3 Dynamics

Theseus had a thirty-oar ship that the Athenians continued to value greatly after his death because it played a key role in the religious ritual they offered annually to Apollo. Because it had to remain seaworthy and pristine for that ritual, they constantly refurbished it. As the wooden planks of the ship decayed, the Athenians removed them and put in fresh and stronger timber. At some point in time, then, in the centuries in which the ship sat in the Athenian harbor, Theseus’s ship came to be constructed of entirely new timber, as each original piece had been removed and replaced. Was this ship made of entirely unoriginal lumber still Theseus’s ship or was it something different? If it was the same, what constituted sameness? And if it was different, when did it become different? Was it when the first plank was replaced, the plank that represented more than half the original ship, or the last plank? At the heart of this version of the Sorites paradox, the question is when is there sufficient change, usually in quantity, so that we have to redefine the entity. In the most traditional form of the Sorites paradox, the question is which grain of sand makes or unmakes a “heap”? That paradox is usually perceived to be about looseness of definitions (what constitutes “a heap”), but it also crucially raises the question about the nature of identity. Let’s consider two paradoxes that philosophers have identified as they contemplated what to make of Theseus’s ship.

The first paradox is about the continuity of identity. Plutarch noted that Theseus’s ship “became a standing example among the philosophers, for the logical question as to things that grow; one side holding that the ship remained the same, and the other contending that it was not the same.”¹ Was the ship that had no original timbers still the ship of Theseus, or was it now another ship? Given the gradual process of change (what Plutarch calls “growth”), what precisely would make it the same or different? One paradox, then, involves the question of what to make of identity in the face of change.

A second paradox involving Theseus’ ship was formulated by Thomas Hobbes, in his discussion of similarity and difference, and specifically when he addressed the question of how to mark the “beginning of individuation.” Suppose “some man had kept the old planks as they were taken out” of the ship, Hobbes speculates, and then put “them afterwards together in the same order in the shape of the original ship.”² Which is now Theseus’s ship—the one built through replacement planks or the one reconstructed of the original timbers? The question that intrigued Hobbes was the question of whether matter or form determined the beginning of individuation. Hobbes’s paradox is what to make of identity in the face of replication or transference or some other event in which an original identity is now no longer indisputably assigned to one thing.

The question that most modern philosophers have addressed is what to make of personal identity in either of these conditions. People, on the whole, although they may sometimes be wooden, are not constructed, disassembled, and reassembled with planks, and so the question is not the same when we think of continuity of identity in persons. For some, I suppose, human cells may constitute the equivalent of planks, and the idea of cellular change—which, of course, is a process that continues throughout our lives as our cells die and are replaced—might provide an example of Theseus’s Paradox for humans. This problem of continuity in the identity of persons led David Hume to note that there is “no question in philosophy more abstruse than that concerning identity,” and that it would require “recourse to the most profound metaphysics to give a satisfactory answer to it.” Hume’s answer, at the end of the first book of A Treatise, is that personal identity depends on the “three relations of resemblance, contiguity, and causation,” and, at the beginning of the second book, that the idea of “self” is produced by passion (i.e., strong, reflective impressions).³

Contemporary philosophers have devised different ways for us to understand what is at stake in questions concerning continuity in identity. One set of explorations involves questions in cases in which there is a discrepancy between numerical and qualitative identity, unusual cases in which we are asked where we can identify the person when there are two or more seemingly identical bodies and/or consciousness. Derek Parfit proposes several fascinating thought experiments involving teletransportation in which two bodies contain identical physical and psychological states. He also proposes some profoundly intriguing thought experiments in which the brain of one identical twin is transplanted into the body of the other, or each hemispheric lobe of one identical triplet into the bodies of the other two. What “person” would the twin that now has both body and brain be, and where can we place the “person” whose lobes have now been distributed to her two triplet siblings? Bernard Williams likewise offered several fascinating thought experiments in which two persons exchange bodies (including memories and character) and then have to determine which of the bodies (the one that had originally housed them or the new one) they would prefer to receive a reward of $100,000, and which to receive physical torture.⁴ These, then, are questions about how or when or where to affirm personal identity in cases of replication or transference. These are questions in the same form as Hobbes’s: which one of these two is Theseus’s ship?

A second set of questions takes up the issue Plutarch identifies involving “sameness,” or continuity over time. When is something the same and when is it different? What sort or quality or degree of change makes something into something other than it had been? Implicit in this question are two issues: first, whether and how we can affirm a unified life of continuous personal identity, given the flow of time and the process of change; and, second, what to make of events of discontinuity in assessing personal identity. Consider the answers John Locke provided for these two issues.

Locke maintained that only in consciousness can we find a criterion for a continuous sense of self: “in this alone consists personal Identity, i.e. the sameness of a rational Being: And as far as this consciousness can be extended backward to any past Action or Thought, so far reaches the Identity of that Person.” Whatever time may do, it is consciousness that creates the continuity of self. It “is the same self now it was then,” Locke concludes, “and ’tis by the same self with this present one that now reflects on it, that that Action was done.” To remember a performed past action is to affirm the selfsame identity of the person who then performed and now remembers it. Locke then took up the second question about discontinuity. But suppose, Locke muses, someone “wholly loses the memory of some parts” of his life, “beyond a possibility of retrieving them.” What happens to the “self” that is now no longer able to reflect on past actions that the same body performed earlier? Locke’s answer is that because “it is possible for the same Man to have distinct incommunicable consciousness at different times,” it is then possible that “the same Man would at different times make different Persons.” For Locke, that amounted to stating that the person who occupies the later consciousness (the one that forgot the original action) ought not to be punished for the actions of the original consciousness that performed that action. To punish the forgetter for the actions of the actor is the same as punishing “one Twin for what his Brother did . . . because their outsides were so like.”⁵

It was likewise the issue of how to determine when it is just to punish a person who might be construed as not the same person who committed the crime that led Robert Nozick to propose a series of thought experiments and devise a series of scenarios and models for how personal identity may be imagined to change over time in a way that we can say that someone remains the same or becomes a different person. Nozick affirms that there is some way in which we can talk about sameness in identity, which he terms the “closest continuer theory.” In the end, what Nozick offers is a tentative and pragmatic solution to an apparently insoluble problem. The “closest continuer theory” is, as he puts it, “the best Parmenides can do in a Hericlitean world.”⁶ It is, in other words, an answer that allows us to perform social transactions—making promises, abiding in contracts, resolving agreements, assuming or ascribing responsibility—in a situation in which time leaves everything in flux.

Locke, Williams, Nozick, and Parfit all take up the first paradox about Theseus’s ship, and they all recognize that change over time raises challenging questions about the continuity of personal identity. Each proposes what might provide a sense of personhood over time, while acknowledging the possibility that the same individual may be manifest as a different person at different times. They perhaps differ in what that may mean in the pursuit of justice.

