TRANS Nr. 17: 8.20. - John N. Williams: Beyond Minority Report: Pre-crime, Pre-punishment and Pre-desert

Utilitarian ethics are apparently incommensurable with other ethical perspectives. Faced with a choice between maximizing general benefit to society and committing an act of injustice, those of us who reject utility in favor of justice are powerless to change the viewpoint of someone who rejects justice in favor of utility. Since there are no higher ethical principles that overarch both principles of utility and principles of justice, both sides must run out of reasons when deciding which principles should be put first. But the unreasoned decision is ineluctable, because there are possible cases in which the principles conflict.(1) This kind of incommensurability in ethical values emerges again in the attempt to justify punishment. A desert-theorist â one who justifies punishment in terms of what the offender deserves â simply seems to have fundamental values incommensurate with those of a deterrence-theorist ne who justifies punishment in terms of the utility of the deterrent effect of the punishment. In what follows I examine a fantastic, yet coherent scenario which puts the desert-theorist back into dialogue with the deterrence-theorist. Both should judge that the form of punishment considered â punishing a person for an offence he will commit â is morally wrong. On both sides, there are good and bad arguments available for this common judgment. But it turns out that all the good arguments available to the deterrence-theorist are parasitic upon good arguments available to the desert-theorist. Therefore the desert-theorist has the better justification of punishment in general.

In the film Minority Report, (2002) adapted from Philip K. Dick's short story (1991), three mutants with the ability to predict only the actions and states of mind of citizens up to the next two weeks ahead with a 99% degree of accuracy, are used by the Department of Pre-crime to arrest and detain would-be offenders before they get a chance to carry out the actual crime. In this scenario, the mutants predict that the would-be criminal will commit the crime unless the Department prevents that crime. So while the mutants predict the would-be crime with a very high degree of accuracy, they do not predict their own success in preventing it. It is important to note that the mutants' near-perfect predictions do not amount to knowledge; firstly the mechanism by which they form them is obscure and secondly, they are unable to provide justification for their predictions â although inductive justification for them is available to the Department in the form of a long record of almost unbroken success.(2) The mutants are unfailingly honest; they never withhold or conceal information. Because the Department knows that the mutants are such reliable predictors, it is certain beyond reasonable doubt (from now on, 'reasonably certain') that would-be criminals will be criminals.

This scenario raises the question of whether the Department is ethically justified in treating citizens this way. A desert-theorist maintains that authorities have a duty to give their citizens what they deserve. She may argue that a citizen who is detained in order to prevent him from committing a crime he would otherwise commit is ipso facto innocent of that crime. Preventive detention is not punishment. Yet it penalizes the innocent detainee. So the authorities have a duty to compensate him or at least ensure that his life is only minimally inconvenienced. In Dick's scenario, the mutants' predictions include the specific future time of the would-be crime. Once the citizen has been detained longer than that time, it is impossible for him to commit the crime that was predicted. Since there is no point in detaining a citizen in order to prevent a crime that he can no longer commit, he should then be released.

By the desert-theorist's lights, if detaining the citizen counts as punishment, this must be because he is being punished at least for having intended to commit the crime (including a crime of passion, where the intention is very brief) or for planning to commit it as well. His detention will impose a penalty upon him for a period that he deserves for having intended or planned the would-be crime. The desert-theorist now faces the task of deciding how to balance or combine these periods of detention, one as punishment for planning the crime and the other as preventing it. One way would be to detain the citizen until the time of the would-be crime, while compensating him for this penalty and then detain him for an additional period without compensation, as punishment. Another would be to release him after the date of the would-be crime and then penalize him with a fine. Yet another would be to release him after the date of the would-be crime but reduce the compensation or comforts of detention in proportion to the gravity of planning the crime.

By contrast, a utilitarian response will differ in two ways. Firstly, the utilitarian will say that the authorities have a duty to do whatever it takes to maximize general benefit to society, one that may override any duty to the innocent detainee. For example, if compensating the detainee for his detention would cause general unhappiness or social unrest that outweighs the happiness of the as-yet innocent detainee, the utilitarian must withhold compensation. Secondly, the utilitarian will naturally justify punishment in terms of deterrence. So she may justify the detention as a penalty that deters others from committing similar crimes or even deters them from planning to commit them.

