1 Introduction

In October 2012 a UK television news broadcastFootnote 1 revealed a number of allegations of sexual offences committed by Sir Jimmy Savile, a radio DJ and TV presenter regarded by many as a “national treasure”. Jimmy Savile had already been dead for a year before the programme was aired. The police eventually investigated these allegations in what was dubbed “Operation Yewtree”.Footnote 2 In the course of the investigation an ever-increasing number of people recounted having been sexually abused by Savile in their childhoods; it also emerged that Savile may have been involved in other types of criminal wrongdoing. By January 2013, the Metropolitan Police Service—the police force leading the operation—estimated that Savile was involved in approximately 450 cases.Footnote 3 A report was published under the title “Giving Victims a Voice”.Footnote 4 Empowering victims was indeed the explicit purpose of the investigation and the report.

In the meantime everyone associated with Savile (he was the patron and founder of a number of charities, many of which were related to children) began to dissociate themselves from him. Statues were removed, footpaths and conference halls were renamed, charities closed down. There was even a suggestion by the Prime Minister of stripping Savile of the knighthood bestowed on him in 1990 for charitable services.Footnote 5

These practices are essentially punitive at the very least in the sense that they constitute an active disapprobation of the acts of the dead wrongdoer. Many may simply have wanted to wash their hands of Savile but the ways in which they went about it, mostly by removing his name from the historical record and destroying his reputation, also amounts to inflicting a sort of punishment on him.Footnote 6 This, in turn, presupposes that it is possible to hold the wrongdoer accountable for the wrongdoing.

One might object that it is inaccurate and inappropriate to speak of punishment with reference to actions on the part of civil society and that it is only punitive action emanating from the state that should be subject to the kind of scrutiny that I propose to develop in this article. The response to this preliminary objection is that the state always either is or ought to be involved in and concerned with such practices in one way or another. To name but a few reasons, first, the state is under a duty to recognise victims of wrongdoing as precisely that. Second, it is also under a duty to protect the reputation of alleged wrongdoers. Third, there may be circumstances in which investigating the guilt of the dead and passing judgment on their culpability is inescapable. Say, for example,—and this did happen in the case of Jimmy Savile—that one is held responsible as a participant in a crime of which the principal perpetrator is already dead. For a court to be able to hold the accomplice accountable, it cannot but examine the potentially culpable acts of the deceased principal offender.

Importantly, these practices reveal an intuition, which seems to be quite widespread: namely, that criminal wrongs do not disappear with the wrongdoer’s death nor is the need to condemn the wrongdoer eliminated. State institutions must therefore be able somehow to address these wrongs. This article is placed against the background of treating the deceased as the type of subject capable of receiving blame and the sense that a response to wrongs is required even after the death of the wrongdoer.Footnote 7 The main purpose is to begin to explore whether any philosophical support is available for such practices and intuitions.

I will explore from a theoretical perspective two questions, which have hardly been addressed in the literature at all. First, may the dead be punished or, to phrase it a little more broadly, may the state respond to wrongdoing in a way that can be understood as distinctively punitive after the death of the wrongdoer? Second, may the dead be held accountable for crimes they are alleged to have committed during their lifetimes? In the course of answering these questions I also hope to raise some issues relating to the pure variants of some general theories of punishment.

Three preliminary notes are necessary. My argument is largely based on the assumption that death marks the complete obliteration of the physical person and his/her consciousness. This is simply an assumption based on the current state of our knowledge, which I take to be acceptable to all. It does not amount to taking a view on the possibility of an afterlife. It therefore does not prejudice the conclusions regarding the political institution of responding to criminal wrongs, which I hope will be acceptable independently of one’s beliefs about our state after death.

Secondly, the reason I speak of “punishment” is because I do not have in mind punishment in its common sense understanding of bringing about some reduction in welfare but rather as some response, which still counts as specifically punitive and therefore distinct to other kinds of responses, to wrongdoing and as a response to wrongdoing.

Thirdly, the section on whether it may be permissible to hold a deceased wrongdoer accountable in court is underdeveloped in relation to the section on whether there may be a conception of punishment which may apply to such wrongdoers. The main reason for this is lack of space. I do, however, think it is important at least to raise the issue and highlight the main questions and principles at play.

2 May the Dead be “Punished”?

Let me start with the question of punishment specifically. Consider a defendant (D), who was put on trial, found guilty of having committed an offence and sentenced appropriately. However, before beginning to serve his sentence, D dies. Are there still grounds for imposing some kind of penalty? The answer to this revolves around one’s approach to the point and content of punishment.

2.1 The Deserving Dead

Some believe that punishment is deserved by wrongdoers, precisely because they have committed a wrong. They also generally believe that wrongdoers deserve something specific, that is to be made to suffer for their wrong.

Desert here is an action-guiding concept. To say that D deserves x because of his action φ is not only to make a judgement regarding the value of φ but also to say that D ought to be given x. This is so in cases of praise or rewards. To say that D deserves a pay rise means that, if P is in a position of giving the pay rise, then P has a putative reason to do so. Should P fail to give D the pay rise, then D is entitled to demand it. This much should be obvious. Things are not different though when what is deserved is not praise or reward but blame and suffering. If D has done wrong and therefore deserves punishment, then, if P is in a position to impose punishment, P has putative reason to punish D.

