The claim that the problem of punishment is a genuine problem is the claim that there is a genuine problem accounting for the moral permissibility of the state’s practice of punishing people for breaking the law.  The argument for the claim is straightforward: punishment involves the state’s treating people in ways that it is normally impermissible to treat people (e.g., taking their money, restricting their freedom of movement, forcing them to perform hard labor, killing them).  Given that these ways of treating people are normally impermissible, there is a genuine need for a satisfactory argument that explains why it is permissible to treat people in these ways when (and because) they have broken the law.

The Principle of Separability Argument: one reason for denying that punishment stands in need of moral justification rests on what Martin calls “a methodological principle of separability.”  On such a principle, politics and morals “are logically distinct.  Politics is not essentially, or logically, a special case of morals; and political theory is not an applied form of ethics” (1970: 254).  Since legal punishment is an essentially political institution, it follows that on this account “punishment does require a political justification.  But this is the only kind it needs” (1970: 255).  To justify punishment politically is to show “its ‘necessity’ within a particular system of political institutions and principles,” and “[o]nce the job of political justification is done,” Martin urges, “it is difficult to see any role for independent moral judgment” (1970: 254, 255).

First Objection: the argument rests on a confusion between two different senses of “justification”.  In one sense, to justify a practice is to establish that the practice is morally permissible.  In a second sense, to justify a practice is to establish that the practice serves a good purpose.  Martin’s defense of the claim that punishment needs a political justification but no moral justification is a defense of a claim about justification in the second sense.  Since legal punishment is a political institution, that is, the argument maintains that one can only show that it has a purpose by showing what its purpose is from within the political system as a whole.  Indeed, Martin at one point writes that “[t]o justify punishment is to display its rationale” (1970: 254) and this is clearly to speak of justification in the second sense.  But the problem of punishment is concerned with justification in the first sense.  The fact that punishment involves the state’s deliberately inflicting harm on some of its citizens does not raise the question, “does this practice serve a good purpose?” but rather the much more fundamental question, “is this practice morally permissible in the first place?”

Second Objection: The argument can succeed only if it is true that so long as a political institution is politically justified, it cannot be the case that its moral permissibility stands in need of justification.  But if this were true, then all sorts of clearly reprehensible practices could not be criticized on moral grounds.  The practice of slavery, for example, or of exterminating members of ethnic minorities in concentration camps, could clearly be “justified” in the sense that it could have a “rationale” from within a given political system, even a system in which political leaders are democratically elected.  If the principle of separability argument is correct, then we cannot criticize such practices as morally impermissible.  But this is plainly unacceptable.  And so the principle of separability argument must be mistaken.

The Logical Entailment Argument: A second argument against the claim that the practice of punishment generates a serious moral problem runs as follows.  There is no need to provide a moral justification for the state’s right to punish those who violate the law because the state’s right to punish such people follows simply as a matter of logical analysis from the very concept of a legal requirement.  Fingarette’s version of the argument runs as follows: (1) “If, when I will contrary to law, the law can or will do nothing, then . . . so far as the law is concerned my will is unqualifiedly unconstrained in this regard.”  But, (2) “the very point of a legal requirement upon me is that I am constrained, unfree, required to obey.”  Therefore, (3) to say that there is a legal requirement that I behave in a certain way just is to say that if I don’t behave in this way, I will “suffer some constraint upon my will.  Anything less than this makes the notion of ‘requirement’, as distinguished from ‘request’, unintelligible” (1977: 509).

First Objection: The claim that the concept of a legal requirement entails a punitive response to violations is false.  It may entail that that the state cannot “do nothing” in response to violations, but there are many other things it can do in response to violations that are non-punitive but which still redeem the state’s claim that they are violations of genuine legal requirements.  Examples: the state could (a) publicly denounce offenders, (b) banish them, (c) compel them to make restitution to their victims, (d) actively prevent people from violating requirements.

        First Response: these alternatives are not as practical/desirable as punishment
        Reply: the claim that they are logically possible suffices to undermine the argument
        Second Response: on some definitions of punishment, these are forms of punishment
        Reply: the argument attempts to show that the problem of punishment is not a real  problem even on the kind of definition
        of punishment developed in class.  And on  that definition, these are not forms of punishment since the harm involved is
        not intended.

Second Objection: even if the entailment claim is true, the argument based on it begs the question.  Now the problem simply becomes: how can it be permissible for the state to impose legal requirements (rather than other sorts of legal edicts, which would only entail restitution, etc., but not punishment) given that requirements logically entail a commitment to treating people in ways that it is normally impermissible to treat them.
 
Third Objection: a possible counterexample that a critic might appeal to: In many countries, including the United States, it is illegal to commit suicide.  The law, that is, requires its citizens to refrain from killing themselves.  There is nothing incoherent or unintelligible about the state’s requiring people to refrain from committing suicide and at the same time recognizing that it cannot punish people for violating the requirement.  And so, once again, there is no reason to think that the existence of a genuine legal requirement logically entails the existence of a practice of punishing people who violate it.