Dick's scenario may be seen as a thought-experiment that sharpens differences between a desert-theorist of punishment and a deterrence theorist. It does not provide a basis for deciding in favor of one or the other. But we can make the example a little more fantastic. Suppose now that the mutants can also see that the Department will not only remain powerless to prevent the crime but will also be powerless to punish the offender after he has committed it. So one salient difference from the original scenario is that the mutants predict that the individual will (with 99% likelihood) commit the crime, whatever the Department does to try to prevent it, so while they predict the actual crime with a very high degree of accuracy, they also predict their own failure in preventing it.

For example, suppose that the mutants let the Department know the following facts:

Suppose further that the Department has technology of a particular kind. It enables it to

identify Charles and inflict pain him so long as he remains within its jurisdiction. However, this pain cannot stop Charles from committing the battery. Moreover, the technology does not allow the Department to locate or apprehend Charles.

For example, suppose that all citizens have chips implanted in their brains. So long as a citizen remains within range of the Department's jurisdiction, it may identify that citizen and transmit a signal that causes him physical pain. We might even suppose that the implant gives the Department a way of constructing a visual depiction of the wearer's suffering that may then be broadcast to the general population as a deterrent. Nonetheless this technology is unable to tell the Department where that citizen is located.

Since the Department knows that the mutants have a reasonably long record of almost unbroken predictive success, it is reasonably certain that Charles will commit the battery tomorrow. It follows that it is in a position to be reasonably certain that nothing will happen that is inconsistent with Charles committing the battery, such as being killed by the painful transmission.

Although fantastic, this scenario is coherent â or so I will argue. And it confronts us with an interesting moral question. Let us grant for the sake of argument that punishing Charles (by inflicting pain upon him) after the battery, would have been justifiable.(3) Since the Department cannot 'post-punish' Charles, should it not now 'pre-punish' him by inflicting physical pain upon him today, rather than not punishing him at all? If the Department adopts pre-punishment it would still be reasonably certain that Charles will commit the battery. Accordingly it need not fret over the fact that there are no trials or public proof of guilt.(4)

In non-fantastic, ordinary cases, we might judge that punishing someone for a crime that he will commit is wrong because we cannot be reasonably certain that he will commit it. In the scenario we have envisaged this uncertainty is removed. Confronted with our new scenario, most of us will judge that it is wrong for the Department to pre-punish Charles and that what makes it wrong is not just the fact that we cannot we reasonably certain that he will commit the crime.(5)

Here is how I will proceed. I first defend the possibility of pre-punishment against two objections. I then consider arguments that could be used by the desert-theorist, and those that could be used by the deterrence-theorist to show that pre-punishment is wrong.

The desert-theorist may deploy three such arguments: that (1) pre-punishment constitutes a lack of respect for the offender's autonomy to make morally significant choices, that (2) it entitles the future offender to commit the crime and that (3) it is always a case of punishing someone who is innocent. I shall argue that there are two versions of (1) the first of which fails and the second of which may or may not be viable depending on what else we think about the nature of punishment, while (2) is mistaken and (3) is clearly effective.

Turning to the deterrence theorist, it appears that his most compelling argument that pre-punishment is wrong is that it cannot actually deter. However, I shall show that this argument is mistaken. Nonetheless, the deterrence-theorist has other arguments available, namely that (1) the deterrent aim of pre-punishment is immoral, that (2) there is no justification for thinking that the offender will be deterred, and that (3) pre-punishment entails the absurd consequence that if, counterfactually, there were a way to prevent the crime, it would be wrong to prevent it.

I show that these three reasons available to the deterrence-theorist are good ones,

but that they presuppose an appeal to desert. So the fantastic example reveals that the desert-theorist has the advantage. Hence my main conclusion: the desert-theorist has a better justification of punishment overall, even in ordinary, non-fantastic cases. Consistently with the conclusion that there is no pre-deserved punishment, I conclude by showing that Fred Feldman's (1995) apparent examples of pre-deserved charity, reward and apology are spurious.

