A Homily For Easter

I am trained primarily as a philosopher, not (primarily) as a theologian or even a biblical scholar. Despite my limitations, I’d like to offer some brief, informal reflections on Christ and Christ’s Atonement.

I used to wonder about the scope of the atonement. Christ died, but for whom? I now wonder about the mechanism of the atonement. How does the atonement save? This is an extraordinarily complex question, and I do not expect to resolve it, at least not in a way I find completely satisfactory. But I have hope that some progress can and has been made.

I confess that I am unsure how to interpret various passages allegedly related to how the atonement works to save. I believe that this skepticism justifies me in constructing a largely philosophical atonement. (After all, if the biblical evidence is currently mysterious to me, but non-biblical evidence is not, why should I not use that latter evidence to construct a theory of the atonement? I find this acceptable.) 

But a strictly philosophical atonement – that is, a theory of the atonement constructed purely on non-biblical grounds – seems unfitting for an Easter homily. I would like to say much more about what I think the biblical authors envision for the atonement, but the most I can claim for these reflections is that they make a great deal of sense to me; that is, when I read these passages, I take them to mean such-and-such. But I am unprepared to offer exegetical defenses of my readings. I am not equipped to do that, and (as I just confessed) I am already quite skeptical that I can make plausible heads or tails of the biblical data. 

This places me in the somewhat unfortunate situation of saying, essentially, ‘Well, I do not claim that this can be defended exegetically in extreme detail, only that things seem to me as being such-and-such upon initial reflection.’ I make no claim, then, that my reflections on the biblical data are such that you should regard them as plausible or binding; I haven’t the evidence to do that. Instead, I make the (much) more modest claim that this is how the biblical data often looks to me, and I welcome you to adopt whatever seems right to you about my reading. If nothing about my reading seems right to you – indeed, if it seems quite wrong to you – recall my confession of skepticism, and recall in particular that I have not said that you should accept what I say about the biblical reading.

When contemplating the atonement, we would do well to remember several things: (1) the Jewish sacrificial rituals and their parallelisms to the death of Christ; (2) the model or models of divine forgiveness in the scriptures, as well as the divine character and nature; (3) the nature of sin and its effects; and (4) how a successful redemption is portrayed in the scriptures. This is far from exhaustive, but it represents what I take to be a plausible set of theological guidelines for atonement theorizing.

We see in the Jewish sacrificial ritual a rite of moral disassociation. I have sinned, but I want to be made clean. I therefore associate with an animal of some kind (though not always: flour and fruits were sometimes used) in such a way that I pass onto it the consequences of my sin (spiritual and perhaps even physical death), and God accepts this and allows me to avoid those consequences for the present. 

Did Christ do this for us? I suspect in some form he did, and that it looks something like this: By freely becoming incarnate and living among us and eventually being tortured and murdered, Christ truly felt the consequences of life apart from perfect, Trinitarian harmony. (Life apart in this sense at least: being among those who were in a state of wicked rebellion.) By associating with the God-man through faith and obedience, we are made righteous and justified in the sight of God. But 'being in' Christ does much more for us than prior sacrificial rites did. Not only do we avoid the consequences of life apart from God, but in fact we are raised to a new life, made new creatures, and we participate fully in the divine life.

Theologians sometimes talk of the ‘transfer of righteousness,’ or the ‘imputation of Christ’s righteousness.’ Understood as an instantaneous sort of thing, this makes little sense. (Do I already possess Christ’s righteousness? If so, then why do I sin as I do? Is it merely a declaration of my present righteousness? Whence my sin? Whence the need to live rightly? Whence God’s more substantive standard of righteousness, namely, actually being righteous?) But understood as a Spirit-guided, grace-infused transformation of our character, this transfer makes a great deal more sense. It is the working of divine grace in our hearts, without which there is nothing good whatsoever.

The Gospel of Luke (15:11-31) tells the story of the Prodigal Son, who leaves home and squanders his inheritance. He returns home and admits (rightly!) that he does not deserve to be regarded any longer as his father’s son. But his father – and this is crucial as a model of divine forgiveness, as I suspect was intended by Jesus, because a number of atonement theories seem to betray the lesson here – accepted him with open arms. There was no punishment; there was no animal sacrificed for sin (only for celebration: if it were a sin offering, it wouldn’t have been consumed). There was only forgiveness and love, and an overlooking of past offenses in light of sincere repentance. The older son complains that he and his brother have received unequal treatment. The older son has always obeyed and received not even a goat; but the younger son has disobeyed terribly and has received the fattened calf!

We do well to see the point: repentance equals things out; it restores, and equal treatment is again morally appropriate. I think this justifies at least a modest skeptical attitude toward the complaint that punishment is necessary for complete restoration. The parable makes no mention of it, and even seems to oppose it, both by marked omission (quite an omission in a Jewish culture obsessed with sacrifice) and a denial of the elder brother’s complaint of unfair treatment. (Shouldn’t he receive more? Or shouldn’t his disobedient brother receive some punishment? Let’s treat all like they deserve!)

What is sin? Speaking broadly, it is a disruption to the prescribed divine-creature relationship. The law is meant to guide us to God, to show us what God requires of us. Missing the mark is missing perfect communion with God. (It is also failing to commune properly with others, but we may safely ignore this for now.) Sin means separation; it is why we are on one side of the Temple veil, and God on the other. Correcting this is what salvation is about, and likewise for the means to salvation. (Is it any puzzle that we call it the ‘at-one-ment’?) Salvation is, for us, about obedience, and obedience (in large part) so that we unite with God once more, being partakers in the divine nature. 

Whatever we say of the atonement must explain this. It is at best an incomplete atonement that fails to explain how we really are restored in a substantively transformative sense. The sum of our atonement theory must go beyond the law: it is not enough that we are declared righteous; we must be made righteous. For without being made righteous, we are not fully reconciled; we are still evil, still unclean, still missing that wonderful and perfect communion with God.

What did Christ do for us? He showed us the way to God, and did so in a way that did not downplay our own sufferings. (I say our suffering would be objectionably downplayed if only Christ’s sufferings were sufficient to recompense God.)  He lived his life in perfect obedience, and by associating with him through faith and obedience we are brought into the Kingdom. Was this something we could do on our own? In one sense the answer is ‘yes,’ but in another sense (a highly important sense) the answer is ‘no.’ It is true, I think, that we could restore ourselves to God in such a way that fits us for eternal communion with Them, but it is false that we could do this without divine grace. Christ did not need to suffer and die to repay God, but instead suffered and died to live rightly throughout tribulation, loving even his enemies to the end – even to an end on the cross.

It is beautiful that Christ’s life inspires us, and that God’s grace (through Christ’s example, among other things) fits many of us for that Love which formed us, which let us go our own way, and which plucked us from sin and darkness into light. We are saved by God, for God, and in God. For this reason, it is right to celebrate Good Friday and the atonement, but wrong to celebrate the injustices done to innocent Christ. Christ’s life was love, and so was his enemy-forgiving death (‘Father, forgive them’).

Christopher Hitchens, before his death, scolded then-Archbishop of Canterbury Rowan Williams for his plea to love our enemies. Hitchens informed Williams that Williams is welcome to love his enemies, but that he (Hitchens) would not. It’s too dangerous. (Perhaps this is why the way is narrow.) Let us be grateful for Jesus, whose love found its way into the heart of even the Roman executioner at the cross.

May we forgive others as Christ forgave us, loving, forgiving, reconciling. Even to the end. Even to a cross.

"The purpose and cause of the incarnation was that He might illuminate the world by His wisdom and excite it to the love of Himself"

--Peter Abelard

"Does Desert Justify Intentional Harm?"

[I defend pacifism in this paper against considerations of desert. This is only a partial defense of what I take to be a prima facie plausible pacifist view.]

The pacifist view I defend entails the following:

Pacifist View              It is never morally impermissible to harm a person, on grounds of desert or otherwise, unless harming the person is consistent with that person’s long-term health and is not intrinsically wrong-making.

My argument for the pacifist view takes aim at considerations of desert. In the minds of many philosophers, an individual who freely commits moral wrongdoing brings about that she deserves punishment. Moreover, the fact that she deserves punishment entails that the appropriate punishment done to her (if indeed the punisher punishes because of the moral wrongdoing) is morally permissible. We will call this view the "desert view”:

Desert View               On grounds of desert, it is (at least sometimes) morally permissible to harm a person[1].

The desert view attempts to render the pacifist view false by demonstrating that its modal claim is too strong. That is, the pacifist view claims that intentional harm to persons is necessarily morally impermissible, yet the desert view claims this is not always (and thus decisively not necessarily) the case. Thus, the pacifist view and the desert view are incompatible theses since they differ on key modal claims of permissibility.