I say the reason is putative because some desert theorists believe that there may be other reasons trumping the desert-based reasons to punish. It is not entirely clear what these reasons may be but, at the very least, they must be of the same kind; they must be reasons pertaining to the deserving person qua person, which is what grounds desert in the first place. In the case of rewards, perhaps the deserving party turning down the reward is a good enough reason not to insist in giving the reward. In the case of punishment, perhaps something like mercy would outweigh desert. There may be good reasons not to punish someone who is, say, terminally ill. However, consequentialist or pragmatic considerations are not of the same kind and do not have the same force. If D deserves to be punished, it is not a good enough reason not to punish him that prisons are overcrowded or that he has escaped and is difficult to recapture. For retributivists, all wrongs ought to be punished.Footnote 8

It follows that, if a wrong is not punished and stays with the community, something is amiss and those who have the reason to punish have failed to act on the right reasons.

The death of one who has been proven responsible for a wrong and therefore deserves punishment is not a good enough reason to trump the desert reason to punish him. For it to be so, it must be the case that the deserving party’s demise results in either the wrong committed and/or the desert relation (i.e., that the wrongdoer deserves to be punished) being cancelled out. I do not see how either can be true.

First, the wrongdoer’s death cannot undo the wrong because the wrong is not an attribute of the wrongdoer for it to be obliterated along with the latter.

Second, desert would be cancelled out altogether with the wrongdoer’s death if the personhood of the dead were completely obliterated. But this is not so. Some of our personhood survives our physical existence. Our reputation, the ways in which we have interacted with others and the relations that we have forged, the things we have created, exist after our physical demise and still bear our mark. All these things that we leave behind can be interfered with and this is an interference with the extension of our person. This is what makes it meaningful—and I do not think that a retributivist would be able to contest this—to still speak of the dead as persons and to say that they deserve praise, reward or blame. Interference with our extensions after our death also amounts to us—note: in that specific sense of what “we” are after our death—being treated kindly or harshly, even though we are unaware of it.Footnote 9

What death may have an effect on, however, is the very possibility of punishing the dead. Whether this is the case depends on how one understands punishment.

The type of desert theory that I have in mind here takes a strong stance not only on the reasons for punishing but also on punishment itself. If we “count noises”, to quote Mitchell Berman,Footnote 10 most retributivists consider suffering to be the proper desert object and not simply hard treatment. Now, perhaps one can be subject to hard treatment or even harm while unaware of it (and this includes the dead). Hard treatment alone, however, or anything else that is not experienced by D does not satisfy the requirement set by desert theory. Perhaps the surviving part of the personhood of the dead can be subject to hard treatment but the dead cannot suffer.

Nor is it available to retributivists to argue that the death of the wrongdoer counts as sufficient punishment. First of all, death itself, i.e., the state of not existing any longer, may be harmful, because it necessarily means that the dead miss out on the opportunity to enjoy things that they have and could have enjoyed or continued to enjoy.Footnote 11 But death itself, i.e., the state of being dead, is not a form of suffering.Footnote 12 Now, imagine an alternative scenario in which D falls ill and suffers in his illness before being punished. It is not uncommon for people to think that in such cases the convicted “got what he deserved”. This may be partly true in that it may be the case that the suffering caused by the illness is more or less the same as the suffering that would have been inflicted by the state. What makes a crucial difference though is that the harm is not inflicted by the stateFootnote 13 and, more importantly, it is not inflicted in response to the wrong committed. Since the desert source, i.e., the state, is implied in the retributivist version of the desert relation, if there is no connection between the suffering and the desert base, then the retributive aim of punishing is not achieved.

A retributivist may, however, counterargue that there is nothing in the desert claim, i.e., the claim that the wrongdoing is sufficient reason for punishing the wrongdoer, to determine the desert object, i.e., what is deserved by the wrongdoer, as suffering specifically.Footnote 14 The desert object could, therefore, be something that might apply equally well to deceased wrongdoers. To counter this, a direct connection must be established between desert and the requirement that the wrongdoer experience his punishment. I believe that such a connection can be made.

The desert subject in the retributivist version of the desert claim is the conscious person with the capacity to reason.Footnote 15 To deserve something implies a degree of reciprocity (which is explicit in the very term “retribution”). One deserves something by virtue of a basis, generally something one has said or done. As I argued above, it does not follow from the deserving party’s demise that the desert relation expires. However, a necessary condition of reciprocity, which is that the deserving party has the same basic constitutive features as when the desert basis was established, has been eliminated with the desert subject’s death. To the extent that retributivists insist that desert is the only reason for punishing, then they must accept that the desert relation remains incomplete. It is also revealed that, when we do blame or praise the dead, we do so for reasons other than their “deserving” to be blamed or praised. I explore some such reasons in Sect. 2.3.