Someone might object that the scenario we have considered is incoherent because an essential part of the meaning of the word 'punish' is that punishment is for a past offence (Cottingham 1992, 662). This objection parallels an older utilitarian response to examples given by the desert-theorist in which following the principle of utility commits us to punishing the innocent. The utilitarian might respond by insisting that such cases cannot arise, since by definition, punishment is not punishment of the innocent (Quinton 1969, 58â9). So since there can be no cases of punishing the innocent, there can be no cases of punishing the innocent that the principle of utility must bring about. But it is now widely accepted that this definitional manoeuvre misses the real issue (Ten 1987, 14â17). The mere insistence on the definition begs the question against a desert-theorist who holds that there is no reason why the term 'punish' cannot be extended to apply to the innocent. Likewise, even if our use of the verb 'punish' is normally backward-looking, that may simply be because we have not yet considered forward-looking cases like that of the highly accurate mutant predictors. So the mere insistence that punishment is essentially backward-looking just begs the question of whether such a case is possible. Whether suffering is inflicted on those who are innocent or on those who are future offenders, the real issue is not whether we should call the infliction 'punishment' but rather whether it is morally justifiable.

There is also a second argument that pre-punishment is impossible. This is that because future offences cannot take place now, there is literally nothing for which we can pre-punish, so since punishment is always punishment for something, pre-punishment is never punishment at all. The invalidity of this argument is demonstrated by a parallel argument against the possibility of pre-payment. It is an obvious mistake to argue that since services rendered in the future cannot be rendered now, there is literally nothing for which we can pay, so since payment is always payment for something, pre-payment is never payment at all. Although there is a sense in which there is now nothing for which we are paying, it does not follow that we are now paying for nothing.

We may conclude that there is no reason to think that pre-punishment is impossible.

A broadly Kantian argument that the Department would be wrong to pre-punish Charles is that doing so constitutes a lack of respect for the offender's autonomy or capacity to make morally significant choices. For example, Saul Smilansky objects to pre-punishment on the grounds that it fails to acknowledge the possibility that the person who is punished will make a choice that constitutes 'last-minute moral improvement' (1994, 52), namely to refrain from committing the crime. Likewise, Feldman writes (1995, 77):

But the Department's reasonable certainty that Charles will batter his victim tomorrow does not entail that Charles will have no choice tomorrow whether or not to batter him.(6) All that it does entail is that the Department is reasonably certain of the way that Charles will use that choice, namely to batter his victim. Even if the Department knew that Charles would batter his victim, this would still not rule out Charles' ability to choose tomorrow not to batter his victim. Of course it is logically necessary that if the Department knows that Charles will batter his victim tomorrow then that is what Charles will do tomorrow. But this necessity lies only in the connection between what is known of tomorrow and what is true of tomorrow. It would be fallacious to conclude that since the Department knows that Charles will batter his victim tomorrow then it is logically necessary that he will do so tomorrow. This modal fallacy is demonstrated by noting that although it is logically necessary that if Smith is a bachelor then he is unmarried, we cannot legitimately conclude that since Smith is a bachelor, it is logically necessary that he is unmarried. Moreover, if the Department's knowledge that Charles will batter his victim tomorrow did entail that Charles will not make a free choice tomorrow then it would be reasonable to claim that its knowledge that he committed the battery yesterday entails that he did not make a free choice yesterday either. It is not as if the Department's knowledge, foreknowledge or correct prediction of Charles's battery forces him to commit it. Rather, Charles's making the choice he does or will make is merely a necessary condition of the Department's knowing or foreknowing it. Likewise, just as Charles's choice yesterday to commit the battery is what allows the Department to know that he has so chosen, so Charles's choosing tomorrow to commit it is what allows the Department to correctly predict that he will commit it.

However, autonomy might form the basis of a subtler argument that pre-punishment is wrong. This argument starts with the claim that the infliction of pain cannot count as punishment unless the offender knows what crime it is for which it is punishment. But this means that when pre-punishing Charles, the Department must inform him that he will commit the crime. Since this might come as news to Charles himself, he appears to have grounds to complain that the Department is not treating him seriously as if he is an autonomous agent â despite the fact that he still is autonomous.

However it is not so clear whether punishment does require that the person who is punished know what offence has attracted the burden involved. Consider a father who sends his errant son to his room with the words 'â¦ and while you're there, think about why you are being punished'. Would the son's failure to work this out mean that the confinement isn't fair punishment â or even not punishment at all? If it wouldn't, then the argument seems to have no legs.

