The Right to Punish

Hobbes’s plan is to begin with a definition of the term “punishment” and then to spend the bulk of the chapter discussing what can be derived from this definition. He starts as planned, but the second paragraph has a digression addressing a problem with the right to punish. After addressing that problem, he carries on according to his plan.

I spent a lot of time on the digression in the second paragraph because I think it involves an interesting slip on Hobbes’s part regarding the term “right.” Then I tried to make the case for the significance of two distinctions that Hobbes draws: one between subjects and enemies and the other between punishment and hostility.

What is the problem with the right to punish?

In 28.2, Hobbes says that he must address a problem about the sovereign’s right to punish. Unlike all of the other rights of sovereignty, this one, it seems cannot be derived from the social contract. A sovereign without the right to punish would be worthless, so this is a big deal. The solution, according to Hobbes, is that the sovereign’s right to punish is the sovereign’s right of nature (see 14.1). Since the sovereign has this right prior to the social contract, the right to punish does not have to come from the social contract.

I said there are three things wrong with this.

First, Hobbes is equivocating about the meaning of the term “right.” When he says that “the right which the commonwealth … hath to punish is not grounded on any concession, or gift, of the subjects” (28.2), he is using the term “right” to refer to claim rights. But when he says the sovereign’s right to punish is the right of nature, he is using the term “right” to refer to a liberty.

If the sovereign’s right to punish really is just a liberty, then there never was a problem in the first place. If it must be a claim right, then the solution is inadequate. (I think it’s the former, for what it’s worth.)

Second, I complained about the fit between the sovereign’s right of nature and the sovereign’s right to punish. The sovereign’s right of nature is the right to do whatever the sovereign thinks is necessary to preserve his or her natural life. Since most crimes do not involve the sovereign’s life, the connection between the right to punish and the right of nature looks tenuous.

Rosie, Kamyab, and Harry together made a pretty good case for an indirect connection: if the sovereign does not punish crime, the state will fail and the sovereign’s life will be at risk. That looks like a promising answer to me. I think Hobbes’s position on this point could be defended, but I still feel that it is awkward.

Third, I said that the right to punish must be given to the sovereign in the social contract. To make my point, I noted that the definition in 28.1 requires that punishment come from a public authority. Someone who has not been made sovereign literally could not punish: it would be logically contradictory.

What is the right to punish?

The sovereign’s right to punish has at least two parts:

  1. Permission: the sovereign would do nothing wrong in punishing.

  2. Status: the sovereign alone is capable of meeting the definition of punishment

The sovereign does not gain permission to punish (A) through the social contract, but does gain the status to punish (B) that way.

Who cares about the definition of punishment?

Anyone strong enough can hurt another person. If the sovereign has the strength to hurt people, why does it matter whether doing so meets the definition of punishment or not?

I think it does matter and the reason why can be discovered in two distinctions that Hobbes repeatedly invokes: the distinction between subject and enemy and the distinction between punishment and hostility.

The difference between punishment and hostility is that punishment is limited by the law in a variety of ways. Most importantly, punishment can only be used in response to a crime. By contrast, the sovereign may treat even innocent enemies with hostility.

What Hobbes is trying to do with all of these definitions is show how the sovereign could use violence in punishment without triggering the spiral of violence characteristic of the state of nature. The idea is that if the sovereign sticks to violence within the definition of punishment, the subjects will not be provoked to try to overthrow the sovereign. Punishment, as Hobbes defines it, gives subjects the ability to peacefully control the use of violence against them: comply with the law and you will be safe. Hostility, by contrast, offers no such assurance: the sovereign may strike enemies whenever doing so seem advantageous. The only thing in your control when faced with a hostile enemy is the choice about whether to strike the enemy first.

This is a pretty good argument for why the sovereign has to keep to lawful punishment and, in particular, why the sovereign should not hurt innocent people. But it is not obvious why only the sovereign can be eligible to punish. Why couldn’t anyone punish, provided they did not hurt innocents?

When Hobbes insists that only an authorized sovereign is capable of punishing he is effectively claiming that the system of restrained, lawful violence can be maintained only if the sovereign is the only one that can punish. It is not obvious to me exactly how he means to argue for this point. It is certainly a reasonable thing to believe. I do not know how the spiral of violence could be contained if punishment could come from a variety of different sources. But I wish Hobbes had devoted more time to thinking this through.

Didn’t Hobbes say the sovereign can hurt the innocent without injustice?

Will brought up the fly in the ointment: Hobbes’s insistence that sovereigns may, and often do, harm or even kill innocent subjects without injustice (see 21.7).

So on the one hand, Hobbes gives the sovereign strong reasons for sticking with the strict limits on punishment. But, on the other hand, he insists that sovereigns would not treat their innocent subjects unjustly if they killed them.

I think he is trying to have things both ways. He wants sovereigns to gain all the advantages that come from a lawful state, especially the ability to use force without sparking a spiral of violence. But he also wants sovereigns to be able to act outside the law when they think it is necessary to do so. And he is realistic enough to know that many sovereigns will make terrible decisions about how to use this power. But he despairs of finding a way to hold sovereigns accountable for their bad behavior. So he concludes that this is simply an inconvenience of life under the state that must be accepted.

Hobbes did not manage to square these two aspirations with one another in a satisfying way. But no one else has done so either. We want our states to follow the law but recognize both that we sometimes want the state to act outside the law and that we have little recourse against states that do so for bad reasons.

Key concepts

  1. The problem Hobbes sets for himself in 28.2
  2. The distinction between punishment and hostility
  3. Hobbes’s conflicted views on punishing (or simply harming) the innocent


Hobbes, Thomas. (1651) 1993. Leviathan. Edited by Mark C. Rooks. British Philosophy: 1600-1900. Charlottesville, VA: InteLex Corporation.