So this is where this leaves pure retributivism: D is judged as deserving punishment, from which it follows that the state is under a duty to punish D by making him suffer. D, however, escapes punishment because he dies, which makes it impossible for him to suffer. For some retributivists, especially those in the Kantian tradition, this allows for a wrongful state of affairs to be perpetuated. But it becomes clear that even the less metaphysically charged versions of retributivism fail to provide those to whom they ascribe the duty to punish with the resources to do so in a potentially significant class of cases of wrongdoing.

A necessary consequence of this is that desert theory is also unable to provide any satisfaction to the victims of a dead wrongdoer as well to as the rest of the community. This, of course, is not pure retributivism’s main aim. Such good consequences are a “happy surplus”, to quote Michael Moore.Footnote 16 Nevertheless, it must be somehow part of retributivism’s aims as a complete theory of punishment just as it must be somehow an aim of every theory of punishment, because we necessarily make sense of wrongdoing and punishment at least partly in terms of their impact on the community, which has been wronged.

2.2 The Fearful Living

Another way of looking at retributivism’s failure is in terms of its excessive focus on the individual wrongdoer and on the experiential, though non-consequentialist, aspect of punishment. The (typically thought of as an) alternative to deontological, pure retributivism is a consequentialist conception of punishment.

I assume that the most plausible consequentialist justification of punishment, as well as the most relevant to the question of what to do about criminal wrongs after the wrongdoer’s death, is a deterrence-oriented one. Of course, not individual deterrence in this case but general. In this view, punishment is justified in terms of its good impact on the community through providing disincentives to would-be offenders. Many objections to this view can be and have been raised. I will suggest that a certain conception of general deterrence can provide an appropriate conceptual and practical framework to posthumous blame and punishment. However, some of the familiar, general objections to consequentialism do eventually resurface. I will address these objections by placing the general deterrence argument  in a different framework in the following section.

The argument from deterrence boils down to the claim that the threat and prospect of punishment provides agents with a prudential, self-regarding reason, which, at least in most cases, is assumed to be sufficiently strong to outweigh most of the motivations to commit wrongs held by rational agents.

This claim rests on at least the following foundational presuppositions: (1) everyone can and does engage in instrumental reasoning; (2) prudential, self-regarding reasons can outweigh any other reasons or what may seem as different kinds of reasons can be reduced to prudential reasons; (3) prudential, self-regarding reasons have some weight for everyone; (4) the prospect of punishment is a prudential, self-regarding reason to which everyone can respond in a way that will guide his or her actions away from the wrong. This in turn presupposes that (5) it is rational to want to avoid being punished after death.

There is much that is controversial about assumptions 1–3 but I will not go into these controversies now. I will focus on whether general deterrence theory (GDT) can consistently and plausibly hold 4 and 5 in their combination.

Is it rational to want to avoid punishment after death? The conception of rationality on which GDT seems to rely is belief-based, because its main aim is to manipulate the motivations of people rather than to structure the reasons they have independently. So, for GDT, if one has good reason to believe or in fact believes that punishment after death can negatively affect one’s well-being, then it is rational to try to avoid that punishment.

One possibility is that GDT must rely on justified belief. In this case, it would be justified to threaten people with punishment to the extent that people have good reasons to believe that punishment is a bad thing that will happen to them posthumously.

This partly involves the belief that the likelihood of the state punishing people after their death is sufficiently high, as unconsummated threats of punishment do not have the same motivating force as ones that one can predict will be consummated. Let us suppose that this is the case and that threats of punishment are not empty. The more important question is whether we are justified in believing that punishment after death is something bad that we should prudentially try to avoid by acting accordingly during our lifetimes. Since justified belief is an objective matter, the grounds for such a belief must be worked out philosophically from the standpoint of the GD theorist.

If one does not believe in an afterlife that somehow mirrors our existence on Earth, then it is seems reasonable to not care about what happens after one’s death, especially if one is motivated fully and solely by self-regarding reasons. For this to be rational, it must be the case that our deaths mark our complete annihilation as persons, which, as I argued earlier, is not the case.

The question then is not whether we are still somehow present after our deaths but whether this sense of being present is one on which GDT can rely in order to justify posthumous punishment. It seems to me that the key is again whether the hard treatment must be experienced by the punished. I think that this plain version of GDT would require that it does. Although our personhood outlives us, our conscious selfhood does not. For the purposes of the GDT claim, the addressees of the threat of punishment must believe that punishment will have an impact on their ability to act on their preferences, which is clearly absurd after their deaths.

As I stated early on, I have been assuming that death is the end. But, of course, very many people believe that it is not and this should be considered when trying to justify punishment in terms of belief not least because people believing in an afterlife raise the claim that there is good reason to do so. This does not seem like the kind of claim on which GDT can rely. Whether there is such a thing as an afterlife (a question largely linked to the question of god’s existence) is fraught with controversy. If GDT took it upon itself to adjudicate between all the reasonable beliefs about whether there is an afterlife and what it may be like, it would remain inconclusive and debilitated.