I think it should be clear why considerations of desert are incredibly relevant to questions surrounding the permissibility of intentional harm. When it comes to intentionally harming a particular person, there are strong intuitions to suggest that the entity in question needs to be guilty of something – that is, in receiving harm, the entity in question is receiving harm because she did something morally wrong.[2] Note carefully the word “needs.” Suppose an agent has performed no moral wrongdoing (that is, she is innocent) but is nonetheless subjected to much intentional harm. On ordinary readings of the word “innocent,” this agent is such that she has not (freely) acted in any way such that she is the proper recipient of harm. She has done nothing to justify any harm done to her. In a word, she does not deserve it.

Some philosophers are inclined to think that even innocent agents can be permissibly punished. At least sometimes, this is argued on the grounds that an agent’s innocence is overridden or outweighed by other, higher moral considerations. One such example is general welfare. If in order to save the citizens of the United States I must intentionally harm – by torture, let us suppose – a particular innocent citizen (that is, a citizen whose actions, wherever they are temporally located, do not themselves justify such torture), then it is permissible for me to do so. It seems to me that such positions are ultimately misguided in virtue of a host of counterexamples, all of which (at least in my estimation) suggest that extrinsic considerations of this sort do not entail permissible action. Consider the example of Jesus of Nazareth, whom we will suppose was morally perfect. The crucifixion and overall torture of Jesus, even if it were necessary to save the whole of humanity, strikes me as a far cry from being permissible. Likewise, raping an innocent woman for the purpose of satisfying a cruel serial killer who holds your family hostage is impermissible. As I see it, intentionally harming innocent persons is intrinsically and without exception morally impermissible. No doubt some will disagree with this position. Although I am willing (and able) to argue for this position, I will not do so here. Suffice it to say, if efforts to show that it is permissible to harm innocent persons fail, then it seems plausible to suppose the only alternative to ground the permissibility of intentionally harming such entities consists of desert. It is to desert, then, that I presently turn my attention.

It will be helpful to qualify the desert view in order properly to evaluate it. As it stands above, the desert view merely commits itself to the somewhat vague position that the condition of desert is sufficient for permissibly harming to a person. But in fact, the desert view and its defenders are advocating something far more specific. The following is one such articulation:

Desert View               For some agent S, action a, time t, and person R, S freely and permissibly brings about intentional harm to R via a at t if (i) R has done something to deserve a at t, (ii) S believes this and on this basis performs a to R at t, and (iii) the fact that R has done something to deserve a at t is itself sufficient to justify the permissibility of a to R at t.

This needs some explanation and defense. First, we are concerned only with those agents who freely bring about intentional harm. Agents who do so unfreely might be doing something wrong, but this says nothing about whether or not they did something for which they are blameworthy (or praiseworthy), since (after all) either their causal powers were not exercised (perhaps they were simply bumped by a car and inadvertently slapped a pedestrian), or because they were not aware of what they were doing (perhaps they were forced into a hypnotic state and then forced to slap someone), or the like.

Second, we are concerned with agents whose actions – again, wherever they are temporally located[3] – are such that they were impermissibly done by the agent in question, in such a way that those actions entail and ground the permissibility of punishment. Again, we do not want to punish innocent agents, or agents whose actions are not such that they entail and ground the permissibility of punishment. For example, we do not want to punish innocent persons precisely because they are innocent and consequently have done nothing for which we can justifiably punish them. Additionally, agents whose actions were impermissibly done by them yet have no bearing on the current punishment (perhaps because they are unrelated, or because the actions in question have already been forgiven, atoned for, or punished) are not justifiably the subjects of intentional harm. A man who murdered his father yet is intentionally harmed for his honest day’s work is not justifiably punished, since the punishment is not done for the right reason. And it is because of this reason that the agent who performs a to R at t needs to know why she is a-ing. If S intentionally harms R without knowing that R has done something wrong, then S intentionally harms R on grounds other than desert. According to the desert view, this is impermissible, and it is not difficult to see why. Intentionally bringing about harm to an agent who, as far as I know, does not deserve to be intentionally harmed, seems morally problematic for me insofar as there is no clear alternative ground for justification[4].

Third, the desert view suggests not only that the actions of R are necessary to justify S intentionally harming R, but are also sufficient. Initially, condition (iii) seems problematic since it claims to be uniquely sufficient to justify intentional harm, but then conditions (i) and (ii) seem either necessary (in which case (iii) is not itself sufficient) or unneeded. But this is not problematic for (iii) since (i) and (ii) are not necessary conditions. they are merely descriptive of the view. For our purposes, we will keep (i) and (ii) in mind although focus our attention on (iii). According to (iii), the mere fact that R has done something wrong is sufficient for me to intentionally harm her. This seems present, for example, in our legal system and in some commonsense notions of justice. A criminal who has committed wrongdoing has made himself liable to harm, and the man who attacks me without provocation has acted in such a way that self-defense on my part (which we will suppose includes intentionally harming him) is now morally justified. In short, they have “done enough” to be intentionally harmed in a permissible fashion.

Everyone, or nearly everyone, has done something impermissible. Everyone, or nearly everyone, has lied, cheated, taken advantage of someone, and the like. For the sake of simplicity, we will suppose that intentional harm, being grounded in desert, is a case of retributive justice, and that retributive justice (or punishment) should be at least roughly proportional to the wrongdoing committed. Again, for the sake of simplicity, let us suppose that the collective wrongdoings of the average person (or perhaps the average wrongdoing of the average person) deserve a swift slap in the face. According to the desert view, and in particular condition (iii), it is permissible for me to go around giving swift slaps in the face to everyone, or almost everyone. But this is morally problematic. Consider the following argument, which I call the Argument from Moral History:

(1)   For any agent S, act a, time t, and recipient R, where R is a person: If S freely brings about a to R at t, then if a is such that it is inconsistent with the immediate or long-term health of R at t, then S brings about a for R at t for either (y) reasons independent of R’s desert-property or (z) reasons dependent on R’s desert-property.

(2)   If (y), then, necessarily, S's freely bringing about a to R at t is morally impermissible.

(3)   If (z), then, necessarily, S's freely bringing about a to R at t is morally impermissible.

(4)   Therefore, for any agent S, act a, time t, and recipient R, where R is a person: If S freely brings about a to R at t, then if a is such that it is inconsistent with the immediate or long-term health of R at t, then, necessarily, S's freely bringing about a to R at t is morally impermissible.

Recall that by “desert” I have in mind any property which is such that the instantiation of that property in a person is necessary to ground the permissibility of harming that person. Necessarily, if anyone harms a person for any reason at all, that reason is either such that it is altogether is irrelevant to the desert-property or is relevant to the victim’s desert-property. If it can be shown that neither of these sorts of reasons is sufficient to ground the permissibility of harming persons, then the desert view is false and the pacifist view is true.

Premise (1) is universal. It concerns every possible agent, every possible act, every possible time, and every possible recipient who is a person. It is also a conceptual truth: If S harms R freely, then S harms R intentionally and thus with some reason. That reason is either independent of R's desert-property, or it is not. If it is not, then it is dependent on R's desert-property. What does it mean to say that a reason is "dependent on" someone's desert-property? the dependence is not metaphysical, but rather cognitive: S bases his treatment of R on some feature(s) of R's desert-property. More precisely, S's reason for harming R depend on R's desert-property only if S believes that some morally relevant feature(s) F of R is such that, because of F, S chooses to bring harm to R. For instance, perhaps I decide to harm you because you (unjustly) killed someone, or perhaps I slap you in the face because you (unjustly) insulted my wife.