Change, then, the idea of alteration simply through the fact that we live in time, is one problem (we will hereafter call that simple change). An additional problem is that we also change by virtue of our acts, behaviors, decisions, and other modes of becoming. Consider the situation in which someone is penitent for a previous behavior. If we believe that “repentance wipes off every crime, especially if attended by an evident reformation of life and manners,” then that act of being penitent alters the person who is so. By this logic, then, when criminals repent some would argue that they “cease to be criminal.”⁷ A significant change, then, not one effected simply by the passage of time (or the gradual replacement of planks), but one effected through a meaningful resolution and alteration in one’s habits and disposition, renders us different than we were before, and the question for justice is whether that difference should matter in the apportioning of punishment. (For future reference, we will refer to this second form as dramatic change.)

This metaphysical paradox raises some important questions about the concept of retribution because it highlights the conflict between retributivists’ belief in the necessity of punishing the guilty and the equal necessity of not punishing the innocent. Consider how this dilemma proved particularly challenging to Kant’s retributivist position. Kant first notes the paradox. An impenitent sinner deserves the punishment meted out to him, but once that sinner undergoes repentance and has become a “new man” then “the punishment cannot be considered appropriate.” The problem, then, is how “satisfaction” has been “rendered to Supreme Justice, in whose sight no one deserving of punishment can go unpunished.” If Supreme Justice requires punishment, and the penitent is no longer the right person to be punished because he is no longer the same person, what can be done? Kant resolves the dilemma by suggesting that the “punishment” is “adequately executed” in the process of repentance itself (the “conversion”). In the act of penance, the “old man” is put to death and the “new man” emerges.⁸ The penitent is no longer deserving of punishment because the process of dramatic change itself has punished the older version of him.

These metaphysical questions about the continuity of identity in a process that involves change are fundamentally at the heart of the topic of forgiveness. How does forgiveness change those who offer or receive it, and what necessary changes have to precede forgiveness as given or accepted? For most commentators, the dilemmas involved in thinking about the changes required and effected in forgiveness lead them to pose metaphysical answers to the metaphysical problems. Let’s consider the standard case, in which the offender apologizes and the offended forgives. Erving Goffman provides an influential and elegant account of what happens when someone apologizes. The individual who apologizes “splits himself into two parts, the part that is guilty of an offense and the part that dissociates itself from the delict and affirms a belief in the offended rule.”⁹ That division—based on temporal change—allows us to understand an apology as effectively saying: “I am sorry for being the person who did that to you; I am now a different person who recognizes the wrong of the person I used to be.” There is a similar split or division in the person who forgives. For most theories of forgiveness, an act of forgiving requires some magical property, what the Greek New Testament calls metanoia; that is, a change of heart. What had been one sentiment becomes another. What the person who is injured had felt, usually resentment and other vindictive passions, is now transformed into some other felt emotion, benevolence, love, something that indicates a forgoing of resentment.

Someone who apologizes, then, becomes a different person by dint of that apology. Likewise, someone who forgives becomes a different person. Both are subjects of dramatic changes. There is still some mystery involved—and I think it is an irreducible mystery—because the person who apologizes must come to the recognition that what he did was wrong, prior to the split, presumably, and the person who forgives must decide that she will forgo the resentment she had felt. The answer that we are split, in other words, does not show us when the change happened—it does not reveal which plank was the decisive one that rendered the ship something other than it was.¹⁰ And I am not sure it is at all possible to show when such change happens. Perhaps an advanced state of neuroscience and the development of equipment able to measure such changes might in the future be able to tell us. This question matters, in any case, only if we wish to affirm that dramatic change is the only way that forgiveness can happen. If a change is dramatic, we might be more pressed to know or need to know what climactic event transforms one identity into another.

As it happens, almost all writers argue that the practice of forgiveness does in fact depend on dramatic change, and that, indeed, any practice that is based on simple change is not forgiveness, even when it resembles it. Perhaps, then, we might apply these metaphysical questions to the three elements in forgiveness—the person forgiven, the event for which forgiveness is granted, and the self who forgives—to see whether dramatic change is indeed the only way for us to understand what forgiveness is.

Meir Dan-Cohen has recently proposed what he argues is a possible solution to the problem of how “to find a conception of change strong enough to support and explain the cessation of negative attitudes toward the offender without being so strong as to disrupt identity.” Penitence, he argues, does a particular kind of work on what he calls “the self’s temporal shape.” When one engages in “revisionary practices,” such as penance, one gives rise to “a new version of self from which the wrongful act is excluded. When this version is inhabited and enacted, it replaces the older one as superior or more authoritative.” As long as “this version is adhered to,” the “misdeed is indeed excluded from the self and does not cast a shadow.” “But the new version does not obliterate the other one,” he warns; “it only supersedes it.”

Dan-Cohen identifies two scenarios that test the limits of the new identity. First, if “one insists, cruelly or obtusely, on unearthing the older version and restoring it, one is not strictly speaking mistaken but merely cruel or obtuse.” So, if either the victim (in an act of accusation) or the wrongdoer (through excessive conscience) insists that the wrongdoer is the same person he used to be, then that insistence makes that person the same. Second, this system does not work, Dan-Cohen insists, “when the wrongdoing looms so large within the offender’s self that removing the wrongdoing would not result in a viable or recognizable version of that self.”¹¹

These are considerable limitations because they place the power to define whether or not change matters in the hands of either a cruel and obtuse perceiver or in the hands of an undefined one. In other words, change becomes solely a matter of perception, which in turn makes penitence a matter whose effect is determined solely by the perceiver. If someone commits a wrong and is sincerely penitent for it, is that person a different person or not? Dan-Cohen seems to want to have it both ways. Yes, that person is different; but if someone says that she is not, then she is not. Noting that whoever ignores the difference is cruel or obtuse does not obviate the fact that we then leave it in the hands of the cruel and obtuse to determine whether change has happened. The second limitation returns us, I think, to the problem we addressed in our comments on The Sunflower and will further discuss in the next chapter regarding the category of the unforgivable. Who can determine that a wrongdoing is the sort that definitively defines a self or not? How do we decide what magnitude of wrong renders an identity static, and beyond change? As is the case with the unforgivable, this is a matter of subjective or popular opinion, in other words, it is a rhetorical category based on the concept of some undefined perceiver.

For us to understand what is at stake in this problem Dan-Cohen addresses, we need to return to an earlier moment in the debate about forgiveness and see how the problem was first formulated. In an important 1973 article in The Proceedings of the Aristotelean Society, Aurel Kolnai wanted to know in what sense we could say whether “forgiveness is possible at all,” given what he calls “the Logical Paradoxy of Forgiveness.” Consider the options we have when we are confronted with a wrongdoer who has committed an injury against us. The first choice is whether to pursue justice or not. If we choose justice, we then seek a way to inflict a punitive injury on our wrongdoer. If we choose not to pursue justice, there are several ways that we can forgo that option. Having outlined what modes of remission are available to us, Kolnai highlights what practices are not forgiveness, but resemble it and are often mistaken for it. We can condone an injury, excuse it, be indifferent to it, forget it, or in some other way allow it to be remitted. Kolnai insists that these are all practices that are not forgiveness precisely because these acts of remission are not acts in which we have a change of heart about the wrong and the person who inflicted it. If we condone an action— if we accept what happened as something that happened—there is no change of heart, which for Kolnai is the only important change that makes something an act of forgiveness. Given these premises, Kolnai then defines the horns of the paradox, using two characters, Ralph the wrongdoer and Fred the injured.