But perhaps GDT does not need to adjudicate between beliefs in that way. Perhaps all that it needs to do is to track beliefs firmly held by a sufficiently large section of the population at large, which it can then systematise as social psychological data and judge accordingly whether the threat of posthumous punishment has any weight in people’s motivational structures. In that case, it is a question of numbers. If a sufficient number of people believe for whatever reason that being punished on Earth after death will be a bad thing that will happen to them, then GDT has everything it needs to threaten with posthumous punishment, because it is rational for people to be motivated by and act on that threat.

I do not see anything incoherent about this but, at the same time, it seems unworkable. First of all, it requires a wide overlap of beliefs, which is hardly attainable in contemporary societies. Second, it is exceedingly taxing because it requires constantly tracking people’s beliefs, relating them to each other, systematising them and so forth in order to draw up any punitive policy. For a theory of punishment that capitalises on efficiency to the extent that GDT does, this is a serious problem.

It appears that the versions of GDT explored so far share one shortcoming with pure retributivism in relation to responding to wrongs after the wrongdoer’s death, namely their experiential orientation. Retributivism, on the one hand, cannot justify punishment that is not experienced by the wrongdoer. GDT, on the other hand, cannot establish deterrence because the threat of punishment as an evil does not have any purchase, if the addressees of the threat cannot build the threat into their motivational structures because they will not be around to experience those evils.

However, there are still ways in which GDT is able to provide a viable justification for posthumous punishment, while still relying on justified belief and without abandoning its consequentialist and self-regarding orientation.

One might argue that, even though it may be irrational to worry about things that happen directly to us after our death, it is rational to worry about things that happen to others as a result of our punishment not as an expression of selflessness but because of the impact that these consequences on others will have on us.Footnote 17 Such bad things can and do happen to others. Posthumous punishment will almost certainly affect, for instance, the reputations and financial positions of surviving family members just as punishment during one’s lifetime does. The question then is whether it is rational for D to worry about the impact that D being posthumously punished will have on others.

There are at least two ways of thinking about this. First, no matter how egoistic GDT holds us to be, it is still the case that even the most hardened individualists will care for and have an interest in the well-being of at least some other people, with whom they are closely linked (children, lovers etc.). If the consequences of D’s criminal wrongdoing have an adverse effect on the lives of those to whom D is closely linked, this is something of concern to D. But, of course, as I have already argued, since D will not be there to experience this negative impact, then it can make no difference to D’s motivational disposition. Nevertheless, being aware of the eventuality of bringing unhappiness upon D’s loved ones may make a difference to D’s motivation during D’s lifetime. In other words, worrying now about the consequences that D’s actions will have for D’s loved ones in the future may go some way towards deterring D from offending.

If this seems a little far-fetched (not least because the eventuality of the adverse impact on others’ lives seems too remote for it to make much of a difference to most people’s motivational disposition), a second way of thinking about the same idea is arguably a little stronger. Even if we concede that we are completely self-regarding and egoistic (not a view that I share but accept for the sake of the argument here), we will still have to accept that our lives and well-being inescapably depend on the well-being of others in at least two very weak senses. First, for us to carry out the majority of our activities, it must be the case that others are in a position to contribute to the completion of our plans (while pursuing their own plans, not out of altruism). Second, and more importantly for my purposes here, many of the activities that we pursue, from making chairs to conducting research on long-term macroeconomics, are worth pursuing because their results will be enjoyed by others even after our deaths.Footnote 18 It would follow from this that, if we know with relative certainty that our wrongful actions will adversely affect other people’s ability to enjoy the product of our current activities after our death, we will be motivated not to commit wrongs because this will result in our activities losing much of their value for us now and therefore our well-being will be negatively affected during our lifetimes. Note that this argument is still experiential not in the sense that we will be there to experience the consequences of posthumous punishment but in that the prospect of the negative consequences of the posthumous punishment will negatively affect our life plans. It is still obviously consequentialist. It is also arguably self-regarding, because it concerns the effects on us and it does not presuppose an exclusively other-regarding concern for others.Footnote 19 Note also that punishment is an independent reason for not offending rather than enforcing a reason which one would have had independently, because it introduces a fact (the reduction in the well-being of those to whom our activities are linked), which would have been unavailable without the punitive intervention of the state.

Once again, there is nothing conceptually incoherent with this argument. However, there may be two independent reasons for which it needs to be qualified. First, for it to work, punishment of the wrongdoer and the consequences for the personal network of the wrongdoer’s must collapse into each other. In other words, the impact on the people linked to the wrongdoer would have to be significantly serious, indeed punitive, for its threat to have any purchase in deterring the possible wrongdoer. Perhaps a GD theorist would not be squeamish about effectively punishing the innocent but, nevertheless, this possibility should make us pause and think twice about how attractive such a proposition is. There may be also a related reason from within GDT making the argument less attractive. Namely, it is questionable whether people would be motivated to simply acquiesce to such disproportionate (and intuitively unfair) laws thus undermining such laws’ very effectiveness.