The next premise, (2), is that it is necessarily impermissible for S to harm R for reasons independent of R's desert-property. If S harms R for reasons independent of R's desert-property, then S harms R for reasons independent of what R is morally responsible for. But clearly, R should not be harmed for things for which she is not morally responsible. Consider a case in which there is a government conspiracy. Suppose the CIA assassinates the President of the United States and needs a scapegoat. the CIA, being incredibly resourceful, can make just about anyone seem to blame for anything. Thus, it is of relatively little consequence whom they select as the scapegoat. Almost randomly, they select an innocent person (call her Lacy) whom they will subsequently capture, torture, and kill, all for "shooting the president." Lacy is clearly not responsible for the death of the president, and for that reason it is impermissible to harm her as if she were. It is always wrong to harm innocent persons, and it is always wrong precisely because they are innocent. Furthermore, it is necessarily true that, if S harms R for reasons independent of R's desert-property, then S harms R for something R is not responsible for and, therefore, S is unjustified in harming R.[5] Judith Jarvis Thomson has proposed a counterexample to the truth of (2). Consider the following passage by Thomson:

...let us look at a second hypothetical case, which is like the first except in this respect: the driver is entirely without fault for what he is doing. How can that be, given that he is chasing you around the meadow in a truck, trying to run you down? Well, let's suppose some villain had just injected him with a drug that made him go temporarily crazy. It is not his fault that he is going to kill you if you do not blow up the truck, he is not villainously aggressing against you; but he is aggressing against you, and he will in fact kill you if you do not blow up the truck. Does morality permit you to blow up the truck? I think it does: I think self-defense permissible in this case - which I will call Innocent Aggressor - just as in the case I called villainous aggressor.[6]

Perhaps my intuitions simply differ from Thomson's here, but I (frankly) do not at all see that it is permissible for you to kill the innocent aggressor. In my view, this case constitutes a bona fide moral dilemma, one in which (as I suggested above) you are neither blameworthy nor praiseworthy. For moral agency requires moral freedom, and moral freedom requires that one be morally responsible; but clearly, one is neither praiseworthy nor blameworthy in cases where one does not have the power to act morally. Put differently, in the terms of 'ought implies can,' someone is morally obligated by pacifist view only if that person is a moral agent - that is, has it within her power to act according to pacifist view. This reply might be pushed back against on the grounds that, in the absence of pacifist view being obligatory, acting inconsistently with pacifist view is morally permissible. But this would miss the point since pacifist view is obligatory only for agents, and a necessary condition for moral agency is moral freedom.

The third premise is also plausible. The only relevant moral criterion dependent on a person's desert-property is immediately relevant blameworthiness. That is, if I am able to justify harming a person, where my justification depends on the person's desert-property, then, since that person's history is necessarily limited to either morally permissible features and morally impermissible features, and since the former is not (as we have seen) grounds for justifying harm, the latter must be the only morally relevant criterion.

It might be replied that it is unclear what is wrong with the view which suggests that it is permissible for me to go around intentionally harming everyone, or almost everyone. After all, there is at least some reason to believe that desert does and should make a difference in how we treat others, including when considering whether someone should be punished. But this objection is problematic on several grounds. First, in this case, it seems we have no good reason to suppose that we need to know where the desert view has gone wrong in order to know that it has gone wrong. Second, if pressed, we might simply reply that the desert view is simply wrong about the entailment relation: Mere wrongdoing doesn’t entail that the wrongdoer is permissibly intentionally harmed any more than mere wrongdoing entails that the wrongdoer should receive $1,000. In my view, this is a decisive objection to the desert view as it stands. If the desert view is true, then it is morally permissible for me to intentionally harm almost everyone.[7] Since that’s false, so is the desert view. I will now attempt to modify the desert view in such a way that it escapes this objection.

Perhaps what is implausible about the desert view, as it stands, is that it entails that just anyone can permissibly inflict harm on almost everyone. In the minds of many, including many philosophers, one’s position with respect to the wrongdoer is significant when it comes to permissible action. For example, it seems impermissible for me to harm my boss’s husband on her behalf in virtue of the fact that his offense is not against me. It is, in a manner of speaking, morally inappropriate for me to confront him. To use another example, consider (again) our legal system. Plenty of people condemn vigilante justice. What we need in order to have any sensible and permissible system of punishments (and rewards) is to establish an authority figure whose job consists of the responsibility to punish (or reward) the members over which she governs. This is why certain members of the government, but not each and every citizen, are permitted, morally, to harm intentionally. Thus, we might modify the desert view as follows:

Desert View*             For some agent S, action a, time t, and person R, S freely and permissibly brings about intentional harm to R via a at t if (i) R has done something to deserve a at t, (ii) S believes this and on this basis performs a to R at t, and (iii) the fact that R has done something to deserve a at t from S is itself sufficient to justify the permissibility of a to R at t.

So, on this modified desert view*, mere wrongdoing is insufficient for S to permissibly intend harm to R via a at t. Rather, S needs to be in a certain relation to R via a at t such that R deserves a from S at t.

Unfortunately, this modified desert view* suffers from similar problems as the earlier desert view. Limiting the number of individuals who are permitted to punish, or limiting the number of individuals whom it is permissible to punish, merely localizes the problem. Consider, first, the agent who would otherwise act on behalf of his boss. On the modified desert view*, he does not act on behalf of his boss, but rather intentionally harms only those who have, say, harmed him. Let us suppose, further, that the intentional harm he inflicts is either equal to or less than the harm committed against him, and that the man in question is me. Because my eye was poked out cruelly by my boss’s wife, I now can permissibly poke out her eye. Or because I was unjustly fired by my law partners at Jerk & Jerk, I can now permissibly send them extremely cruel hate mail. the number of our offenders is often many, and so the number of individuals we could permissibly punish according to the desert view* is also many. these numbers likely increase when we consider the example of the government official. But again, it seems false to suppose that the government official can go around and permissibly slap or terrorize just any offender.

It might be objected that this relies on the contingency of those who have harmed us, or on the number of persons under government rule. But that is false, since the modified desert view* entails that it would be permissible to punish agents in appropriate relations. So, in some possible world in which our offenders are infinite, if the desert view* is true, it would be permissible for us to punish each and every offender. Thus, it is entirely appropriate to object to it in theory.

Perhaps where both the desert view and the modified desert view* have gone wrong is in suggesting that just any wrong, or any wrong committed in relation to some particular agent, justify punishing the agent who freely committed that wrong. But this is false: We do not permissibly punish people who have lied because they have lied, say, and we do not permissibly punish people who eat unhealthily because they eat unhealthily. these are insufficient to justify intentionally harming the perpetrator. they are not serious enough to merit punishment. It is, rather, severe or serious wrongs[8] that justify intentionally harming a perpetrator. Thus, we might revise the desert view* as follows:

Desert View**           For some agent S, action a, time t, and person R, S freely and permissibly brings about intentional harm to R via a at t if (i) R has done something serious to deserve a at t, (ii) S believes this and on this basis performs a to R at t, and (iii) the fact that R has done something serious to deserve a at t from S[9] is itself sufficient to justify the permissibility of a to R at t.

I assume that it is plausible to suppose that there are such things as serious wrongs that can be plausibly distinguished from mere wrongs, or at least that there are such things as more serious wrongs and that they can be plausibly distinguished from less serious wrongs. For instance, it is plausible to hold that rape is a serious wrong, or a more serious wrong than telling a “white lie.” I therefore think it unreasonable to deny the desert view** on those grounds.

However, the problem of intentional harm being permissibly committed too liberally remains. For in some possible world there are thousands of persons who have committed serious moral infringements, to us or under the leadership of some ruler. But it does not therefore follow (contrary to the desert view**) that each and every one of those thousands of persons is permissibly the subject of our intentional harm. Indeed, it seems utterly absurd to suggest otherwise.

An equally serious problem concerns proportionality. If intentional harm is to be our outlet of justice, then we must handle it well. Let us begin by supposing that the harm we intend should be equal to the wrong committed. If that is true, then if every perpetrator has raped, they should be raped (or suffer some equivalent). But to do this seems quite beyond the bounds of what is morally permissible. the problem intensifies if we suppose that the harm we intend should be greater than the crime committed. However, perhaps the best route to take is to suggest that the pain we inflict should be less than the crime committed. That is, it should be disproportional to the crime committed. We might slap him across the face once or twice, or toss water on him and shock him (extremely briefly) with a small voltage of electricity. Aside from seeming moderately cruel, such acts of intentional harm fail to take seriously the gravity of rape. Rape, if it is to be taken seriously, must[10] be taken seriously in a way and to the degree that rape is a serious thing. Since neither equal nor unequal acts of intentional harm are morally permissible on this account, the desert view** is implausible.

One might think this sort of objection is at least equally as bad for someone who opposes the desert view**. After all, one who denies that it is ever morally permissible to intend harm to someone is committed to saying that, in cases of rape, the perpetrator should not be subjected to intentional harm. But in fact, none of this entails that cases of rape cannot (or should not) be taken seriously. It merely denies that cases of rape cannot permissibly (and thus should not) be taken seriously in a way that involves intentional harm. there are numerous problems with such an approach, as I have already suggested. Alternatives ways of taking serious crimes seriously include intense and prolonged rehabilitation, imprisonment, a lifetime of community service, banishment, etc.[11]. So, denying the desert view** does not entail that one cannot take rape, or serious crimes in general, seriously. I will now consider general objections to the view I have defended.

the first objection I will consider is a variation on the last objection. It seems extremely plausible to think that desert is both an ordinary and a permissible feature of our everyday lives. Teachers give students what they deserve, whether that consists in grades or in time spent with the principal. It seems quite permissible for a teacher to fail a given student if that student is not meeting fair and required standards for passing the class, or test, etc.