If Ralph is genuinely penitent—and done everything implied in that state, from apologizing, offering restitution, and mending his ways—then what is Fred to do? Can he now forgive? For Kolnai, Fred cannot forgive now because Ralph’s penitence has caused the injury to disappear. Forgiveness, writes Kolnai, “has now lost its ground and raison d’être . . . there is no room for it, seeing that there is nothing to be forgiven.” We cannot then forgive the genuinely penitent.

If Ralph is not penitent and continues injuring Fred or continues in pursuing the same goals or values that produced the original injury, then Fred cannot forgive him because to do so is to condone rather than forgive because the thing that would be forgiven continues to persist. We also then cannot forgive the impenitent.

Here, graphically, is the situation we face. If Ralph hits Fred with a stick once, and then repents sincerely, then Fred cannot forgive him because the event has disappeared. If Ralph hits Fred with a stick repeatedly, and doesn’t repent, then Fred cannot forgive the action because forgiving without evidence of the wrongdoer’s penitence is condonation. Kolnai concludes that, given these premises, “forgiveness is either unjustified or pointless.”¹² It is unjustified without sincere penitence, and pointless with sincere penitence. Later philosophers would take up the questions raised by Kolnai, some using this paradox to argue that forgiveness is impossible (as does Jacques Derrida) and some that it is possible only in light of a cosmic-religious event (as does John Milbank).¹³ In Kolnai’s example, the Ralph who injured Fred and the Ralph who apologized to Fred are sufficiently different that to punish the later Ralph for the crimes of the earlier one would seem unjust. It follows, then, that granting forgiveness to someone who was sufficiently different than the person who committed the wrong is likewise wrongheaded—not unjust, but unnecessary. Change, then, in this case, the change of heart in the wrongdoer, renders forgiveness redundant. We will return to this point.

There is another metaphysical paradox implied in Kolnai’s explication, although he does not explicitly designate or explicate it as such. When he says that “there is nothing to be forgiven” when Ralph is sincerely penitent, presumably he means that the event (the thing) is no longer pertinent. It was significant when it happened, but Ralph’s acts of penance and restitution have lessened its significance to the point of making it disappear. The first metaphysical paradox in forgiveness, then, concerns the status of the person who is to be forgiven; the second paradox involves the event for which forgiveness is sought. Is it, like the penitent soul, changed so much by the act of penance that it is no longer the same thing? Or, to put it in the stronger terms implied in Kolnai’s account, is it no longer even a thing because it is nothing? It seems worthwhile to explore what this claim might mean—when an event (say an immoral act) is transformed through a countervailing moral act into a nonevent, a something into a nothing.

What Kolnai is saying, then, is that one act (penance) can “undo” an earlier act (the wrongdoing) and render it null. Many contemporary philosophers have persistently argued that this claim is indeed viable, and valuable. The term many employ is “undo”: forgiveness, many of them argue, can undo what has been done. What they probably mean by that odd usage is that forgiveness can make us forgo the effects of the harm. If Fred feels resentment at Ralph for what he did to him, and then forgives him, and his forgiveness is as full and sincere as Ralph’s penitence, then Fred frees himself from the vindictive passion. His resentment has been undone. It occupied him for a while, and now it has dissipated. Dan-Cohen states it pithily when he suggests that we can imagine a “contraction in the temporal scope” to be meaningful when we change “our attitudes” and “the significance we attach to past events.”¹⁴

The philosopher who has been most influential and oft cited on this question has been Hannah Arendt, who wrote in her 1958 book, The Human Condition, that forgiving “serves to undo the deeds of the past.” Faced with a world in which time flows in only one direction, what she calls “the predicament of irreversibility,” and faced with the laws of cause and effect, action and reaction, humans confront a world in which one act can determine a whole historical sequence of reactions. Say that Fred decided not to forgive Ralph for hitting him with a stick, but instead took up his own stick and beat Ralph with it. Ralph then retaliates, Fred reacts, and eventually one kills the other. The next generation, Ralphson and Freddaughter, then takes up the battle, with renewed vigor; as the feud expands, the community descends into a vicious cycle of vengeance. An act of forgiveness prevents that development. But even in less dramatic scenarios, we can imagine how an act of forgiveness frees us from the self-loathing that we might feel at having hurt another. If we are sensitive enough, our single act of having hurt someone haunts us and, in a way, imprisons us. In Arendt’s terms, if we are not “released from the consequences of what we have done, our capacity to act would, as it were, be confined to one single deed from which we could never recover.”¹⁵ The rest of our lives would be reactive, not creative. For Arendt, forgiving as “undoing,” is liberating because it clears the ground for new action (without which the human condition becomes inhumane). Arendt’s forgiving, then, is not a mystical conception of undoing that requires a miracle of some kind that would seem to necessitate a reversal of time.

We cannot confidently say that of other contemporary philosophers, like Vladimir Jankélévitch, who maintains in his 1967 book, Le Pardon, that by “the grace of forgiveness, the thing that has been done has not been done” (Par la grâce du pardon la chose qui a été faite n’a pas été faite). Jankélévitch is quite unclear on this point because he wavers between asserting that what is done cannot be undone and affirming that what is done can be undone. In the end, he does not come to a resolution. Nor does he use “undone” in the psychological sense we have noted above—that what is undone is Fred’s resentment. He means it literally. What has been done has been undone, and forgiveness is able to accomplish this miracle because forgiveness, as he writes, “cuts the temporal continuation.”¹⁶ There is no “predicament of irreversibility” in this world as there had been in Arendt’s, precisely because time is discontinuous. There are other philosophers who share with Jankélévitch this same mystical conception of time and likewise affirm that forgiveness, as Emmanuel Levinas writes, has “the power to efface, to absolve, to undo history,” or, as John Chryssavgis puts it, that through forgiveness “the past can be undone,” or in Milbank’s words, that in “a certain way we can go back in time and redo the things that still darkly echo in the present.”¹⁷

We have, then, a range of different philosophical positions about how forgiving can undo the past. In the first, it is the psychological effects of the hurt that are undone (the folk wisdom of “forgive and forget” captures this sensibility). Forgiveness permits us to accept those who hurt us back into our lives. In the second, that is, Arendt’s, it is the political effects of the harm that are undone. Forgiveness allows us to escape the doom of having all our actions determined by one original act, and instead permits us to pursue novel actions and beginnings. And in the third, the mystical, it is the event itself that is undone. It is properly the third that is metaphysical, in both the colloquial sense of that term, that is, unmoored from physical and temporal restraints, and in the philosophical sense, of inquiry into the ontological or existential status of being.