On the other hand, suppose that the basic claim is correct, with the result that the Department must inform Charles that it is about to inflict pain upon him as punishment for the battery he will commit tomorrow. In that case, is Charles is entitled to think that the Department is treating him as if he has no choice but to batter his victim? We might object that he is not so entitled, by simply rehearsing the argument above that a correct and reasonably certain prediction of how he will use his choice is consistent with his still having it. But we should admit that even if the Department is in a position to make that rather sophisticated philosophical objection, then it should recognize that as an ordinary citizen, Charles is unlikely to appreciate the force of it. I conclude, not that the second argument from autonomy is untenable, but rather that it is not clear whether it can be made to work.

A desert-theorist might claim that the reason why pre-punishment is wrong is that it is like purchasing an indulgence for future sin. More precisely, this is the claim that by pre-punishing Charles for battering his victim tomorrow, the Department has given him the right to batter his victim tomorrow.(7)

This claim might be supported by a commercial analogy. The desert-theorist might see punishment as payment, and so see pre-punishment as a kind of pre-payment. But just as a person who has pre-paid for goods has a right to them, so in being pre-punished, Charles has the right to batter his victim tomorrow. Thus it can never be justifiable to pre-punish, because it is always wrong to punish someone for doing what he is entitled to do.

Christopher New however, claims that punishment is not like payment, thus the commercial analogy is misleading. He points out that the purchaser has a right to his goods, because he has made an implicit or explicit contract with the vendor to receive them upon payment. But, according to New, no such contract exists between the offender and the authorities (1992, 38).

But this objection is not decisive. Some might hold that there is an implicit social contract between the offender and the authorities; all members of society, as potential offenders, have implicitly contracted to submit to punishment for an offence if they commit one.

We could instead object that the desert-theorist's conclusion â that it is never justifiable to pre-punish â does not follow from the fact that it is always wrong to punish someone for doing what he is entitled to do. For setting pre-punishment aside, it is always wrong to punish someone for doing what he is entitled to do because the person in question already has the right to do what he is then punished for doing. In pre-punishment however, things are different, for there, according to the argument from entitlement, the person who is pre-punished only acquires the right to commit the crime after he has been punished and so was not, at the time of punishment, already entitled to do what he was punished for doing.

The desert-theorist could give a different reason why it is wrong to entitle Charles to commit the battery tomorrow, namely that so doing will entitle Charles to batter his victim. But since his victim deserves not to be battered, it is wrong for Charles to batter him. So pre-punishment is wrong because it is wrong to entitle someone to do what is itself wrong.

But this reply is incoherent, since it is impossible to give anyone the moral right to do what is morally wrong. So there can no such entitlement.

This last claim â that it is impossible to give anyone the moral right to do what is morally wrong â might be challenged by the following example. Suppose that I share a house with my friends. Today is my turn to do the laundry, so neglecting to do it would be mildly wrong. I learn that my brother urgently needs to go to hospital. I promise to take him there, although other members of my family are willing to take him as well. Relying on my promise, they become unavailable to take my brother to the hospital. It now becomes my duty to take him, because I have promised to do so and I am the only person available. But taking him to hospital means that I must neglect to do the laundry. By making the promise, I have given myself the moral duty, and a fortiori, the moral right to do what is wrong, namely neglect to do the laundry.

Even if this example is successful, it may simply be avoided by substituting the plausible claim that it is impossible to give anyone else the moral right to do what is morally wrong. Anyway, we may point out that my duty to take my brother to the hospital is much more important than my duty to do the laundry. Since I cannot discharge both duties, acquiring the duty to take my brother to the hospital releases me from my duty to do the laundry. So although I have entitled myself to neglect the laundry, neglecting it is not wrong.

A better objection to the claim that pre-punishing Charles for battering his would-be victim entitles him to commit the battery, is found by comparison with post-punishment. For if pre-punishing Charles for battering his victim tomorrow now entitles him to batter that victim tomorrow then post-punishing him for having battered a victim yesterday must now entitle him to have battered that victim yesterday. But this entails the absurd result that no form of post-punishment whatever may be morally justified, since it is always wrong to punish someone for doing what he was entitled to have done.

The temptation to make the incorrect claim that pre-punishing Charles for the battery entitles him to commit it, arises from the correct observation that pre-punishing him does entitle him to immunity from further punishment if he does commit it, as is highly likely. Charles cannot be fairly both pre- and post-punished, because nobody can be given fair and full punishment more than once for the same crime. But it does not follow from this that pre-punishing Charles gives him the right to commit the battery. Charles may have a moral right to immunity from post-punishment, but that is because this immunity is not wrong, but indeed only fair. But nobody, including Charles, has a moral right to do what is morally wrong.