Likewise, many of our friends have lied to us, cheated us, taken advantage of us, etc. People make mistakes, but when they make mistakes of certain sorts and there is no good reason to think they have now become trustworthy, it seems plausible to suggest that they deserve to be trusted less.

Consider, finally, the example of dangerous criminals. In some very real sense, dangerous criminals should be quarantined. they are a danger to themselves and others. Surely these criminals deserve to be less trusted by society; surely they deserve to have their crimes (or their potential for crimes) taken seriously in such a way that placing them in prison or some other correctional institution is deserved.

It seems to me that these accounts are correct. It is permissible to give students failing grades if, in fact, they deserve them. It is permissible to trust your friends less if, in fact, they deserve less trust (presumably as a result of being untrustworthy). And supposing that incredibly dangerous criminals (or any criminals at all) actually are responsible for their crimes, it is permissible to treat them more cautiously and even to restrain them from being too engrained in society.

It is essential to realize that such an admission for a critic of the desert view** is not in the least problematic. All the critic of the desert view** claims is that intentional harm is not justified (or does not appear to be justified) by considerations of desert. He is not committed to the far stronger thesis that desert does not justify anything. Consequently, the objection as it stands gives us no good reason to abandon our criticisms of the desert view**. Denying the view simply does not come with the cost of denying that considerations of desert are morally relevant and difference-making in treatment other than intentional harm.

Consider one rather different sort of example in which a teenage boy is in a severe car accident and is fatally wounded. EMTs arrive on the scene and realize that the boy’s internal organs are now externally visible. Even worse, his organs are in bad shape. the boy is conscious and in immense pain, screaming. He has only thirty more minutes to live, and those in utter agony. He asks that the EMTs give him a fatal dosage of morphine to easy his pain and to deprive him of thirty minutes of misery. Legal concerns aside, it seems the EMTs have a moral obligation to do what is in the boy’s best interest (all other things being equal), which in this case entails that they should administer the morphine. It seems just as plausible to suppose that the boy, in virtue of his suffering, deserved something better; he deserved to be relieved of his pain. Fred Feldman offers a similar case for consideration:

Suppose there are two sick children are in the hospital. Suppose each has a painful disease. the first has suffered for several months, and has been quite miserable. Yet the doctors are perfectly certain that she will soon become quite well, and in a short time will be fully recovered with no lingering effects. the second has also suffered for several months, and has also been miserable. In this case, however, the prognosis is different. the doctors are perfectly certain that he will soon die[12].

Feldman goes on to suggest that, in his estimation, this child deserves the offer made by the Make-a-Wish Foundation: namely, “an all-expense paid trip to Disneyland.” Feldman thinks that, with respect to the trip to Disneyland, the second child “would be the more deserving, precisely because he is going to suffer the greater misfortune.” By this, I assume Feldman means to endorse that considerations of desert are relevant to relieving the suffering of individuals, that they (in some sense) deserve to be compensated for the pain they have endured. This is certainly compatible with depriving of misery the person in question, ending her suffering by ending her life. But it seems the pacifist view is incompatible with this. Thus, the pacifist view is mistaken.

Interestingly, the pacifist view, as it stands, is compatible with terminating the lives of persons provided that doing so is not intended to be harmful. In cases like the boy in the car crash or Feldman’s case, it seems that what the boy and the sick patient deserve is not intentional harm, but rather intentional help or intentional compensation. Those who endorse the pacifist view can easily accommodate such considerations.

I have claimed to be critiquing the desert view** as a means of defending the pacifist view. But it should be realized, as I mentioned early in the paper, that it is contentious whether the failure of the desert view** entails the success of the pacifist view. That is, it is unclear whether there are morally relevant considerations other than desert which justify bringing about intentional harm to persons. Perhaps factors other than the agent – i.e., extrinsic to the agent – are morally relevant: the agent’s family history or geography, the commands of God, the general welfare, and such. I haven’t the space to discuss these possibilities, although it seems to me that none succeeds. But it seems to me that, if the pacifist view is to succeed, it must show that positions like the desert view** are implausible. I have done my share of the work here by criticizing the desert view**.

Another possible response on behalf of the defender of the desert view** is that considerations of desert are necessary but not sufficient conditions for permissible intentional harm. Of course, that is just to admit that desert views, as they stand, are false. But perhaps it is the case that desert, if supplemented with something else, suffices to render it permissible to intentionally harm the agent who meets those conditions. Unfortunately, this sort of objection seems to ignore the fact that this paper has been an exploration of supplementing considerations of desert with other considerations in order to generate a sufficient condition. Moreover, there is some reason to be skeptical that there are any such other considerations. We have modified the relevant factors in significant ways: the type of wrongs, the type of punisher (i.e., agents who stand in a certain relation to the subject of punishment), and the type of punishment. Perhaps another variable we might change is the number of punishments that can be administered. Maybe it is permissible for me to intentionally harm only one or two people a year, say, or at least for there to be some limit to the number of people I harm. This is a very straightforward way of avoiding my objections, but it is also unmotivated and post hoc. If it is permissible for me to harm only a limited number of people a year, in virtue of what is this is the case? If it is not in virtue of something about the agents, then it is unclear what good reason we have for intentionally harming one agent rather than another[13]. If it is in virtue of something about the agents, what is it? Furthermore, what good reason is there to think that there are additional categories beyond those explored here? Although perhaps not unassailable, it seems to me that this dilemma is problematic for defenders of the desert view**.

            No doubt there will be attempts to modify option (z) so as to avoid this implausible consequence. But as I see matters, any such attempt is bound for failure. For it seems to me that the only ways of attempting to reconstruct (z) are to suggest either that (z) does not have this implication actually, but merely possibly; or, that (z) does not have this implication even possibly. the first attempt is, in my view, irrelevant, for the counterfactual If some circumstance C occurs, then (z) would 'tell us' that it is permissible to go around slapping everyone in the face. This counterfactual consequence is sufficient to render the first response unsuccessful, since implausible counterfactual implications are (at least) almost as morally unpalatable as actual-world, non-counterfactual implications.

The second response, which is (or would be) an attempt to modify (z) so as to avoid the implication in question, is also, I think, something we need not concern ourselves with, at least not too much. One might suggest that (z) is prima facie permissible, but in cases where the number of persons is so significantly large, there are ultima facie reasons for refraining from punishing others. This strikes me as horribly ad hoc. Moreover, consider a deeper problem, one which I will illustrate with an analogous, but distinct, principle. First, consider the principle ‘An act of torture is justified provided that you do not torture too many people.’ We might call this a prima facie principle of justification for torture. It has a caveat, namely, that the number of persons tortured not be excessive. Perhaps this is correct. However, there is no reason to believe it, and (consequently) it remains unjustified. But even this response is inadequate, since it does not effectively rule out possible revision; and, if it does not rule out possible revision (where, of course, the revision ‘solves’ the problem at hand), then it does not show that it is necessary that (z) fail. If that is true, then (3) is unjustified.

Essentially, then, a revised (z) - call it (Z) - would (and must) amount to, entail, or make probable the principle that ‘it is permissible to harm guilty persons, unless there are too many of them.’ Are there any plausible or possible candidates for (Z)? I cannot think of any, but that does not entail that there aren't any[14]. For the moment, then, I find nothing intrinsically unacceptable about this alternative. Although it carries a strong scent of being ad hoc, I cannot find anything impossible about (Z) that is immune to my criticism. However, this does not entail that (3) is unjustified, since a case for (Z) has yet to be proposed and, consequently, its plausibility is (as yet) indeterminate.[15] On this front, I commend pacifists to find a way to demonstrate that (Z), which (as things currently stand) might avoid my criticism, inevitably fails. I find myself presently unequal to the task.

Despite all this, the Argument from Moral History seems to me to be a good argument, one which carries considerable plausibility despite a possible exception. (Recall that this possibility is merely epistemic; it is not, so far as I know, metaphysical.) For even supposing that (Z) is possible, it is unclear whether it is plausible. (Obviously, if it is plausible, that is problematic for the argument.) But (3) can be modified in such a way to make room for this epistemic possibility, and such a modification need not detract from the power of the argument.

---

[1] I offer this weaker version since the stronger version is (in virtue of being stronger) less plausible, since it appears to rule out the possibility of overriding considerations. As I will argue, this weaker desert view needs to be weak if it is to have much hope of escaping my objection.