Let’s now return to Kolnai. Whatever these philosophers just discussed may imagine is involved in the “undoing” of the forgiven event, they all conceive of the undoing as the result of the forgiveness. That is not what Kolnai claims. It is important to note that Kolnai’s account assumes the change to take place not when forgiveness is granted (forgiveness serves to undo the deeds of the past), but before it is granted (the deed is undone prior to its being forgiven). It is not the forgiven event that has disappeared, but the event that is now no longer necessary or possible to forgive because it has disappeared. This is a different kind of undoing. Now, according to Kolnai’s account, it is penance that undoes the past deed that in these other philosophers’ models forgiveness had undone. We can first ask a basic question: why or how does penance do the work of forgiveness when it involves only the penitent? The answer must be that it is self-forgiveness. The penitent soul forgives herself and therefore does not require the forgiveness of the victim (which is now, as Kolnai says, “pointless”). If Kolnai objects and claims that the penance alone—without any self-forgiveness—is what undoes the event, then our follow-up question would be: what does Kolnai mean by “penance” because penance is explicitly and has historically been a beseeching for forgiveness? If asking for forgiveness is the only thing required to undo the yet-to-be-forgiven event—without forgiveness given by the victim or the penitent—then we would have to affirm that for Kolnai forgiveness is not “pointless.” It is unimportant. In that case, there is no paradox, just a dismissal.¹⁸

In any case, we have seen the metaphysical questions involved in asking about two categories of the forgiveness scenario. First, what can we make of the changed identity of the wrongdoer whose penitence might be described as having changed them sufficiently so that they cannot be described as the same person who committed the wrongdoing? Can we forgive someone who is arguably not the same person who hurt us? Second, what can we make of a potentially dissolved wrong deed for which we ask for or are granted forgiveness? In the non-mystical models previously cited in which forgiveness undoes the psychological or political damage of the past deed, the deed disappears from the field in which future action takes place. In Kolnai’s account, which either smuggles in forgiveness or is mystical in a different way, it disappears before it is forgiven. What is the existential status of an undone event? In the non-mystical models, where forgiveness undoes the past deed, there is a residue in that dissolution. The past event is not dwelt on nor recalled, and it does not play an undue role in future relations. What makes it undone is an unspoken agreement that it will be undone and will remain unimportant in the constitution of that relationship. We have explored, then, the metaphysical questions about the nature and meaning of change in two of the categories—the forgiven person and the forgiven event. We can now turn to the third.

The third problem in this metaphysical tangle is the status of the forgiver. Why does someone forgive, or change from injured to forgiver? What alterations does that person go through to move from feeling resentment to forgoing resentment (and, again, resentment here stands in for the range of reactive attitudes)? That precise change—what happens to permit forgiveness to happen in most models, except perhaps the debt-cancelling one—is crucial because it determines, for most commentators, whether what happened was forgiveness or not. One kind of change constitutes forgiveness, while another does not. The change must be meaningful in a specific way, not only dramatic (based on a resolution, say) but also done for moral reasons. It is worth examining what that model of forgiveness entails and what is presupposed in it. Let’s consider what most writers posit as the two most obvious forms of what they argue appear to resemble some kind of faux forgiveness but are not forgiveness at all: condonation and forgetting. If the victim condones or simply forgets the injury, then forgiveness, they argue, has not occurred.

As Macalster Bell put it in a recent article, any “adequate philosophical account of forgiveness” will “provide a basis for distinguishing between meritorious forgiveness and objectional condonation.” That is a necessary condition. Another, Bell adds, is that the account provides “some basis for distinguishing between forgiveness and other ways of going on after wrongdoing, such as . . . forgetting the wrongdoing.”¹⁹ Forgetting and condoning have largely become the established concepts that help us identify just when forgiveness fails, the conditions that make it possible to identify what we can then classify as deluded or misfired acts of forgiveness. Let’s examine more deeply just what is assumed about condonation and forgetting that makes them not forgiveness.

It is worth noting, at the outset, that “condonation” did not always denote an objectionable means of forgoing retribution or acquiescing in our degradation. As Glen Pettigrove reminds us, condonation used to mean something considerably closer to forgiveness. The Oxford English Dictionary defines “condonation” in one usage as “the pardoning and remission of an offence or fault,” and cites a passage from 1858: “Condonation meant a blotting-out of the offence imputed, so as to restore the offending party to the position which she occupied before the offence was committed.” Dr. Samuel Johnson defined it in 1755 as “a pardoning, a forgiving.”²⁰ To condone, then, was indistinguishable from to forgive. Now, however, for most of the writers who mark the distinction, what has occurred instead of forgiveness is that the victim has either developed or discovered an absence of self-esteem. They have accepted that they are the type of person who can be injured in this way and accepted that injury without demanding retribution or apology or any other form of restitution. But is that necessarily the only or best way to understand condonation? I would argue that it is not.

We should note that condonation is a tricky concept, and a judgmental one. It would be odd to hear anyone say, “I condone what you have done.” It is rather what we say about another person: “They have condoned that behavior.” And so condoning is what we think other people are doing, but rarely what people think of themselves as doing. And the judgment is often unsparing. “Condonation is thus virtually ‘conniving’ and immoralistic,” Kolnai writes, as well as being “undignified and self-soiling,” and carries with it, he concludes, an “innuendo of spineless accompliceship.”²¹

Like other acts of judgment, the accusation of condonation is based on appearances. In some cases, no doubt, those judgments (without the cruel comments) may be correct and what appears to be is in fact the case. Some assuredly do lack self-esteem (or, more likely, adequate power) and are resigned to accepting injuries at the hands of those who do have the power to hurt them. But in other cases, those appearances are likely deceptive.²² There are, just as assuredly, those who may accept what happens as part of their conscious decision to exist uncomplainingly in the world, or feel a high degree of tolerance toward the behavior of others, or exist in some other frankly spiritual mode in which they neither rail against the events of the world, nor demand restitution for them, but either stoically or transcendentally allow them to pass through their compassionate lives. Admittedly, that is probably not the attitude of many who condone wrongs committed against them, but it is the attitude of some, and what that attitude says, I believe, challenges our conceptions of forgiveness, if our conceptions are too narrowly focused.