We may conclude that neither pre- nor post-punishment entitles the person who is punished to commit his crime.

The argument from entitlement to crime is not a viable route for the desert-theorist. It remains unclear whether a version of the argument from autonomy is on the cards. However, there is a clearer and more obvious reason the desert-theorist should endorse why pre-punishment is wrong. This is that pre-punishment is always a case of punishing someone who is innocent. For if the Department punishes Charles today for the battery he will very likely commit only tomorrow then surely today Charles is innocent of the battery. The desert-theorist holds that a person deserves to suffer in proportion to his moral culpability and so will say that Charles deserves no suffering today.(8)

New objects to this argument. He agrees that it is wrong to punish an innocent person but denies that Charles is innocent when he is pre-punished. He claims that that in place of our ordinary perspective on punishment, we should uphold the principle that 'it is wrong to punish someone for an offence which he never commits' (1992, 37). Since New agrees that it is wrong to punish the innocent, he is committed to accepting

But there are two serious problems with this redefinition of guilt. The first is that it conflicts with our intuition that we cannot punish someone for what he is (including being a future criminal) but only for what he has done. The second is that the redefinition makes it impossible to give a coherent account of the idea of moral decay. For if we accept that someone is now guilty of the crimes of his future selves, then we are forced to say that nobody can ever pass from a state of relative innocence into a state of relative guilt. Thus a baby is as guilty as any of its future selves, since any crime it commits in the future will disqualify it from innocence of that crime at any earlier point in time. Since we need to make sense of moral decay, we should not abandon our ordinary perspective on innocence, according to which, Charles is innocent of the crime for which the Department pre-punishes him. So the argument that pre-punishment is wrong because it is always punishment of the innocent, is a strong argument after all.

Indeed the argument from innocence is the chief argument available to the deterrence-theorist for showing that pre-punishment is wrong.

In the next two sections, I consider what ammunition is available to a deterrence-theorist who wants to show that pre-punishment is wrong.

The deterrence-theorist might be tempted to say that pre-punishing Charles is morally unjustified because it can have no deterrent effect. But as New points out, 'in pre- as much as in post-punishment the penalty imposed may deter potential offenders, as also the actual offender, from committing other offences in the future' (1992, 38). If the Department broadcasts a visual depiction of Charles's suffering, this may well deter other citizens from planning similar batteries by letting them know that if they do so then they too will be pre-punished. Nonetheless the deterrence-theorist might insist that the deterrent effect of pre-punishment is necessarily more restricted than that of post-punishment, because it cannot deter the offender from committing the very same crime for which he is being punished. New assumes that this claim is true, but points out that it is also true of post-punishment (1992, 38). Nothing, including post-punishing Charles, can now deter him from having battered his victim yesterday, unless we accept backward causation. Thus even if it were true that pre-punishing Charles cannot deter him from battering his victim tomorrow, this would not pose a special difficulty for pre-punishment.

But New is wrong to assume that Charles cannot be deterred from committing the very crime for which he has been punished. Admittedly, since the Department is reasonably certain that he will commit the crime, it should be reasonably certain that pre-punishing him will not deter him from committing it. Nonetheless it might still be mistaken. Thus it does not follow that Charles will not be deterred from committing the crime tomorrow. We have already established that there is no reason to deny that Charles still has a genuine choice tomorrow whether to batter his victim. And since the Department's prediction might be mistaken, Charles might use that choice to refrain from battering his victim.

Indeed there seems to be an example where pre-punishing Charles does deter him from committing the very crime for which he has been punished. Suppose that the unpleasantness of his pre-punishment leads him to reflect on his would-be criminal career. He recognizes that since pre-punishment has been instituted, he will probably not continue to evade punishment. He also forms the reasonable expectation of further doses of pre-punishment for each of the batteries he had intended to carry out each day after tomorrow. Although he knows that he has nothing to lose by battering his victim tomorrow, he might know himself well enough to see that it is now in his interest to drop the habit. For example if he knows that the thrill of battering an innocent victim on any one occasion will reinforce the compulsion to batter other innocent victims, he may sensibly reason that his best course of action is to desist immediately from any batteries whatever.