[2] By ‘desert,’ I also mean to include any other property p which is such that, if p is instantiated in a person, that person is permissibly harmed on the basis of p. This includes properties such as being liable to harm, as in when a person freely engages enemies in warfare. Although I will speak almost wholly of desert in terms of moral wrongdoing, this is purely for purposes of clarity and concision.

[3] There is some debate as to whether future actions are plausible candidates for desert. I stake no claim on this issue, although it seems to me odd that actions which have not yet occurred can currently (that is, right now) ground the current fact that someone deserves to be intentionally harmed.

[4] And even if there were, we are supposing (currently) that other prospects are dim.

[5] This is consistent with the counterfactual “If S knew some relevant fact F about R’s desert properties, then S would be justified in harming R on the basis of F.” But we are supposing here that S lacks a justified belief that F is true.

[6] Thomson, Judith Jarvis. "Self-Defense," Philosophy and Public Affairs, Vol. 20, No. 4 (Autumn, 1991), p.284.

[7] Rights theorists might press that a necessary condition for permissibly harming persons with desert-properties is that you have a right to harm those persons. Perhaps persons in government, for example, have the right to harm (certain) persons with desert-properties, whereas the everyday citizen lacks that right. But then it would be permissible for persons in government to go about slapping people in the face in NYC. It is difficult to see why that would be less counterintuitive.

[8] It might be wondered why ‘serious wrongs’ done by a free agent makes that agent more of a candidate for permissible intentional harm than ‘mere wrongs’ done by those agents. I admit that there is something mysterious here, but for current purposes the suggestion is that serious wrongs are intrinsically difference-making with respect to bringing about permissible intentional harm.

[9] I have kept the modification ‘from S.’ However, it can be removed without affecting my upcoming objections.

[10] And by ‘must’ I mean to identify a moral obligation. We should take rape quite seriously, since it is a serious wrong.

[11] Each of these could be done in ways that include intentional harm. I am referring to utilizing these approaches in ways that don’t include intentional harm.

[12] Feldman, Fred. “Desert: Reconsiderations of Some Received Wisdom,” in Mind 104 (413), p.71.

[13] And recall: the defender of the desert view** will have to say that only the agents in the limited category are permissibly harmed. So, it will do no good to suggest that agents outside the limited category could be permissibly intentionally harmed, since that would be merely to return us to earlier criticisms.

[14] Actually, I can (sort of) think of one. (Z) might be argued for on intuitive grounds, perhaps in the following way: ‘(Z) is permissible only if there is no outweighing cost.’ the defender of (Z) could then go on to suggest that harming too many people is an outweighing cost, using the same intuition I have been using. But how can the defender of (Z) account for this intuitive truth? If it is permissible to punish a guilty person, why would it matter (short of destroying or severely debilitating humanity) how many persons are harmed?

[15] More specifically (and carefully), the epistemic situation (as I see matters) is this: We have good reason to believe there are no exceptions, unless (Z) is true; but we have no reason to believe (Z), and (in fact) have an initial presumption against (Z); thus, we have good reason to believe there are no exceptions, and no good reason to believe there are exceptions.

In Defense of Affirmative Action

I take it that racism is a clear-cut moral wrong, or at least a case of wrongdoing which is quite close to being clear-cut. But what is wrong with racism? Racism is wrong because it entails or includes acts which are designed to favor a particular person or group over another person or group based on an irrelevant criterion, where that criterion is the person's or group's race/ethnicity. To see this, consider the historical mistreatment of African Americans. Many African American persons were used as slaves in the United States, deprived of various privileges to which they were entitled and subjected to various harmful treatments, and all because of their skin color. But there is, quite clearly, nothing morally impermissible about having a certain skin color. Even if there were, since African Americans are not responsible for their skin color, we would be unjustified in punishing them for their skin color, or for treating them as if they were responsible even if we did not believe they were.

I think plenty of politically conservative persons have a moral distaste, even a severe moral distaste, for racism. Even more, I think politically conservative persons desire the elimination of racism, and (in at least some cases) have no problem apologizing, on behalf of themselves or others, to victims of racist behavior. In these respects, I think we are very much on the same page. I would find myself in a state of moral outrage (and, I think, rightly so) if a politically conservative person observed some racist behavior but thought (or declared) it wrong to comfort the victim, or apologize.

But politically conservative persons often stop there, and I think that in doing so they are mistaken. To go further than this, in the minds of many conservatives, would be to engage in what is often called reverse discrimination. Of course, discrimination is broader than racism, but it includes racism. It also includes sexism, homophobia, ageism, and the like. If someone is a victim of racist behavior (say, because the person is African American), it would be wrong for us to privilege that person because he or she is African American. Apparently, many politically conservative persons believe that this is the chief problem with affirmative action, which seeks to favor certain "minority" groups, or groups which are discriminated against. Moreover, they often view this as an insurmountable problem; there just isn't any way for affirmative action to avoid the problem. I disagree, and let me explain why by appealing to a thought experiment.

Suppose a teacher has ten children in her tiny, Midwestern classroom. As it turns out, nine children have brown eyes and one child has blue eyes. All of the brown-eyed children are named George, and the blue-eyed child is named Alice. One day, the ten children go out to play on the playground. While on the playground, the Georges surround Alice, discover she has blue eyes, and subsequently beat Alice because she is blue-eyed. The Georges and Alice return to the classroom, and the teacher discovers that Alice has been beaten up. The teacher believes Alice has been treated unfairly, and she spends extra time talking with Alice, trying to comfort her and make Alice's day better.

By spending extra time with Alice and showing her special treatment of sorts, did the teacher do something unfair to the Georges? I do not think so. Alice clearly merited some extra time on the teacher's part due to her special need. What's more, the teacher did not favor Alice because Alice had blue eyes, but only because Alice had been abused. So it would appear that the teacher's special treatment of Alice was, in this case, morally justified.

Consider contemporary (and historical) parallels. African Americans and other groups are discriminated against in morally unjustified ways. It is (or would be) permissible for them, like Alice, to receive special treatment in relevant ways. For example, it would be permissible to offer your extra room to an African American family over a white family if the African American family had been discriminated against in their search for housing, but the white family had not. This doesn't entail that the African American family is morally superior to the white family; it merely entails that the African American family has some morally relevant features that the white family lacks, features in virtue of which there is stronger moral reason to help the one family over the other. (Consider a similar example: Family A hasn't eaten in over a week, whereas Family B is pretty hungry because they had an early lunch. Is Family A better than Family B? No; Family A is just in greater need than Family B.) Likewise, many women and elderly persons are discriminated against. There is thus a relevant distinguishing feature between them and those who lack experiences of discrimination (or something of equal importance).

Here is another possible objection, this time leveled against the parenthetical example above: Surely it is not Family B's fault that Family A hasn't eaten in over a week. If that is true, then Family B should not be punished for the circumstances of Family A. There is a straightforward response to this objection: Family B isn't being punished for the circumstances of Family A. Rather, there are stronger moral reasons for feeding Family A instead of Family B, and on that basis we would be justified in feeding Family A over Family B.

There are, I think, ways to show that affirmative action is not morally justified in today's world. For example, one could deny that any or much discrimination occurs. The evidence, however, strongly suggests otherwise, so this move is (epistemically) unjustified. For example, the phenomenon in social psychology known as Shopping While Black shows that African American persons are frequently accused or suspected of shoplifting, despite a lack of evidence on the part of the accuser(s) or suspicious persons. Likewise, examples of female subordination abound. Women are frequently paid significantly less than their (equal) male counterparts, abused, raped, and mutilated.

One could also suggest that, although there are better moral reasons for favoring one group over another, the reasons are not strong enough, or not the right kind of reasons, to justify favoring one group over another. Perhaps there is some legitimacy to this in a hypothetical sense. After all, if I was placed in a situation where I had to decide whether to kill Person A or Person B, even if Person A was a slightly better person (in a moral sense), it is doubtful that would justify killing Person B. Or perhaps the economy is such that jobs are needed and slim, and it would be wrong to favor an African American over a white person for the job even though the former, but not the latter, had been discriminated against. (If these judgments are incorrect, of course, it is a problem for the critic of affirmative action.) But I fail to see that the reasons for favoring one group over another aren't good enough reasons overall. Perhaps the example of the economy is plausible, but it is doubtfully true with respect to everything (e.g., schools, political offices, housing, etc.). Moreover, even if it were true with respect to everything, this would (it seems) be a merely contingent problem with affirmative action, not a necessary one. That is, affirmative action would be unjustified because of the contingent circumstances in which we find ourselves, but (clearly) those circumstances could be different and if they were, the same objection against affirmative action wouldn't hold true.