That spiritual, even stoical, attitude can be characterized as a forgiving one, but not forgiving in the sense most people usually reserve for the act of forgiveness. It is an attitude in which someone assumes a benevolent and patient stance in advance of the wrongs that they expect because of the kind of world in which we live. Robert C. Roberts has coined the term “forgivingness” to describe this virtuous “disposition to abort one’s anger (or altogether to miss getting angry) at persons one takes to have wronged one culpably, by seeing them in the benevolent terms provided by reasons characteristic of forgiving.”²³ While Roberts sees “benevolence” as the background attitude behind forgivingness, Margaret Holmgren has recently identified “humility” as another virtue enabling us to forgive unconditionally and almost in advance of the harm done us. Genuine humility, Holmgren writes, “will prevent us from adopting a judgmental perspective toward the offender” because the humble soul recognizes that such judgment reveals that we “attach too much importance and status to ourselves” and “fail to recognize what is of great importance outside of ourselves.” Instead, the virtuously humble will regard the wrongdoer “with respect, compassion, and goodwill, thereby adopting toward him an attitude of unconditional genuine forgiveness.” Christine Swanton has suggested that the virtue of universal love is manifest in “an ongoing preparedness to forgive all who have wronged one.”²⁴ What such a stance has forgiven, we can say, is the human condition, and it has forgiven that condition in advance of the future wrongs that will be committed. And that, for most writers on forgiveness, is just what makes it not forgiveness, as for them, forgiveness cannot be proleptic or future-oriented, nor can it be general. It must be retrospective or past-oriented, and specific. We will return to this point shortly.

Likewise, in the case of forgetting, those writers argue that even though the resentment the injury caused has abated or disappeared, forgiveness has emphatically not happened. Writers who insist that forgetting is not forgiving—even though their minimum threshold definition of forgiving has been met in the victim’s forgoing of resentment—argue that the victim must let go of resentment in a particular way or with a specific kind of attitude. Forgetting, they argue, is not a moral act, while forgiving is. And so, the act of forgoing resentment must be conscious. To employ the terms we introduced earlier, for them the change cannot be simple but must be dramatic to be meaningful. Forgetting is a product of simple change; time passes and we are occupied with other things and forget our resentment (unless something provokes us and reanimates it). Forgiving is a process of dramatic change; we affirm that we are forgoing resentment in a conscious and concerted way.

These contested forms of releasing resentment (by condoning or forgetting the injury) raise some significant questions about forgiveness. Let’s start with condonation. To condone a wrong someone committed against us, Kolnai writes, is to be “clearly aware of the wrongdoing” and to disapprove of it, but “deliberately” to refrain “from any retributive response to it.” Someone who condones does not overlook an offence; she “acquiesces in the offence.” Paul Hughes refines this definition by distinguishing among various kinds of condoning, between “condoning as accepting,” in which one overlooks and thereby approves of the wrongdoing, for instance, and “condoning as tolerating,” in which one overlooks while nonetheless disapproving of the wrongdoing. In the end, Hughes defines four ways of condoning wrongdoing and identifies one of these as “potentially indistinguishable from forgiving.” When one overcomes “personal anger” toward the wrongdoing, he concludes, what one does (condone the wrong, but also overcome the anger it raised), then the condonation is almost the same as forgiveness. While seeming to offer some latitude, though, Hughes concludes by stating that “where forgiveness is synonymous with condoning, it is unvirtuous forgiving insofar as it is done for a reason incompatible with the demands of morality.”²⁵ Condoning can sometimes resemble forgiving, he suggests, but they are morally distinct phenomena in the end. For Hughes, the reason that condoning cannot be forgiving is that forgiveness, for him, requires the injured party to forgo moral anger, and that this forgoing of moral anger must be a resolute act to a specific event.

If we think of forgiveness as having only form, constituting one practice, and operating always in one way, then we would have to agree that condonation (in its modern definition) cannot be forgiveness. If, for instance, we believe with Richard Swinburne that forgiving works as a “performative utterance” in which forgiveness consists in saying “I forgive you” or making other expressions or gestures understood by both the forgiver and the wrongdoer as forgiveness in that context, then, we understand why condonation cannot be forgiveness. We do not usually articulate our condonation, and what constitutes forgiveness in this model is the performative utterance. For that reason, Swinburne defines condonation as treating “the act as too trivial to need forgiveness.” Likewise, if our focus like Brandon Warmke’s (at this instance in his exploration of the multiform practice of forgiveness) is on “the ways in which forgiveness alters norms” both for the forgiver who releases the wrongdoer from what were previously permissible modes of regard and treatment, and the forgiven who is released from them, then we can see that implicitly there has to be a statement of some sort that communicates the forgiveness that now changes the norms. For that reason, Warmke adds, these norms cannot be altered to “something resembling an ex ante state” by “forgetting about what happened or by condoning that behavior.”²⁶ In these cases, then, where the focus is on interactions between forgiver and wrongdoer, and in which forgiveness is either performative or an agent of normative change, condonation has no place.

If, however, we agree with Warmke that forgiveness “is a diverse and diffuse practice, admitting of a multitude of modes or ways of forgiving,” then we can indeed see condonation as a form of forgiveness when we focus on other kinds of interactions, particularly the interactions within the forgiver herself.²⁷ We will see that it is not so much the case that what is forgiven by condonation are not acts “too trivial to need forgiveness.” They are acts that do not warrant the sort of resolute forgiveness that requires performative utterances. That does not mean they do not need forgiveness, but that they do not need that kind of forgiveness we offer on different occasions to qualitatively different kinds of wrongdoing.

Let’s turn to a biblical representation of forgiving that seems to disagree with the accounts that maintain that acts of forgiving are distinct from the virtue of forgivingness, and that only resolute acts can count as forgiveness. When Peter approached Jesus and asked him, “Lord, how oft shall my brother sin against me, and I forgive him? Till seven times?” Jesus responded: “I say not unto thee, Until seven times: but, Until seventy times seven” (Matthew 18:21–22). Jesus seems to be implying that one should always forgive—or, for the literal-minded, forgive each person 490 times.²⁸ Yet some have read this injunction as urging something other than forgiveness at all. David Konstan argues that because Jesus does not specify what attitude the offender should assume (i.e., be penitent and offer atonement), what Jesus is actually urging Peter to do does not properly constitute forgiveness, but rather that Peter adopt “a posture of general charitableness.”²⁹ What Konstan describes as a “general charitableness” is what other philosophers have followed Roberts in calling “forgivingness,” or what I have described as that spiritual or stoical attitude of accepting what the world offers because one has forgiven the human condition and the future acts that those who have fallen into it will commit.

Why, then, is it held that a general and proleptic act of charitableness cannot be thought of as forgiveness? Consider the kinds of reasoning that often precedes and makes possible a single act of forgiving; most scholars identify two in particular. People think that they will forgive because they, too, are flawed beings in need of forgiveness, or that the wrongdoer deserves to be seen in a larger context that evaluates his one wrong against other beneficent deeds or behaviors.³⁰ The first looks at the self in its depth, and the second looks at the other in its fullness. The first discovers a self that should not hold too exalted a sense of being aggrieved because it, too, will assuredly fail on occasion, and the second affirms an other whose overall life has been largely benign and whose singular failing does not significantly change that. What these reasons exhibit, then, is precisely an attitude of general charitableness; that is, they accept the human condition for what it is.