An alternative is to formulate a GDT-oriented justification of posthumous punishment in terms of how one’s well-being now will be affected by events at the point of one’s death and after it.

The most relevant and plausible way of thinking about this is in terms of informed desire satisfaction.Footnote 20 Some believe that our well-being does not necessarily depend on our experiencing pleasure but on the extent to which our desires are fulfilled. In other words, one may not only be harmed without knowing it or ever finding out about it; one may also lead a fulfilled life without knowing it. Suppose that I have the informed desire that my students, whom I know well and to whom I have talked on many occasions about their futures, succeed professionally. If these students never get in touch with me again after graduation, my desire will still have been fulfilled if they do succeed professionally even without me experiencing any such satisfaction. The same idea could be extended to events that take place after one’s death. If I have the informed desire that my daughter become a Member of Parliament, I die and then she is indeed elected, my desire will have been fulfilled.

The possible ramification of this for posthumous punishment is not that my punishment will affect my well-being after my deathFootnote 21 but that being punished after my death will affect my well-being now, because I know that punishment after death will frustrate the desires and plans that I form during my lifetime. This, the argument would go, should be sufficient rationally to motivate me to not commit criminal wrongs during my lifetime.

Before considering this argument in more detail, let me highlight one general problem about the informed desire satisfaction conception of well-being, a problem which will resurface a little later. In James Griffin’s words: “… one’s desires spread themselves so widely over the world that their objects extend far outside the bound of what, with any plausibility, one could take as touching one’s own well-being”.Footnote 22 There are two further and interlinked (and relevant to the issue in question here) extensions to the same argument about removing the experiential requirement. First, the range of things that may affect our well-being is vast. Second, our well-being may then depend too extensively on events, choices, successes and failures, which we cannot control.

I can see three ways in which the threat of posthumous punishment may frustrate D’s lifetime desires in a way that may have an impact on D’s motivations: (1) the punishment may threaten to frustrate the desire that D wanted to satisfy by committing the crime; (2) the punishment may threaten to frustrate a wider range of D’s desires, which may be unconnected to the crime; (3) the punishment may threaten to frustrate a general meta-desire, which underlies a sufficient number of D’s specific lifetime desires and which is assumed to be shared by a sufficient number of people.

Option (1) may have some purchase in the context of crimes, which are committed in order to satisfy long-term desires. Suppose that D embezzled a large amount of money to guarantee the future of his children. The problem here is that very few of the crimes that we would want to prevent by threatening with punishment after death are of this sort. The desires, if any, that, say, sexual offences or offences against the person satisfy are generally satisfied immediately through the commission of the offence itself. Therefore the scope of the threat of punishment would be rather narrow and fail to prevent the commission of a sufficient number of offences.

Option (2) avoids this error by dissociating the threat of punishment from the crime-specific desire. If, say, D commits a sexual offence, the threat could be directed to other desires of D’s such as his desire that his children are well off with the wealth that he has accrued. This poses some problems. It seems very difficult to single out in an agent-independent way the long-term desires, which a would-be offender would not want to frustrate. In other words, it is very difficult to draw any generalisations regarding desires that a sufficient number of people will share. One way around this would be for the relevant rule to be so general as to be applicable to everyone’s circumstances without specifying in advance which desires the punishment would frustrate (something like “benefits that D is proven to have valued the most during D’s lifetime or their equivalent will be removed from D’s estate”). This could work better assuming that it is feasible (which it may be but it will certainly be exceedingly difficult) to work out how various desires are structured for each individual offender. In any case, legal rules that remain indeterminate to such an extent would also pose serious rule of law problems. It could also easily have the opposite effect, because, if one does not know which desires eventual punishment will disappoint, one will either not have a strong enough incentive to comply or one may even have a perverse incentive to not comply repeatedly. Moreover, most surviving desires will be other-related, which raises the problem of unfair impact on third parties, which I identified earlier when considering whether the justifiability of threatening harming others connected to the wrongdoer as punishment for the latter (imagine, for example, that my daughter is removed from her seat in Parliament because of my wrongdoing).

Option (3) is similar to option (2) in that the threat is relatively general and left to be specified in light of each wrongdoer’s circumstances. The main difference is that this kind of desire is assumed to enjoy priority over other desires that we may form. It can also be specified to a greater degree than the general desires I discussed above. The most obvious meta-desire that I can think of is good reputation, which also includes good posthumous reputation (what the Greeks called hysterophemia). It is plausible to think that it is a desire that we all share to one degree or another. It is also prior to other desires in that, if it goes unsatisfied, satisfying other desires loses much or all of their value (most people would not want to be wealthy but disgraced) and in that not satisfying it may in fact jeopardise the satisfaction of other desires (this applies to one’s reputation while alive but also to hysterophemia; many of our long-term desires largely depend on the maintenance of our good reputation).