Another move would be to suggest that affirmative action interferes with free will. Some people, after all, do discriminate unjustly, but that hardly makes it permissible for us to interfere with their decisions. But this response has an extremely counterintuitive consequence, namely, that the Emancipation Proclamation was unjustified. Clearly, stopping slavery (or trying to stop slavery) is morally justified.

Yet another move would be to accept the examples but conclude that they are valid only with respect to individuals, but not to groups. I think it is clearly possible for there to be a "group example." Suppose that a very bright group of students - call them the Copernicans - are kicked out of school because they do not adhere to geocentrism. Suppose that a more scientifically sensible principal comes to rule the school, and on Character Recognition Day, the principal awards the Copernicans, and only the Copernicans. (The award can go only to one person or group, you see.) There are other intelligent students of course, many of whom are just as intelligent as the Copernicans. (Let us suppose, however, that they are a rather intelligent group of geocentrists.) Would the principal be justified in recognizing the Copernicans? 

Those unpersuaded by this example should consider another example: Two college applicants, Judith and Michael, are competing for the last available student scholarship at Now-We're-Moral University. Both are exceptionally bright, hard-working, and have identical grades and ACT scores. But Now-We're-Moral University was previously called Driscoll University, an extremely sexist institution. They turned Judith down a number of times because she was a woman, but were willing to accept males with the same (and even lesser) qualifications. Would the committee at Now-We're-Moral University be justified in giving the scholarship to Judith over Michael? In light of prior circumstances, it seems far more fitting, and indeed permissible, for Judith to receive the scholarship over Michael. Again, it's not that there's something about Michael that is undeserving; he is quite deserving. But in light of her prior horrible experience with the university, Judith is more deserving that Michael.

Before I close, I will offer yet another example of justified "preferential" treatment. Military physicians are sometimes confronted with cases in which they must perform triage. If a physician has limited pain medication and must decide whether to administer it to only one of two patients, the first who will die if he doesn't have his pain relieved, and the second whose migraine will continue if his pain isn't relieved. It seems fairly obvious that it is permissible, and surely even obligatory, to administer the medication to the first patient.

I close with a final example, an appeal of sorts. Suppose you go out for lunch and discover an African American woman (to continue with our primary example) purchasing lunch. The manager, discovering that the woman is African American, refuses to serve her lunch. I think you would be morally outraged. Suppose that a short time after this, another woman steps up to purchase her lunch, but is a few pennies short of what she needs. You have only a few dollars and could pay for either person's lunch, but not both. Would you be morally justified in paying for the first woman's lunch over the second woman? I suspect your answer, if you understand the example and are honest, is the same as mine. If I am correct in this judgment, then it will likely seem true to you, as it does to me, that we can (and sometimes do) have good reason for favoring one person or group over another. Quite possibly (and even actually), in the case of a group, that group shares certain biological characteristics, such as sex or skin pigmentation. It should then be obvious what these facts entail for the one who accepts them: a reason to support affirmative action (in some form).

"A Kantian Defense of Pacifism"

[This paper was written as my senior (undergraduate) thesis and was presented to an audience at Cedarville University. I am grateful to the various persons mentioned in this paper, along with many others (including Dean Erickson, Rebekah Hereth, Luke Lewallen, and Jeff Gates), for their significant and helpful contributions. I will soon post another essay, "Advice to Fellow Christian Pacifists," in which I modify several of the views expressed in this essay.]

[1], [2]

The people of the world have gathered together to speak with God in order to determine whether God values a harmonious relationship with them. They have devised a test in which, if God truly values Her human creation, She must create a human person, torture it, send it to hell, and annihilate it. If She does this, the people of the world will, as they sincerely promised, pursue a harmonious relationship with God.

Would God have done something good if She created a human person, tortured it, sent it to hell, and annihilated it? In terms of the general welfare, it seems the answer is ‘yes.’ After all, God would have then acquired the devotion of all Her human creatures. However, it seems that in another sense the answer to this is ‘no’: God would have done something wrong if She treated a human person in this way. Put differently, although we might fairly say, “Well, on the grounds of accomplishing the general welfare, or the greatest good for the most people, this was successful,” we could also fairly say (it seems), “But what a nasty, evil thing to do to a person.”

Our intuitions in this latter sense seem best explained by the fact that we believe that our actions ought to treat human persons well[3] – we believe that our actions ought to be justified in terms of the proper recipients we know those actions would affect. Presumably, this is also why we believe slavery and genocide are wrong: Both fail to justify the actions in terms of the action’s effects on proper recipients[4]. Of course, both slavery and genocide have been defended on the grounds that they can sometimes preserve the good of the general welfare, but this fails to acknowledge our intuitions that slavery and genocide treat persons wrongly. Since our intuitions seem to favor this latter notion, it seems we are at least prima facie justified in endorsing the following moral principle, which I shall call the Principle of Effect (PE):

(PE) For any agent S, action A, time t, and proper moral recipient(s) R, if S reasonably believes that A will affect some R at t in a way that could maintain or improve R’s well-being at t and t+n, then S is morally obligated to justify A as attempting either to maintain or improve R’s well-being at t and t+n.

An example would (surely!) help. If I need to toss a pan full of hot grease out the window of a two-story apartment onto a busy city sidewalk, I probably will realize that doing so will affect human persons, whom I take to be important to treat well. Indeed, I likely will realize that my throwing the boiling grease out the window will result in serious burns to the persons on the sidewalk. Presumably, this action would be extremely difficult to justify in terms of the individuals it affects since it neither maintains nor improves their well-being, or general health. Thus, I likely cannot justify that action in the required way and, consequently, it fails to meet the necessary standards for morally permissible action. PE also rules out doing something (or refraining from doing something) which detracts from a proper recipient’s long-term well-being, given the requirement of “t+n.” Thus, if I implant a device in your chest that I know will cause you to experience extreme pain three years from implantation, I have violated PE: My action failed to attempt to maintain or improve your well-being in its later effects.

It is essential to mention what I mean and don’t mean by ‘well-being.’ A person’s well-being refers to her general health, which includes proper functioning of the organs, a state of general painlessness, those things necessary for psychological health (e.g., sanity, a healthy view of oneself, etc.), and the like. Thus, being shot in the arm is generally problematic for a human person’s well-being since it is problematic for proper blood flow, is antithetical to a state of general painlessness, and causes psychological trauma. Conversely, the pain experienced in scraping an arm or in receiving a flu shot are not ordinarily threatening to a person’s general health. They hurt, certainly, and are problematic in that respect; however, the pain of the flu shot is, in most cases, insufficient to justify the belief (for example) that one’s liver is shot, or that one’s heart is going to fail, or that one is in a general state of pain, and are therefore insufficient to justify the belief that one’s well-being (or general health) is endangered[5].

Consequentialists will perhaps be the first to reject this principle, and so I will deal with them first. The consequentialist might reply that treating persons in this way fails to be ‘right,’ but does not fail to be good (where ‘right’ has to do with treatment of persons and ‘good’ is identical to the overall general welfare). If by this distinction the consequentialist is suggesting that ‘right’ is not a moral category, then the consequentialist and I are not discussing the same issue, since as far as I am concerned the treatment of persons is a moral matter[6]. But perhaps the consequentialist is merely suggesting that the ‘general welfare’ morally overrides the side-constraints advocated by PE. This is severely question-begging since it simply denies without further consideration the sort of cases I have already envisioned. The consequentialist cannot plausibly condemn me for question-begging, since I have not merely assumed that consequentialism is false but have instead offered seeming exceptions to the consequentialist view. These considerations suggest that the consequentialist take seriously the sort of examples provided. However, despite these considerations, consequentialists needn’t get too touchy since PE is compatible with the thesis that the consequences of actions are morally significant. PE simply denies that the consequences are so significant that they override equally significant considerations, like the intrinsic value of persons.