8. Immoral deterrent aim and two other reasons why the deterrence-theorist must judge pre-punishment wrong

There is a better argument that pre-punishment is wrong that has to do with deterrence. Suppose that pre-punishment is to be justified in terms of deterrence. Let us now look at the Department's deterrent aim. In post-punishing Charles for having committed a battery, it is trying to deter would-be offenders, including Charles himself, from committing the same type of crime after the punishment has been given. This must be likewise true of pre-punishment. Thus in pre-punishing Charles for battering his victim tomorrow, the Department is trying to deter would-be offenders, including Charles himself, from committing similar batteries after the punishment has been given. But in pre-punishment, these offences include the very offence for which Charles has already been punished. If the Department succeeds, then Charles will refrain from battering his victim tomorrow. So the Department will have punished a person who, by any definition, is innocent. Thus the Department is trying to do something that is morally wrong. Since trying to do something that is morally wrong is itself morally wrong, the Department is wrong to pre-punish Charles.

It might be replied that we should redescribe the Department as only aiming to deter Charles from committing batteries on all occasions after that for which he has been punished. But in fact, this isn't even true of post-punishment. Suppose that the Department punishes Charles for having battered his victim the day before yesterday. By so doing, it cannot be trying to deter him from battering another victim yesterday. Moreover, the re-description forces us to say that by pre-punishing Charles, the Department is not trying to deter him from committing other batteries before that for which he has been punished. Now the aim of deterrence is much more modest than it should be. This becomes conspicuous if we suppose that Charles has been pre-punished for a crime that lies further in the future, or at least, as far as the mutant's predictive powers allow. There is now a significant portion of Charles's would-be criminal career that the Department cannot even try to change.

A similar argument that pre-punishment is wrong can be given by moral reformers. The reformer hopes that punishment for a crime will improve the moral character of the criminal to the extent that he will choose to refrain from the same type of crime in the future. But in pre-punishment this becomes the strange hope that the person who has been punished will not commit any future crime of the same type, including that for which he has been punished. The most successful form of reformation (or deterrence) is that which takes effect as soon as possible after punishment and which lasts as long as possible with respect to as many future opportunities as possible of crime of the same type as that for which punishment was given. If the reformer's hope is realized, she has punished an innocent person.

There is, moreover, a simple reason why pre-punishing Charles will normally fail to deter him from committing that very crime. As we have seen, although pre-punishing Charles does not entitle him to commit it tomorrow, it does entitle him to immunity from further punishment if he does commit the battery, because nobody can be given fair and full punishment more than once for the same crime. If Charles has faith in the justice of the system, he should recognize that he has nothing to fear by battering his victim tomorrow. So unless Charles has a special reason for not battering his victim tomorrow such as breaking a general compulsion to batter innocent victims, pre-punishing him will fail disastrously to deter him from battering his victim tomorrow. Thus the Department aims to do something, namely deter Charles from committing the battery, that it should recognize is likely to fail. Since the Department has no rational justification for thinking Charles will be deterred by pre-punishing him, it has no tenable moral justification for doing so either.

Since the Department has already pre-punished Charles for battering his victim tomorrow, it would even be unfair for it to prevent him from carrying out the battery, for then it would have punished someone who is innocent of carrying it out. By bringing about the innocence of a person who has been punished, the Department would have brought it about that an innocent person has been punished. To do so is morally wrong. It follows that if the Department's reasonable certainty that it is powerless to prevent the crime were confounded by the unlikely discovery of a way to do so, the Department would still be wrong to prevent the crime. By pre-punishing Charles, the Department has transformed its practical inability to prevent Charles from committing the crime into an absurd moral inability to do so. In other words, it has committed itself to saying counterfactually that if there were a way to prevent the crime, it would be wrong to so prevent it. But the Department would be wrong to say this, because it has a duty to prevent crime should it be within its power to do so.

The desert-theorist must now face the objection that if the Department does not pre-punish Charles then it will allow him to escape punishment altogether. But since punishment is what he deserves, allowing him to escape it is wrong. This objection dissolves under closer scrutiny. The desert-theorist must say that it is wrong to punish someone who is innocent of a crime because such a person deserves not to be punished. He must admit that it is reasonably certain that Charles will deserve punishment that he will evade. This is cause for regret, but not for remorse. For it does not follow that pre-punishing Charles is justified. Since pre-punishing Charles is punishing an innocent person, it is wrong because it is undeserved.