It is plausible that threatening the satisfaction of hysterophemia (or any other meta-desire) will have a deterrent effect. There are, however, at least two independent problems, which go back to the general problems about GDT, which make it a rather unattractive option, if left unconstrained. First, it seems disproportionate to most crimes. One might think that reputational damage is inescapable when one is tangled up as a defendant in the criminal justice system. This, however, is a side effect, which in fact the criminal justice system itself is under a duty to mitigate so as to ensure that the punishment does not exceed the offence (whatever the measure of proportionality may be). Second, targeting one’s hysterophemia directly and widely will also adversely impact third parties and especially those whose well-being depends on the satisfaction of D’s meta-desire.

2.3 The Dead and the Living in Community

Recall that the failure of making sense of posthumous punishment in terms of deserved suffering indicates that we are in need of a justification that looks further than the wrongdoer as the immediate target of punishment while not losing sight of the fact that punishment is a response to a wrong. Although general deterrence offers this and can play some role in grounding posthumous punishment, it cannot serve as the sole or as a freestanding justification because of the limitations flowing mainly from its tendency to punish third parties and its tendency to punish in ways that exceed any intuition regarding the gravity of criminal wrongs.

What is therefore required is a way of thinking about posthumous punishment in a relational, non-experiential way, which may also be able to accommodate constrained consequentialist considerations. One such way is in the terms of a political justification of punishment. To explain this, I will begin by outlining, admittedly cursorily, what I have in mind.Footnote 23

The fact that we are members of an organised, governed political community generates specific, political normative relations between us, which are public/political in nature. Publicity has at least two senses. First, it refers to the capacity in which we relate to each other, that is as members of the political community and in our extensions in the world and not as moral agents. The public character of these relations may determine their form but does not and cannot determine their content.Footnote 24 This can only happen against the background of facts as they develop in each specific political community making our external relations public/political in the second sense. In this step, the question ceases to be formal and philosophical; it is situated in real contexts and becomes an inquiry into the basic and irreducible facts that animate our institutional structure.

There is some disagreement in modern political philosophy as to what the basis upon which the state and law may be constructed. One point of convergence—certainly one shared by theories placing themselves in the post-metaphysical, constructivist tradition—is that late modern constitutional democratic states are based on a political conception of the person as free and equal in the sense that there can be no a priori valuable conception of the good or, in a different formulation, valuable mode of exercise of private autonomy.Footnote 25

This political conception of the person and the political community has various upshots for institutions of criminalisation and punishment. First, the justification and content of such institutions are always public in the sense that they relate to individuals in their capacity as members of the political community (they are political in the first sense identified earlier). Second, the justification of criminalisation and punishment become part of the institutional structure and can only be justified and shaped in relation to it and not in a freestanding manner (they are political in the second of the above senses). An extension of this is that criminalisation and punishment are contingent institutions; there is nothing necessitating the category of crime or the practice of punishment. Third, and this specifies the previous point for our political communities, whichever content criminalisation and punishment are given, they must always be respectful of the political conception of the person as free and equal. Fourth, it follows that crimes can only be public wrongs, i.e., violations of political duties specified by the institutional structure. A further implication of publicity is that the wrongs are of concern to the whole political community and not only to those at the receiving end of the wrong, because it is a disruption of the institutional structure which has an impact on everyone participating in it. From the fact that a wrong is a violation of a duty, it follows that there is a reason to respond to this violation with an accountability-seeking measure. Very importantly, this is already justified to the wrongdoer as it is a term of the institutional structure that he or she has already accepted by participating in it. From the fact that the violated duty is political follows that it is the political community as a whole and as represented by the appropriate institutions that is entitled to respond to the violation. Fifth, what the response will be is not a matter of necessity but of appropriateness. Whether it will be what we normally consider as punishment or something else depends on which measure will best serve the stability of the political community in the face of its disruption through the wrong.

Let us return to our fictional character D who has committed a wrong and died after having been found guilty and liable but before serving his sentence. In the political understanding of crime and punishment, the reason for responding to D’s wrong does not expire with his death. The disruption to the institutional structure and the impact that the wrong has had on the political community by reshuffling normative positions in an unauthorised manner cannot be extinguished by D’s death. In this political conception, the wrong is, of course, attributable to the wrongdoer but also becomes part of the political community.

As I said earlier, the reason for responding to a wrong is already justified to D to the extent that he is a member of the political community and the reasons introduced by the institutional structure apply to him. One may ask, however, how this can be the case after one’s death, given that no reasons may possibly apply to the dead. I think the answer is straightforward. What is justified to the person is the eventuality of a response to his or her wrongdoing (under certain procedural and substantive conditions, some of which I will discuss a little later on). This has already taken place during the wrongdoer’s lifetime and this suffices.

It does not, however, follow from the fact that there is a reason for responding to a wrong even after the wrongdoer’s death that a response is necessary or that it necessarily ought to take a specific form. This is because a response does not “undo” the wrong by restoring the victim’s rights. Whether to respond and how to respond depends on whether a specific response is the appropriate one for cancelling as much of the impact of the wrongdoing as possible. It follows from the public character of wrongs and institutions of punishment that the impact that needs to be cancelled out is the impact on the political community. In short, appropriate responses to wrongs are those responses which can maintain the stability of the political community.