Another possible objection is that rape and torture are morally impermissible because they demean human beings, but intending harm (the objection goes) needn’t always be like that. After all, I can kill someone in war without demeaning her. But presumably what’s wrong with demeaning someone is that such a person ought not to be treated in a demeaning way, which suggests that there is a way we should treat persons. I have argued that a necessary condition for proper treatment is to intend to maintain or improve the person’s well-being in all that you do. If you reject this as a necessary condition, then you would either have to maintain that demeaning someone isn’t always wrong (in which case many of my comments to the consequentialist apply also to you) or that demeaning someone is wrong in a ‘basic’ way and thus doesn’t need my PE for further justification. If you pursue this latter route, then you are committed to the following claim being true: “Demeaning someone is impermissible, but not because it fails to maintain or improve her well-being.” This is extremely counterintuitive since it suggests that demeaning you is wrong but not because it in any way hurts you, thereby suggesting that my treatment of someone (assuming demeaning someone treats someone) isn’t problematic on the grounds of treatment per se, since what’s wrong isn’t how my actions affect her well-being. But what else is there to harm about a person other than a person’s well-being, construed broadly? I suggest that the answer is ‘nothing,’ since it seems to me that if our concern really is with the person as an end-in-herself, the nature of our concern is going to be with that person’s well-being in some respect. It makes little sense to me to say, “It is wrong to torture a human person since a human person is an intrinsic good (or an end-in-herself), but that has nothing to do with the person’s well-being,” since comments like these seem to direct our attention away from the person. But if the moral issues don’t concern the person per se, then we’re back to the problem the consequentialist (and those like him) face: namely, the common intuitions I have offered as examples.

Since a particular conception of intended harm is central to the discussion at this point, I will offer a careful definition:

A is intended harm = def. A is an action (1) freely intended by some agent S (2) toward a proper moral recipient(s) R (3) such that A is intended to bring about something which S takes only to detract from what S takes to be the well-being of R.

It seems to me that PE is logically incompatible with the view that intended harm is at all morally permissible. To see why, consider the following argument, where A is an action (or inaction) that is intended by me to harm a proper moral recipient R at time t:

1        It is morally permissible for me to A to R at t just in case A maintains or improves the well-being of R at t.

2        But it’s not the case that A maintains or improves the well-being of R at t.

3        Therefore, it’s not the case that A to R at t is morally permissible for me.

The first premise is simply a restatement of PE. The second premise is plausible since intended harm purposes to detract from the well-being of a proper moral recipient. Since an action which detracts from the well-being of a proper moral recipient entails that the action neither maintains nor improves the well-being of the recipient, it follows that intended harm is logically incompatible with PE.

What follows from the argument? Presumably, many of the common ways of going about war and self-defense are morally impermissible since those ways frequently intend harm to proper recipients. However, does it follow from PE that killing is wrong? After all, not all killing is the result of intended harm. Consider the case of Sigmund Freud whose later days were spent in immense pain until finally his physician out of compassion gave him an overdose of morphine. Or consider babies born with defects which cause them horrendous pain. Surely in these cases it would be a mistake to consider active euthanasia intended harm. If anything, it seems these are cases where what is intended is improving the well-being of individuals since their pain is removed.

But this is mistaken. If I am on the battlefield and I capture a soldier who will likely be tortured until dead, I would not be intending to maintain or improve his well-being by ending his life[7]. Instead, I am depriving him of well-being in a way I deem less painful than other ways (e.g., torture and subsequent death), yet nonetheless depriving him of well-being.[8]

O B J E C T I O N S

The first objection I will consider is expressed aptly in the words of C.S. Lewis:

You cannot do simply good to simply Man; you must do this good or that good to this or that man. And if you do this good, you can’t at the same time do that; and if you do it to these men, you can’t also do it to those. Hence from the outset the law of beneficence involves not doing good to some men at some times.[9]

And this in fact most often means helping A at the expense of B, who drowns while you pull A on board. And sooner or later, it involves helping A by actually doing some degree of violence to B. But when B is up to mischief against A, you must either do nothing (which disobeys the intuition) or you must help one against the other.[10]

For Lewis and those in agreement with him, PE is potentially impossible to uphold as a principle since it is logically possible (and sometimes actual) in which any course of action will harm someone. Thomas Nagel suggests the following as an escape from such a dilemma:

But what if the world itself, or someone else’s actions, could face a previously innocent person with a choice between morally abominable courses of action, and leave him no way to escape with his honor? Our intuitions rebel at the idea, for we feel that the constructability of such a case must show a contradiction in our moral views. But it is not in itself a contradiction to say that someone can do X or not do X, and that for him to take either course would be wrong. It merely contradicts the supposition that ought implies can – since presumably one ought to refrain from what is wrong, and in such a case it is impermissible to do so.[11]

I see Nagel’s response as implausible. In my view, ‘ought implies can’ is extremely plausible and hence quite a heavy sacrifice to make.[12] An initially possible solution to Lewis-type dilemmas is that they fail to show that PE is false since they fail to show that the sort of intentions (or attempts) specified by PE cannot occur. To illustrate, suppose that a lifeguard is walking down a peninsula when he realizes that two kids are drowning: one to his left, the other to his right. Suppose that the distance between them is such that, if he attempts to save one, he is unlikely to make it to the other in time. Suppose further that he agrees with PE and deliberates in this as to its implementation. He realizes the following three are possible courses of action:

(1)   Attempt to save Child 1 instead of Child 2

(2)   Attempt to save Child 2 instead of Child 1

(3)   Attempt to save neither Child 1 nor Child 2

But he realizes that

(4)   If (1), then he fails to abide by PE with respect to Child 2

(5)   If (2), then he fails to abide by PE with respect to Child 1

(6)   If (3), then he fails to abide by PE with respect to Child 1 and Child 2

In (1), (2), and (3), PE is not upheld since in each case there is a proper moral recipient whose well-being is not intended to be maintained or improved by the lifeguard. So far, PE isn’t sustainable in the Lifeguard Case. However, there is a fourth possible course of action: namely,

(7)   Attempt to save both Child 1 and Child 2

The Lifeguard Case is one in which the saving of both children is unlikely, perhaps even extremely unlikely, but it doesn’t follow that an attempt cannot be made or intended, which is what PE suggests is necessary (and thus always possible). The lifeguard might intend to save both children, possibly carrying that out by sincerely attempting to run from one to the other.

It might be objected that this move is implausible since it is unclear in what sense it can be said that I intend to help the second individual, since my actions are not immediately directed toward that individual[13]. But it seems to me that this is mistaken. Suppose I intend to clean up ‘Christmas trash’: all the paper wrappings, bows and so on, tossing them in the garbage. Suppose I am also running short on time and realize I probably won’t be able to clean up all of the garbage. As I hurriedly clean up, could it plausibly be said that I am not really intending to clean up all the garbage? It seems not, and so the same holds true in the other case: I intend to carry the first individual closer to the second individual in order to help the second individual (as well as the first individual, of course). Because the intention is to help both individuals, it follows that the intention is also to help the second individual. Of course, it might be unlikely in Lifeguard-like cases that both parties can be saved. However, our duties extend only insofar as our capabilities, and so we must simply do the best we can in these cases.

This possible solution is unlikely to convince some[14]. They will see it as a foolish effort since by trying to save two lives one’s total efforts are divided in half, and this is problematic since one’s total efforts are necessary to save even one child. But because of the division, one cannot dedicate one’s whole efforts to one child, and thus both children will die. At least save one, the reasoning might go. There is a possible solution here for that person, a solution consistent with PE. The requirements for PE state that my obligations in these difficult cases apply just in case I reasonably believe that my actions will affect both children in a way that could maintain or improve their well-beings at the time. Since the objector claims this isn’t the case, it would follow that extreme Lifeguard cases fail to meet the requirements for the obligations imposed by PE. Hence, if I am in an extreme Lifeguard case, I am not obligated to attempt to save both children. Other moral principles will surely be relevant here and will serve to guide my actions or another’s actions, but that is for another discussion. It seems to me that this solution is more plausible than the first, and so I tend to prefer it over and against the first solution. However, whichever of these two options one endorses on this point will be consistent with PE: If my actions will make a significant difference, then I should attempt to make that difference; if my actions won’t make a significant difference, then PE does not obligate me to try.

Considerations of desert are among the more popular objections to pacifism, including the variety I defend. It seems those guilty of wrongdoing no longer deserve to be treated well, or at least not as well as those who are innocent, and thus my view is mistaken since I believe that all human persons (including the guilty) should be treated in accordance with PE regardless of their past, present, or future actions. But the first form of the objection is far too strong. It is morally heinous to suggest that guilty persons needn’t be treated well, since all human persons are guilty of something (and thus there is a degree of vagueness as to where one ‘draws the line’) and because it remains impermissible to rape or torture guilty persons, which likely means that we have a duty to justify our actions toward even guilty persons in terms of their well-being. Of course, this need not entail that there are no legitimate applications of desert. If an acquaintance of mine perpetually reveals my secrets to others, she deserves to be trusted less. Thus, considerations of desert are morally permissible insofar as they cohere with PE, and there are quite a few possible instances in which this is the case. Do considerations of desert such as imprisonment and corporal punishment cohere with PE? It seems to me that the answer turns on how such things are carried out. If imprisonment really does fail to maintain or improve the well-being of an individual – if corporal punishment really does fail to maintain or improve the well-being of an individual – then those things are impermissible. However, I am unconvinced that all forms of imprisonment fail to do what PE requires. For instance, some prisons function like rehabilitation centers, attempting through peaceful confinement and methods to reintroduce persons into society. Corporal punishment, if it is shown to detract from someone’s well-being[15], is impermissible since although they might intend long-term improvement for an individual, they are nonetheless impermissible since the means they employ do not meet the criteria for PE. Likewise, torturing someone in order to improve his character (torture-for-improvement) is impermissible since torture can reasonably be said to detract from someone’s general health.