Both the desert and the deterrence-theorist can find good reasons for judging pre-punishment wrong. But the desert-theorist enjoys an advantage over the deterrence theorist. The desert-theorist should say that pre-punishment is wrong because it is a case in which an innocent person is punished, whereas the deterrence-theorist should say that it is wrong because it aims to bring this case about. So the deterrence theorist's position is subordinate to that of the desert-theorist's, for unless it were wrong to punish an innocent person, it would not be wrong to try to do so. The deterrent-theorist, who we may suppose to be a utilitarian, might reply that the world of Minority Report is unlikely to arise. She might add that the disutility of setting up this world outweighs the utility of preventing crime. But this is beside the point. Although my fantastic example requires circumstances that are unlikely to arise, the utilitarian is committed to doing something wrong in them, in virtue of a rule that she uses in all circumstances, including ordinary ones. Even if it would be wrong to set up the world of Minority Report, a utilitarian who finds herself somehow catapulted into it is likewise committed to doing something wrong in its circumstances, in virtue of a rule that she uses in all circumstances, including ordinary ones. That is enough to show that the utilitarian has the wrong moral compass, even in most ordinary circumstances (compare Honderich 2006, 104â109).

The other reasons to which the deterrence-theorist can appeal for judging pre-punishment wrong are likewise subordinate to those of desert. What explains why the Department will probably fail to deter Charles from committing the crime for which he is being punished, is his recognition of the principle that nobody can be given fair and full punishment more than once for the same crime. This same principle is what transforms the Department's practical inability to prevent the crime into an absurd moral inability to do so. Yet this principle is itself a consequence of the desert-theorist's most fundamental principle that a person deserves to suffer in proportion to his moral culpability.

If there are 'pre-deserts' that is, if someone could now deserve to be treated in a certain way because of what she will do or undergo, then it is natural to think that these pre-deserts could extend to punishment. Feldman thinks that there are cases of pre-deserved charity, reward and apology and gives apparent examples of each. Consistently with my claim that there are no pre-deserts of punishment, I conclude by showing that Feldman's examples fail to establish any pre-desert.

In Feldman's first example, the Make-a-Wish Foundation offers a trip to Disneyland to the most deserving sick child, one of which is certain to recover and one of which is certain to die. Feldman holds that

But this example is not decisive. One problem with it is our feeling that positive deserts must be earned, so strictly speaking neither child deserves the charitable trip. But putting this aside we just may as well describe the child who is certain to die as deserving the trip because it is now fatally ill, in which case the desert lies in the present, or as deserving it because it has contracted the fatal illness (which might conceivably be a type of 'earning') in which case the desert lies in the past.

His second example is of soldiers about to embark on a suicide mission. Instead of decorating the soldiers with medals posthumously, the commanding officer decorates them in advance (1995, 73). Feldman correctly dismisses the objection that the commanding officer has rewarded the pilots merely for the past act of volunteering for the suicide mission, but fails to provide a reason. The reason might be that in a parallel case of post-reward, the commanding officer should bestow a lesser reward upon all those who volunteered (by awarding them a less honorific medal) yet reserve a greater reward (the more honorific posthumous medal) to those who had completed the mission. Otherwise we would be forced to say that those who volunteered for the mission but failed to complete it are as deserving of honour as those who volunteered and succeeded. Since completion of the mission is a necessary condition for deserving the greater reward, that is what we should say too in the case where the medals are awarded before embarking on the mission.

But non-consequentialists might refuse to accept this claim. If the moral desert of an action is based on the intentions and dispositions of the agent rather than its outcome, it seems that there is no basis for pre-reward. On this view, the success of the mission was of importance only because it was evidence of the pilots past willingness and ability to complete it. What makes the pilots now deserve to be honoured is not that they will succeed in completing the mission, nor that they have merely volunteered to do so, but because they are now ready and able to complete the mission (a disposition that they may have worked hard to cultivate). On this way of looking at it, if the commanding officer is confident that the pilots are now willing and able to complete the mission, he should award the medal even if the mission is cancelled due to bad weather. But this is not a case of pre-desert.

Feldman's final example is a customs inspector who apologises to a traveler for having to search his bags. Feldman comments

But it is not clear that the officer giving is giving an apology that the traveler deserves. For all the officer knows, the traveler may be carrying contraband. Thus the officer is simply doing his duty in searching the bags and thus is entitled to invade the traveler's privacy. Since he is doing his duty, he is doing no wrong to the traveler, and since an apology is deserved only in virtue of a wrong, the traveler deserves no apology at all. The officer's words are merely an expression of regret in the form of an elaborate courtesy, much as a conscientious doctor injecting a smallpox vaccine into a child might say 'I'm sorry if this hurts'.