What, then, may guarantee stability? Since the content of institutions can only be determined with reference to a specific institutional structure as a whole and since my argument in this article is mostly formal, I will not make any concrete suggestions as to what kind of response may best serve the aim of stability. However, a few things can be said about the general direction that any such institutional responses should take based on the impact that institutional responses are meant to reverse.

Public wrongdoing does not only amount to harming the victim or changing the normative relation between the victim and the wrongdoer. It also marks a change in or a threat to the normative standing of the victim in the political community. Although I do not have the space to develop this idea in detail here, I should highlight that there is no metaphysical overtone in it. The wrongdoer’s acts impact upon the victim’s interests in a way that deprives the latter of the ability to act on the reasons which would have otherwise been available to her. The wrongdoing therefore forces the victim into a different normative role in relation both to the wrongdoer as well as the rest of the community.Footnote 26 How this may be varies from wrong to wrong. For example, the change in the normative position of the victim of a property offence and that of a victim of violent crime is different but what matters is that there is a change (and I mean prior to and independently of the involvement of the victim in the justice system as a victim). The organised response on the part of the state to the wrongdoing addresses this in four interconnected ways: (1) it recognises the change in normative status of the victim by involving him or her as the wronged party in the process of responding to the wrong; (2) it reaffirms the proper normative status of the victim by responding to the wrongdoing; (3) it reverses to the extent possible or makes amends for the actual consequences of the wrongdoing on the victim’s interests; iv) it recognises the change in the normative standing of the wrongdoer and the need to reverse that.

The promise on the part of the state that it will respond to wrongdoing and in fact responding also maintains stability by reassuring members of the political community. It provides assurance to the actual victims by reversing the effects of the wrong to their normative standing. It assures members of the political community at large that, should they become victims of crime, the change in their standing in the community will only be temporary and that their proper status will be reaffirmed. It also offers them some assurance that the actual consequences of their victimisation will be addressed.

Since this aim of responding to wrongdoing is not directed at the wrongdoer as a conscious agent, his or her death makes little difference. It is still appropriate to respond to his or her wrongdoing. To illustrate, recall again the Metropolitan Police’s explicit aim of empowering victims by giving them a voice. To give victims a voice is already to recognise them as victims and this presupposes that someone’s wrongdoing has rendered them victims. So, despite the care that the Metropolitan Police took to emphasise that this is not a case of punishment because the alleged wrongdoer is deceased, to empower the victims presupposes a framework, which can justify blaming and punishing the dead wrongdoer.

A further way of maintaining stability by providing another reason to members of the political community to act on institutional reasons is by providing assurance that others will have reasons not to commit crimes—in other words, by offering some guarantee of general deterrence. This works well and complements the political theory of crime and punishment in normal circumstances. When it comes to posthumous punishment, however, some of the problems with general deterrence that I identified earlier apply here too. Placing deterrence in a political framework does not particularly help with the difficulty in manipulating the motivational disposition of the living if the threatened unpleasant consequence of wrongdoing will never be experienced or with the difficulty in tracking sufficiently generalised beliefs about the afterlife and the impact that punishment on Earth will have on the dead.

Nevertheless, the political conception of punishment can make use of the motivational force of posthumous punishment threatening to frustrate D’s plans during his lifetime. It can also do so without running the same risks as GDT, i.e., punishing the innocent and imposing disproportionate punishment, because the pursuit of the aim of deterrence is constrained by the political framework in which it is placed.

These negative constraints are the following. First, whatever the response to D’s wrongdoing may be, it may not have an undue impact on third parties. Say, for example, that the most appropriate thing to do is to confiscate part of D’s estate. This measure may not interfere with entitlements that D’s heirs would have had independently of D’s death or D’s wrongdoing, because this would be unjustifiable (they have committed no wrong) and because it would undermine assurance in the political community. This is not to say that the response must be necessarily connected to and impact on the wrongdoer’s surviving personhood. The most appropriate course of action may be to just “give victims a voice” or for the state to somehow compensate them for their losses. But it will more often than not be appropriate to impact on the wrongdoer. Not for a reason that has anything to do with desert but because it is important that the authorship of the wrong is recognised so that the normative standing of the victim as well as the wrongdoer be restored.

Second, the response must be proportionate. This is not in the sense that there is some metaphysical exchange rate between wrongs and sanctions, as much of retributivist theory holds. Recall that in the political conception of punishment proportionality is mediated and determined by the need to guarantee stability. Punishment in the strict sense, however, is only one among many instruments that can contribute towards that aim. Not only must it be kept as last resort because citizens should be given the chance to discharge their political duties without being coerced, but it must also operate in conjunction with measures which will allow the reaffirmation of the status of everyone as free and equal members of the political community (such as restorative justice measures and so forth). This should preclude, for example, attempts at completely identifying a wrongdoer as a person with the wrong he committed thus overshadowing every other aspect of his life history and destroying his reputation altogether. In cases in which tarnishing one’s reputation is the only appropriate measure, punishment must be done in a way that is proportionate to the need to provide assurance, maintain stability and serves as an effective disincentive.