It is here that the matter becomes somewhat more complex. One could argue that the structure of using chemotherapy on a cancer patient is sufficiently similar to torture-for-improvement to cause problems for PE. Like torture-for-improvement, chemotherapy can be (and often is) used to maintain or improve an individual’s well-being in the long run even though it detracts from her well-being in the short run. But since it’s obviously true that chemotherapy is morally permissible (the objection goes), it can’t always be impermissible to detract from someone’s well-being in the short run. Consequently, it seems that PE ‘casts the net too wide,’ so to speak: It is sufficient to rule out torture-for-improvement, but it fails to discriminate between obviously impermissible things (like torture-for-improvement, which it says is impermissible) and obviously permissible things (like chemotherapy, which it says is also impermissible).

There are grave difficulties for this line of objection. First, it will not do to suggest that torture-for-improvement is morally impermissible on other grounds, since those grounds would have to show that torturing someone to improve her is morally wrong but not because we are morally obligated to justify our actions toward that person. But then what would be wrong is not that a person is harmed and shouldn’t be, since we have no duties to justify our treatment of her in the first place. This requires a drastic re-interpretation or denial of our strong intuitions that we should justify our actions in terms of a person’s well-being. Consequently, it seems implausible to suppose that other justification is available as a plausible alternative. Secondly, either the structure of chemotherapy is sufficiently similar to torture-for-improvement to create a counterexample or it isn’t. If it’s not, then there is no use worrying about this and the objector must try again. If, on the other hand, chemotherapy is sufficiently similar to torture-for-improvement, then I am inclined to suggest that chemotherapy is morally impermissible given that it is sufficiently similar in structure to torture-for-improvement. Put differently, since torture-for-improvement is clearly wrong, if chemotherapy is sufficiently similar in structure, that speaks poorly of chemotherapy. It’s far more plausible to think this than it is to think that chemotherapy’s similarity to torture-for-improvement shows the permissible nature of torture-for-improvement. Moreover, it doesn’t seem implausible to deny that chemotherapy is morally impermissible since it detracts significantly from the short-term well-being of persons, causing vomiting, general weakness, hair loss, immense aches and pains, and (sometimes) depression. It is true that chemotherapy does save lives, but the way in which those lives are saved seems to me to violate the duties we have to those persons. In any case, it is far clearer to me that torture-for-improvement is impermissible than that chemotherapy is permissible[16].

---

[1] My defense of pacifism is Kantian insofar as it shares significant similarities with Kant’s views on respect for persons. Although I believe that Kantianism better supports pacifism than non-pacifism, this is a contentious issue, and even Kant himself seems to have fallen on the non-pacifist side of things. See, for instance, Kant’s Perpetual Peace in Kant's Political Writings (Cambridge: Cambridge University Press), 1991.

[2] I am thankful to fellow members of my Senior Seminar course – Timothy Cuffman, Danielle Early, Katy Hawkins, and Shawn Graves – for extensive comments on various drafts of this paper. I am also thankful for help from Aaron James, John White, Ryan Peterson, Rebekah Jones, Bill Patton, Timothy Ronco, and several anonymous readers.

[3] I leave it open-ended whether anything else falls into this category, such as non-human animals.

[4] By ‘proper recipients,’ I have in mind the following definition:

                X is a proper recipient = def. X is such that any action which affects X is morally relevant.

[5] Danielle Early has cautioned me to make clear that I am by no means advocating that physical pain alone constitutes well-being. The reader will do well to bear in mind the various factors I have mentioned in this paragraph.

[6] And I do not merely assume this or declare it in some question-begging way, but rather utilize cases in which the proper treatment of persons seems clearly obligatory. Either the consequentialist accepts that our intuitions in moral matters have positive epistemic status, or she doesn’t. If she does, then she needs to take seriously the sorts of intuitions I have proffered. If she is not an ethical intuitionist, then it would seem that a conversation beyond the scope of this current paper is in order, but I see no good reason why it is inappropriate to assume ethical intuitionism in this case.

[7] It might be objected on religious grounds that killing some persons allows them to experience the Beatific Vision, or some sort of pleasant afterlife. However, it seems to me absurd to suggest that this provides grounds for permissibly killing “Heaven candidates,” since, if it were impermissible to kill Heaven candidates, one could permissibly walk into a church and slaughter people – a highly implausible position, to say the least. Additionally, the means taken to get there also (likely) violate PE, and thus even the necessary means to killing them is a problem.

[8] The reader ought to remember that by ‘well-being’ I have in mind ‘general health.’ Surely killing someone does not maintain or improve her general health. In fact, it does just the opposite: It takes away health entirely. For example, suppose I turn in a paper required for a course I am taking, which my professor immediately returns, informing me that my paper is in bad shape. Suppose in order to improve the paper, I burn it to pieces in front of him, justifying my actions in the following way: You said the paper was bad; I just got rid of the bad, and that makes things better.” Call it a gut instinct, but I gather that my professor wouldn’t consider this to be an improvement of the paper since there isn’t a paper any longer. The same seems to hold true in the case of killing, even in cases of ‘mercy killing.’

[9] Lewis, C.S. “Why I am Not a Pacifist,” in The Weight of Glory (New York, NY: HarperCollins Publishers), 2001, p.76. I use Lewis primarily since his critique of pacifism is paradigmatic and is well-worded.

[10] Ibid, pp.76-77.

[11] Nagel, Thomas. “War and Massacre,” in The Morality of War: Classical and Contemporary Readings (Upper Saddle River, NJ: Pearson Education), 2006, p.233. In fact, Nagel is contrasting consequentialist and absolutist treatments of war, defending the latter and attempting to provide a means of escape from Lewis-type dilemmas without having PE specifically in mind. Nevertheless, his solution to Lewis-type dilemmas apply to PE since PE is absolute insofar as it is a moral principle.

[12] I will offer one example as to why I think that is the case. Suppose I claimed that human persons, like us, have a duty to think of every possible number at once. This is not a task we could complete (or even conceive of completing), and so it seems implausible to think – on those grounds alone – that we could have such a duty. Against this, some have argued that the following examples suggest otherwise: If I promise a friend that I will meet her for lunch at noon in Las Vegas, but am (unfortunately) on a delayed flight hovering over the city, it still seems true that I ought to meet my friend at noon even though I cannot, and thus ought needn’t imply can. There is much that can be said here, but I will say only this for now: It is far clearer to me that I don’t have a duty to think of every possible number at once (on the grounds of ‘ought implies can’) than it is that I ought to meet my friend for lunch at noon in Las Vegas when I can’t.

[13] Shawn Graves has pressed me on these grounds in conversation.

[14] And I appreciate immense critical comments here from Timothy Cuffman and Katy Hawkins, who prompted me in this direction.

[15] Psychologists have produced evidence to suggest that the effects of corporal punishment are potentially quite harmful. See, for instance, the following articles: Ohene, S., Ireland, M., McNeely, C., & Borowsky, I. W. “Parental expectations, physical punishment, and violence among adolescents who score positive on a psychosocial screening test in primary care,” in Pediatrics, 117 (2006), pp.441-447; Katz, L. F., & Windecker-Nelson, B. “Domestic violence, emotion coaching, and child adjustment,” in Journal of Family Psychology, 20 (2006), pp.56-67.

[16] It is possible to see chemotherapy as a reason for further worry, since the following sort of example seems also to be impermissible assuming PE: Someone is about to be hit by a bus, so I push him out of the road and onto the ground, knowing that he will probably scrape his arm. The worry is that my actions fail to maintain or improve the person’s well-being, since my action leads to the scrape. This is a poor objection on the grounds that it is implausible to think that a scraped arm constitutes a subtraction from well-being. The implausibility lies in the counterintuitive nature of certain statements (e.g., “That flu shot took away my general health!”) and the failure of scrape-like pains to violate plausible criteria for what constitutes well-being, such as psychological health or a general state of painlessness.