed that the subjects authorize the sovereign to punish them. But this argument, which I have simply copied from earlier on this page, seems to show that the subjects cannot authorize the sovereign to punish them.

  1. If the subjects authorized the sovereign to punish them, then they would own the sovereign’s acts of punishment.

  2. If the subjects owned the sovereign’s acts of punishment, then it would be as if the subjects were punishing themselves.

  3. But the law of nature forbids anyone from harming themselves.

    A LAW OF NATURE, … is a precept, or general rule, found out by reason, by which a man is forbidden to do that, which is destructive of his life, or taketh away the means of preserving the same; and to omit that, by which he thinketh it may be best preserved. (Leviathan, 14.3)

  4. Therefore, the subjects cannot authorize the sovereign to punish them (assuming that punishments are harmful, of course).

In class, I said that he could abandon premise 3. True enough: he could. But here is the real answer: you can authorize someone to do something that you do not have the right to do yourself.

As you read this, I am miles away. In fact, I am probably asleep. But I can still hear you saying “What? That’s not helpful at all!”

I know where you are coming from. It sounds odd and it is not obvious how it answers the question. The idea is not difficult, but the path to get to it is a little twisty. Here we go.

You cannot give someone permission to do something that you do not have the permission to do yourself. But you can take ownership of someone’s actions even if you do not have the permission to do those things yourself. That is the crucial point.

It sounds weird, but an example will show that it is actually obvious. Think about criminal conspiracies. Suppose that I hire you to commit a crime, like killing my rival. When you kill my rival I am responsible for what you did. I own your criminal action of killing my rival even though I was not permitted to kill the rival.

Hobbes saw it that way too. To understand the passage that proves it, you need some vocabulary. In the passage, the term “author” refers to the person who authorized a representative’s actions. The term “actor” refers to the representative of the author or, in other words, the person who acts with the author’s authorization. OK, now let’s prove that you can authorize someone to do something that you have no right to do yourself.

When the actor doth any thing against the law of nature by command of the author, if he be obliged by former covenant to obey him, not he, but the author breaketh the law of nature: for though the action be against the law of nature; yet it [the action -mjg] is not his. (Leviathan 16.7)

Similarly, when there are “Facts [actions -mjg] done against the Law, by the authority of another … both the Author and Actor are Criminals” (27.27).

The author authorizes the actor to break the law. When the actor does break the law, the author is responsible for what the actor did; that’s what the author’s ownership of the actor’s actions involves. The author did not have the right to break the law. But the author authorized the actor to break the law. Therefore, it must be possible to authorize someone to do things that one has no right to do.

Remember that this is why we hold people who hire hit men responsible for the murders the hit men commit. When you get to the bottom of it, it’s a normal, obvious thing to say.

Now, going back to the problem, suppose the law of nature forbids you from harming yourself (premise 3 in the argument). Suppose that means you have no right (i.e. liberty) to hurt yourself. You can still authorize the sovereign to hurt you. That is because you can authorize people to do things you have no right to do. In other words, even if premises 1-3 were all true, the conclusion in point 4 would not follow. That is so because the subjects can authorize the sovereign to harm them even if they have no right to harm themselves.

And that is the answer to Helena’s problem.