If the motive that leads people to forgive is general charitableness, what precisely is gained from distinguishing the singular act from that motive?³¹ It would seem to require us to insist on what we can describe as a punctuated sense of history. Forgiveness becomes a precise process with a precise schematic. A wrong is identified as a wrong, resented, and then atoned for, and finally forgiven. The stages are distinctly marked and mapped. I am not sure that is how forgiveness does work in our world, at least that is not my experience of it. I suspect that on most occasions we forgive without recognizing our resolution to do so, frequently without waiting for an apology, and often without offering a ritual declaration of forgiveness. Our lives are not punctuated that way, and the argument that we are not forgiving if we don’t recognize the distinctness of those acts seems to me wrongheaded. What that insistence demands is that each act of forgiveness be resolute; that is, it is constituted by a recognizable resolution that I am forgiving someone for some specific thing. It assumes that forgiveness requires what Avishai Margalit calls “a policy of adopting an exclusionary reason with regard to someone who has wronged us,” and that forgiveness is or must be, as Kolnai argues, a “consciously decisional act.”³²

Two objections could be made to this understanding of resolute forgiveness. First, we know that not all resolutions succeed. There are cases where someone resolves to forgive only to discover later that their resentment was greater than they thought and does not permit them to act toward their wrongdoer in the way we would expect from someone who has forgiven. In other words, the resolution (or reason or decision) in such cases did not actually express forgiveness (unless we hold that a resolution or decision to forgive is a contract that cannot be revoked). Second, we know that forgiveness is not always expressed in such recognized resolutions. We frequently feel a sharp pain of resentment at a wrong, and then quickly, without thinking too much about it, forgo our brief pang of resentment. There might be, at an unconscious or subconscious level, some kind of resolution—some internal marking of the shift from one emotional state to the next—but it is not always discernible. These two objections, then, show us that if we insist on resolution as the marking of what constitutes forgiveness, we need to have an extraordinary faith in the human capacity for self-knowledge to believe that we can determine and be fully aware of what we feel or when we feel it.

Let’s consider now the case of forgetting as being emphatically not forgiving. Hagit Benjabi and David Heyd represent what seems to be a consensus among almost all philosophers who write about forgiveness by stating it explicitly: “Forgetting what you have done to me does not mean that I have forgiven you.” Forgetting is not forgiving because forgiving, like genuine toleration, they write, requires some kind of ‘restraint,” must have a “price,” and requires the exertion of “some effort.” Forgiving and tolerating are “intentional attitudes rather than mere psychological dispositions.” In other words, they conclude, unlike forgetting, forgiving is not an “automatic or ‘natural’” response, but requires “some sort of deliberation, a mental process guided by reasoning and decision.” Kolnai affirmed the same thing earlier, referring to the process of forgetting as “emotional prescription.” We cannot be said to forgive, he writes, if what happens is that after “‘a long lapse of time’ . . . the wound ceases to smart and the original indignation and retributive attitude come to be toned down and finally to vanish, fading away without any (even purely mental) explicit act of forgiveness.” Forgiving is not only “not ‘forgetting,’” Kolnai asserts; it is, in fact, “incompatible with forgetting.”³³ The objection to forgetting, then, is that it is a “natural” process rather than a rational one, and what distinguishes it from forgiving is that it is not a “decision” or “explicit,” even as a purely mental act. It is simply the fading away of something (the memory of the injury) rather than the affirmation of something else (the decision to forgive). It is, in other words, a simple change, and not a dramatic one.

If we grant these premises, and argue that forgiveness does require that dramatic change, that conscious decision, that resolution, we might then inquire when forgiveness happens. Does forgiveness happen before, during, or after the resolution? In other words, do we resolve that we forgive because we realize that we are no longer resentful, or do we forgo our resentment as we resolve our forgiveness, or do we resolve our forgiveness because the resentment has lessened to some threshold level we find acceptable, after which we work to rid ourselves of the residue of it? Forgiveness, then, can be expressed, constituted by, or preceded by a resolution. When it is expressed, forgiveness can be thought of as an anterior process that the resolution brings to light, which is largely what the emotional model of forgiveness holds. When it is constituted by the resolution, forgiveness can be thought of as a genie that requires our naming it to bring it to light, which is what speech-act theorists who affirm the debt-cancelling model hold. When it is preceded by the resolution, forgiveness can be thought of as a contract or promissory note that will be fulfilled sometime in the future, which is what is held by those who argue for a hybrid model involving change in emotions and attitudes to constitute forgiveness.

Some will no doubt object that forgiveness is not a resolution. It is a process, like other psychological processes, like grieving, for instance, which proceeds through stages. To think of it as a resolution is to cast it into a contractual or rational form that does not belong to it, and to which it does not belong. That, of course, is my point too. What I hope this analysis demonstrates is that we are talking about a nebulous process that cannot be marked by stages, at least not schematically. But I think that those who insist on denying that condoning or forgetting is not forgiving—and that is the vast majority who write on forgiveness—must be committed to the unacknowledged presupposition that forgiveness is a resolution. If it is not a resolution, then forgiveness can take the form of general charitableness (where forgiving is not formalized) or forgetting (where forgiving is not even consciously recognized).

So, can condoning and forgetting be acts of forgiveness? The answer, of course, depends on what definition one has for forgiving. My point is not to assert that they are, or should be, but that they sometimes do precisely the same work as do the formal acts of forgiveness. General charitableness can be described as an informal mode of forgiving. While we are not often aware enough of our inner emotional states, we are, I think, sufficiently aware of some aspiration to be compassionate beings who are charitable to others because we recognize ourselves as needing an equal charity. It makes sense for us to characterize that sensibility as the forgiveness felt toward the human condition generally, and then permeating and conditioning our responses to each specific injury we meet. It is informal—as opposed to the kind of formal forgiveness that requires an explicit declaration of resolution—because many of us simply do not feel formal declarations are always appropriate or timely. They might be on certain occasions, but probably not for most.

Likewise, I think forgetting can be described as a casual mode of forgiving. We forget an injury because it is not serious enough to warrant our remembering it; that does not mean that it does not warrant our forgiving it. It means, I think, that we forgive it with less emotional investment than we do greater injuries. Of course, there are cases where we forget an injury without forgiving it, and we feel all the original resentment when we are reminded of it. But there are also cases where we do not feel any resentment at the memory because we recognize that we have already forgiven it. That recognition is belated, because the forgiveness itself has happened beneath our notice. (And, again, if we feel that we have consciously to notice forgiveness for it to be such, then we are back at the resolution model of forgiveness). So, condoning might be forgiving with minimal notice and forgetting with almost none.