3 May the Dead be Held Accountable?

So far, I have been isolating the question of punishment assuming that our wrongdoer D died after having been tried and found guilty. This, however, will happen very rarely. Most cases will be like Jimmy Savile’s and the wrongdoing will be revealed after the alleged wrongdoer’s death. Are there then any grounds for holding the wrongdoer (still ‘D’ but this time he dies before being prosecuted) accountable according to the political conception of crime and punishment?

A good place to start thinking about this is to consider what the main aims of processes of holding wrongdoers accountable may be. The most basic aims are, first, to ascertain facts. Second, it is to apply the relevant law to the circumstances of the particular case. Application of the law requires justification of why the specific defendant is held accountable and ordered to be punished in a way selected and specified from a legislatively predetermined range of measures. It would also seem that these two aims apply to most criminal processes and not only the criminal trial. In fact, use of the trial as a mode of holding people accountable and imposing punishment is steadily decreasing. Nevertheless, I will refer to all accountability-holding processes as “the trial” for the sake of convenience.

The obvious difference between any regular criminal process and posthumous ones is that in the latter the defendant cannot be present. The question then is whether it is necessary that a defendant be present for a trial to be fair and, if so, why. I will approach the question in light of the political scheme outlined above and argue, however tentatively, that there may be ways of holding the dead accountable and that this resonates with many of our current practices. I also focus on D’s presence in person rather than D being properly represented. I take the right to representation not to be affected by D’s death.

If we think of the trial in political terms, we will see that the reason for D’s participation is that, although he is facing allegations sufficiently strong for him to be held accountable, the state must still treat him as an equal member of the political community. This is because D does not stand alone against the rest of the community, which is prepared to banish him and exact its vengeance; D is still part of the community, which therefore has a duty not only to ascertain guilt and responsibility but also to protect D. But the trial is not only a simple two-way interaction of negotiating the exchange of wrong for penalty between the state and D. The process is also the business of the political community as a whole in the much more extensive and substantive sense that having a proper process ascertaining facts and justifying the imposition of some measure is a means of maintaining the community’s stability.Footnote 27 The question then is whether this more complex aim of the criminal trial can be served in D’s absence.

It is arguably easier to answer this in relation to the fact-finding aspect of the trial. To start with, there are good reasons, reasons already animating the privilege against self-incrimination recognised by many jurisdictions as well as the ECHR,Footnote 28 for not imposing on D a duty to testify. But, of course, defendants also have a right to testify and give their accounts of events as they experienced them and as only they can express them.Footnote 29 This right, however, is not so strong and the information that D can contribute to the trial not so valuable as to provide a good reason not to hold people accountable when they cannot be present in person in this process. Although there is symbolic value in allowing D to give his personal testimony (a value linked to treating D as free and equal), the primary value of allowing D to have his say is instrumental towards ascertaining the truth about facts. In this light, the importance of D’s personal testimony is significantly reduced because the information provided by D lacks the objective strength to determine the outcome of the fact-finding process, if uncorroborated by objectively ascertainable data. I should therefore think that it may be of use in an extremely limited range of cases, which makes it possible to compensate with alternative institutional arrangements.

Things are a little more complicated when we consider the justificatory function. The trial does not justify the criminalisation of a certain act itself; this is the task of the legislature. It also does not justify the possibility of D being held accountable. This has already been justified to D during his lifetime and while D was a participant in the political community, the institutional structure of which D is held to have accepted (on the caveat that it largely treated participants as free and equal agents). What ought to be justified to D directly is the application of the pre-existing norm and the imposition of a penalty or some other accountability-seeking measure. A state that fails to do so also fails to treat D as a free and equal participant in the political community.

This, however, cannot be an absolute right. It must be seen within a wider scheme of the ways in which the trial is a manifestation of the way in which citizens must be treated by the state as citizens. The trial is also meant to protect D against unwarranted, not properly public accountability-seeking practices and punitive measures. It is therefore not only meant to give reasons to D as to why he is being punished but also to give reasons to the political community for not punishing D. Seen in this light, the state is not only at liberty to start proceedings against a dead alleged offender but also under a duty to do so. As the case of Jimmy Savile illustrates, the repercussions of allegations of crimes for the personhood of the dead can be very grave. Civil society tends to take measures which are essentially punitive (though not administered by the state) but not preceded by a proper ascertainment of guilt. Even state agencies are on occasion tempted to pre-judge the guilt and responsibility of the dead, especially when the dead’s alleged acts are connected to the wrongful acts of others.

At the same time, the public institution of the trial is also in the interest of the political community as a whole. Once a suspicion has been raised (and especially if this has happened as publically as it did in Savile’s case) that a crime has been committed, there is good reason for the state to try to restore stability and reassure the community. Stability and assurance, however, can only be achieved when pursued institutionally by public bodies operating as representatives of the political community. Therefore on balance, holding D publically accountable is to treat him as well as the rest of the citizens as free and equal participants in the political community.Footnote 30