The contrary view of the matter is that the apology is deserved if the traveler is innocent of carrying contraband. But if so, then whether the apology that was made before the search was justified, will depend on the outcome of the search. This means that the officer would not be justified in giving an unconditional apology. To do so is assure the traveler that he will be found innocent, while at the same time searching his bags. Such an apology is insincere, and thus undeserved, since offering an insincere apology to someone does him a disservice. Thus the officer should really give the conditional apology, 'If your bags reveal no contraband, then I now apologise for invading your privacy'. But such an apology is incoherent. Suppose that the search reveals no contraband. Then it follows that the officer has already apologised before the search for invading the traveler's privacy. But the officer was then in no position to know the outcome of the search. This means that when he made the apology for invading the traveler's privacy, he was then in no position to know that he had made one. This is absurd. Nobody can genuinely apologise if he is unable to know that he is so doing. To avoid this absurdity, the officer should really say 'If your bags reveal no contraband, then I will apologise for invading your privacy'. If the bags reveal no contraband then the traveler will deserve the apology for having had his privacy invaded. Clearly this is not a case of pre-desert either.

1 The obvious case of course is that of Jim and the Indians. Jim is captured by a bandit who tells him that unless he selects and executes one of ten Indians, all ten will be executed. Utility demands that Jim executes an Indian of his choice, but justice demand that he does not, because it is unjust of Jim to execute an innocent person.

2 One reason to set up the example as involving predictions that do not amount to knowledge is to accommodate a possibility relevant to the deterrence-theorist of punishment â that punishing someone for a crime that we are reasonably certain he will almost certainly commit, may nonetheless deter him from committing it. I deal with this possibility in Â§7.

4 Those who think that these elements are needed for punishment to be justified might be persuaded otherwise if we also stipulate that the Department itself is fair and honest. Alternatively we might try building them into the example. Perhaps the general public could also be brought to see that the mutants are a source of certainty beyond reasonable doubt, one legitimately tapped, thus obviating the need for trial. Perhaps trials could be arranged. They would be strange trials indeed: first the would-be criminal is charged that he will commit a crime that the court is not able to prevent. Next, he is asked to challenge the testimony of the Department based on the near-infallible and honest predictors. The first step in this procedure seems to threaten the would-be criminal's autonomy. I will deal with this point in Â§4.

7 Feldman writes 'Suppose a person appears at the police station and offers to pay a fine for speeding. She says that she is going to speed later in the day, and wants to pay the fine in advance so as to avoid red tape and inconvenience ... To accept the money and agree that she deserves the fine would be to acknowledge that she is going to speed, and in effect to grant her permission to do so, and this the officer cannot do' (1995, 77).

8 It might be objected that in our actual legal system, criminal liability is grounded in, and that criminal punishment is therefore justified by, character traits, of which criminal actions are simple evidence. I reply that Charles's battery constitutes his conduct (the act of delivering the punch), his state of mind (his intention to harm his victim) and the result of his conduct (the black eye). So even if the Department knows that Charles now has the intention to black his victim's eye, but has yet to deliver the punch, he is presently innocent of battery. True, his criminal liability is grounded partly in his present intentionâbut only partlyâwhich is why the punishment is not justified merely by his present disposition to punch people just for fun. But couldn't we envision a stricter judicial system in which the formation of the intention is itself a crime? Perhaps, but then what I have said about the crime will hold, mutatis mutandis, of the intention.

8.20. Rationality, Ethical Incommensurability and Existential Communication

For quotation purposes:
John N. Williams: Beyond Minority Report: Pre-crime, Pre-punishment and Pre-desert - In: TRANS. Internet-Zeitschrift für Kulturwissenschaften. No. 17/2008. WWW: http://www.inst.at/trans/17Nr/8-20/8-20_williams17.htm

	Internet-Zeitschrift für Kulturwissenschaften	17. Nr.	Oktober 2011

Sektion 8.20.	Rationality, Ethical Incommensurability and Existential Communication Sektionsleiter \| Section Chair: T. Brian Mooney (Singapore Management University) Dokumentation \| Documentation \| Documentation