Let me address one more set of objections pertinent only to the model of forgetting as forgiving. Forgetting, most believe, is best thought of as “a natural” act, as Benjabi and Heyd claim. We forget because our intellectual equipment can hold only so much, and we dismiss things of lesser importance to make room for those that have more. That is a result of our limited neurological capacities. Forgiving, on the other hand, is a moral act, and, for these critics, that means we need to distinguish it from natural ones. Moreover, forgetting seems not a moral act because it is what Hughes calls a “completely passive phenomenon.” Whereas “forgiveness is active,” Dana Nelkin argues, “forgetting is something that happens to you.” It does not require the kind of “effort” presupposed in forgiveness as commonly conceived, and in fact can seem precisely as a failure of effort.³⁴ Let’s consider these objections.

First, we can address the question of forgetting as “natural,” and therefore not forgiveness, under the assumption that a natural act must differ from a moral one. It is not entirely clear that forgetting is always a natural phenomenon and not part of a moral process. We do, after all, believe that the best form of forgiveness requires us to “forget” the wrongdoing after we forgive it to make as full a reconciliation with our wrongdoer as possible. We urge ourselves and our children to learn to forget certain things to behave more morally toward others. So, forgetting in this case is part of an overall moral activity, in which what seems like a natural act (forgetting) enhances and makes more fully meaningful the moral act (forgiving). Of course, the objection above is not about what happens after forgiving, but what constitutes it, and forgetting, for these philosophers, cannot constitute it because it is a natural phenomenon.

We don’t have to occupy a hard determinist position to be skeptical about too firm a distinction between what is a natural act and what is a moral one. In an essay in which she challenges precisely that kind of hard determinism, Iris Murdoch makes the case for what she calls “a kind of inchoate non-dogmatic naturalism” that, as I read her, does not depend on the strict, dogmatic distinction between an act that is natural and one that is moral, between something that happens at a level we cannot always control and something that has to be resolutely within our conscious control.³⁵ In cases of immoral acts, it seems to make sense because we have ethical and legal systems that require our defining what level of responsibility an agent can be ascribed. Not all ethical systems have made that distinction, of course, and Susan Neiman’s terrific book, Evil in Modern Thought, shows us what sorts of contortions and contradictions emerged as thinkers attempted to make that distinction between natural evils and moral evils. The events that most of us now see as wholly belonging to one or the other of those categories, the Lisbon earthquake as a natural evil, Auschwitz as a moral evil, were not always, or even now, so understood exclusively and finally to belong to one or the other.³⁶

Forgetting, then, might be a natural process, even though it can play a role in our moral activity, and even though the argument that there are unbreachable distinctions between a natural and moral act is both a historical and contingent one.

Second, we can address the question of passivity. It is important to state at the outset what I am not saying. I am not saying that all acts of forgetting an injury are forgiving; some might be, and some might not (which is exactly the case with a resolution or declaration of forgiveness). But some kinds of forgetting are compatible with, and indicate, forgiveness. I do say that those episodes of forgetting can be construed as forgiving, and they can be valued as moral acts. They are moral acts that fall beneath our level of conscious awareness. When we remember an injury that we had long not thought about—forgotten, in other words—and realize that we no longer feel resentment at the person who inflicted it on us, we recognize at that moment that we have already forgiven that person at some point in time. We could, of course, say that the moment at which we remember and become conscious of it is the moment of forgiveness (if we must punctuate our mental life in that way); but that would be inaccurate. It is more accurate for us to admit that the forgiveness happened earlier and in some deeper recess and did not require a punctuated effort of which we had to be conscious or cost us the price or require the exertion that the resolute model holds as necessary.

What it might have required instead is something resembling what Murdoch describes as “attention.” The “moral life,” as Murdoch argued, “is something that goes on continually, not something that is switched off in between the occurrence of explicit moral choices.” It is not the choice or effort or resolve to forgive that constitutes forgiveness, but rather an ongoing and relentless form of attention that in fact constitutes the foundation of our moments of moral choices. If we “consider what the work of attention is like,” Murdoch notes, “how continuously it goes on, and how imperceptibly it builds up structures of value round about us, we shall not be surprised that at crucial moments of choice most of the business of choosing is already over.” It is that kind of attention—which involves recognizing and appreciating the meaning and value of the kind of general charitableness described earlier—that can make an act of forgiveness happen without our noticing it until later. What might be equally “imperceptible” as the work of attention’s building up a structure of values that undergirds our choices are some of the choices it makes along the way. We make moral choices periodically; but those are not the only moments when we are doing moral work. We are as, Murdoch beautifully phrases it, “morally active in the interim.” A “decision,” she insists, “does not turn out to be . . . an introspectable movement.”³⁷ To be morally active, then, and even to make decisions, is not something of which we are or can be always conscious; sometimes, such activity is hazy, sometimes nebulous, and sometimes seems like forgetting. It might seem odd, even paradoxical, to say that attention is not something of which we’re always aware, which is not in the end what I am saying. What I think is more accurate is to say that it is not always conscious of its work. That is the kind of attention I think Murdoch is describing here, and the one for which I am arguing, one which is incessant in a way our conscious minds are not.

We can return, now, to the metaphysical question with which we began; that is, the question about how change over time affects the continuity of identity. We have seen how a certain definition of what forgiveness is not can be challenged and shown to be limited and limiting. What I want to see now is what that definition of forgiveness—based on the denial that condoning and forgetting are practices of forgiving—reveals about the answer to the metaphysical question. It seems to me that what these models of forgiveness insist on is precisely the denial of change. Let me take up two points in particular.

First, in arguing that forgetting and condoning are not acts of forgiving, this model of forgiveness posits that only a resolute and dramatic change—not a simple change—can effectively forgive. If we exhibit general charitableness toward or forget an injury, we permit simple change to do the work; that is, the prior forgiving of the human condition or the ceaseless flow of time is what dissipates our resentment. Second, in arguing that forgiveness serves to undo the wrongdoing, this model posits that the dramatic change alters the past event; that is, forgiveness makes it disappear in some way and, in the words of one commentator, restores “an antecedent moral order.”³⁸  One presupposition of Theseus’s Paradox is that time changes everything; repairs have to be made and planks replaced because time simply flows and wears down whatever exists in time. What this model of forgiveness insists on, however, is that only one kind of change has moral value, and, arguably, it possesses that moral value precisely because it does not accept the concept of ceaselessly flowing time. Those philosophers who deny that forgetting is forgiveness wish to pre-empt time, and those who believe that an act of forgiveness undoes the past event wish to reverse it.

The reason, I think, that they fall into this metaphysical trap is that even the most devoutly secular of these writers continues, implicitly and, I would argue, unknowingly, to draw on a model of forgiveness and a sense of time that are religious. Let’s return to Kolnai’s character, Ralph the wrongdoer, to discover what is at stake. If Ralph is penitent, then, in Kolnai’s account, Ralph is no longer the same person who committed the injury. But how is he different? There are at least two ways we can think about his new identity. In the first, he is different because he has been reformed; he recognizes that what he did was wrong, corrected that wrong, and therefore stands now in no need of forgiveness (according to Kolnai) because the wrong he did is undone; that is, “annulled” or “nothing.” If the event is undone, then logically he is re-formed as the same Ralph he was before he committed the injury. In the second, he is different because he has been transformed; in recognizing his wrong and correcting it, he has gained something that makes him different. He has gained a knowledge of what constitutes a wrong (if he didn’t already know that), and the knowledge that he is capable of committing that wrong, and capable of feeling guilt for committing it, and capable of expressing penance for it. What makes Ralph different, in this second case, is that he is not now reformed (the same old Ralph) but transformed (a new wiser Ralph), because his identity has now been augmented with all his newfound knowledge.

We might add that in this second case, perhaps this Ralph does require forgiveness in a way that Kolnai maintained the first one did not. The first one did not because there is no “thing” to forgive (it has been undone by the penance), while the second Ralph might benefit from forgiveness because that act would complete the knowledge that he has accumulated. It would inform him that his penance was acceptable. Let me put this polemically in the terms employed by those who define the paradox of forgiveness: without that forgiveness from Fred, it is Ralph who cannot know whether he is condoning his own action.

I imagine that most people would intuitively consider it preferable that Ralph be transformed than reformed because we wish for him to be wiser and believe that it is less likely that he will commit another injury if he is so. But, curiously, a great deal of the writing on forgiveness seems to favor the idea of reformation instead of transformation. That, I suspect, is why we find these mystical, metaphysical assertions that with an act of forgiveness the wrong is undone. When Lucy Allais, for instance, declares that a perpetrator “has undone the harm through apology, repentance, penance, and restitution,” she is describing a sense of the past that can be erased. The “heart of forgiveness,” she notes in her title, is “wiping the slate clean.” Nicholas Tavuchis likewise notes that although the laws of physics are what they are, the laws of morality can somehow evade them: “we cannot undo what has been done,” he writes, “only erase it by seeking forgiveness.”³⁹ These are examples of the reformation model; that is, they hold that the past can be expunged and we can start with a new slate. The transformation model holds that the past, if we think of it as a text, is not one that can be erased, but it can be re-inscribed. It is, in other words, a palimpsest.

This desire for reformation instead of transformation is likely the product of a particular religious heritage. In most religions that make redemption part of their cosmic history, the idea of redemption is appealing because it is complete. Sin and all the consequences of sin, including death, are eradicated. What Adam brought into the world through his Fall, Christ will take out of it through his Redemption. In the eschaton, we will be fully renovated, pure, innocent, new beings, not ones who continue to learn the lesson of what it is to sin and repent. That model is considerably more desirable than its alternative. A religion that promoted the idea that we would be partially redeemed, remain conscious of our failings, and be eternally guarded in our actions because we recognize our capacity for doing what we should not do, but have done before, is likely to be much less attractive. Who needs a redemption that sounds more like parole when we can have a redemption that is fuller than a pardon? The problem, though, is that, given those implicit beliefs, we expect forgiveness on this plane—in this world, with our physical sense of time, and our peculiar emotional and cognitive equipment—to do the work that forgiveness properly performs in a very different plane, and by a very different kind of agent. As I have noted elsewhere in tracing the Pauline origin of this idea that humans can forgive like God, such a belief is at the root of much of the incoherence of what forgiveness is and does.⁴⁰

The issue, then, concerns consequences—what it means to inhabit a life in which we must accept them, or what it means that we desire a life in which they are erased. Consider in this light the ending of Macbeth, in which Shakespeare plays with the very idea about undoing deeds, appropriately, in a play about the consequences of a pricked conscience. Lady Macbeth’s almost final words are “What’s done cannot be undone” (V.1.67).⁴¹ This final sentiment of hers—the only other words she utters after them are “To bed, to bed, to bed”—returns us to the topic of her and her husband’s meditation earlier in the play, when each of them had debated just what it means to act. Macbeth, for instance, is troubled by the physics of action, by the fact that all actions have effects and reactions. As he puts it, he would not be troubled by performing an undesirable task—killing the king—if only “th’ assassination / Could trammel up the consequence.” But in this world where things are irreversible—“here upon this bank and shoal of time”—all such acts have effects. Macbeth drily notes this point in a pun explaining his dilatoriness in acting: “If it were done [finished] when ‘tis done [performed],” then he would do it gladly, but he recognizes that as a fond and false hope (I.7.1–6). Lady Macbeth’s final words, then, remind us of what Macbeth knew but denied: that consequences matter (they cannot be trammeled up), and that what is done (performed) is never done (finished) because those consequences continue to produce further consequences, that, in Lady Macbeth’s word, what is done cannot be undone.

Immediately after Lady Macbeth utters that stark reminder of what is a law of physics as much as it is a law of morals, the Doctor who is witnessing her sleep-walking confession comments that what she needs is not physical medicine, but spiritual: “More needs she the divine than the physician.” He completes that thought with a brief prayer: “God, God forgive us all” (V.1.73–74). The Doctor realizes that it is only this divine forgiveness that can heal Lady Macbeth, or anyone troubled by an embroiled conscience; that is, that only spiritual succor can provide a healing salve. But it is also only divine forgiveness that can perform the miracle for which we have seen some philosophers of forgiveness argue, that something done can be undone. When Macbeth talks to the Doctor about healing his wife, he speaks in precisely those terms:

    Cure her of that.

Canst thou not minister to a mind diseased,

Pluck from the memory a rooted sorrow,

Raze out the written troubles of the brain,

And with some sweet oblivious antidote

Cleanse the stuffed bosom of that perilous stuff

Which weighs upon the heart? (V.3.42–47)

What Macbeth wishes for his Lady is just what she has recognized in her somnolent state is impossible—to undo what has been done. He would have her forget her act (plucked from her memory) or have that act effaced from her mind (razed from her brain). If the act itself cannot be undone, the only alternative is to have its agent forget that it was ever performed. What Macbeth wants is Lady Macbeth reformed.

What Theseus’s Paradox teaches us, I think, among other lessons, is that change—simple or dramatic, plank-by-plank or wholesale lumber—is virtually impossible to catalogue. And what that fact of the difficulty of ascertaining when change happens should teach us, among other lessons, is that we need to be modest in assessing what change can mean, and even more modest in denying that certain kinds of changes are not meaningful. Forgiveness, as we’ve seen in our survey of models and methods, is not one thing, and it can take many forms. There are acts of formal forgiveness, acts of informal forgiveness, and acts of casual forgiveness (and, were one pressed to it, maybe acts of business attire forgiveness). They are acts that take different forms, and probably express different qualities of change. What that suggests is that we need a more refined vocabulary rather than a restrictive one, and a greater appreciation for the range of ways in which resentment can be dissipated, attitudes toward wrongdoers adjusted, moral anger dissolved, and even the forgoing of moral indebtedness communicated.