Evil in an indifferent universe.
What is "evil"? Samuel H. Pillsbury titles his book Judging Evil to make an old fashioned (but never out of date) argument--that we should punish evil because it is a good thing to do, indeed that it is the only right thing to do. But can we identify "evil"? (1) In search of an answer, Pillsbury begins the book with the story of a death penalty case he covered as a reporter. (2) Ernest John Dobbert, Jr. "was convicted of the terrible abuse of his four children, abuse so severe that it killed two of them and left another battered and nearly blind." [vii] After hearing the abuse described by the prosecutor, Pillsbury rejected what he calls his "fashionably left" 1960s views to conclude that Dobbert was evil and deserved punishment.
But then the veteran defense lawyer told him the other side of the story: Dobbert was the son of "an abusive father who beat and degraded him." [Id.] The other side of the story suggested that Dobbert was "caught in a cycle of abuse far beyond his power to control. In a strange way, Dobbert loved his children. In prison he had shown himself to be a peaceful, even kind person." [vii-viii]
Pillsbury asks: "Given all this, how could we rage at him? What did he really deserve, this victim of abuse turned victimizer?" That ultimately is the question at the heart of Pillsbury's book. Let us assume for the sake of argument that Dobbert could not have done otherwise, that he lacked the capacity to do anything but continue the pattern of abuse he experienced from his father, who experienced it from his father, and so on. How can we justify imposing terrible criminal penalties on this man who, in prison, was "peaceful, even kind"? One way to describe this view, embraced for example by Peter Arenella, is that we should not punish those who lack the moral capacity to do otherwise than what they did. (3)
Pillsbury left journalism for law school and later clerked for a federal judge. One day while he was "in the elegant chambers office lined with bound reports of federal cases," he heard that Florida had executed Dobbert. He "felt hollow and sick." [viii] Dobbert had committed a horrible crime against defenseless children who depended on him to protect them, but did he deserve death? What if he had no choice but to do what he did?
This essay takes up that question, drawing on the first three chapters in Judging Evil and locating Pillsbury's account in the philosophical literature that finds moral responsibility compatible with causal determinism. The second part of the essay will explore a few of the specific applications of his theory to the law of homicide.
I Free Acts and Determinism: Moral Values and Indifferent Universes
Criminal punishment is one of the most coercive and harmful actions that government can take toward individuals. It must, therefore, be justified by a fundamental interest that society has a right to advance. Conventional wisdom divides these interests into utilitarian and retributive categories. Although each category has numerous variants, for the purposes of my essay, it is important only to understand the basic philosophic divide between the two major categories. That divide can be expressed as just deserts versus consequentialism, retribution versus utilitarianism. Do we give actors the punishment (and praise) that they deserve, or do we decide what to punish (and praise) based on the overall consequences to society?
Pillsbury uses "value approach" to describe the just deserts theory of punishment, and his locution here is meaningful. He wants to remind readers that just deserts values persons for their "choosing abilities." He explains:
Under a Kantian approach to punishment, just punishment is what a person deserves according to the individual's choice to disrespect another. Deserved punishment for homicide respects autonomy in two ways: it respects the autonomy of the victim that was wrongly destroyed, and it respects the autonomy of the killer by taking seriously his or her choice to kill. 
By contrast, the 1960s medical model, with its premise that we should attempt to reform and cure deviance, devalues the individual. This was precisely the point to Anthony Burgess's novel, A Clockwork Orange. Though Pillsbury's exposition of his "value approach" is illuminating, the terminology is a little flat--I have to keep reminding myself what it means--and I will sometimes use "retribution" or "just deserts" in its place.
Just deserts is by far the oldest justification for punishment (the Biblical command "an eye for an eye" manifested this theory over two thousand years ago). Utilitarianism, on the other hand, is a modernist justification that does not require that we value individuals or respect their choices. Thus, it can justify punishment on actors whose capacity to avoid causing harm seems limited at best. Jeremy Bentham was a leading advocate of utilitarianism. From his principle that punishment should be imposed only "in as far as it promises to exclude some greater evil," Bentham derived a corollary: it should not be imposed in cases where "it is needless: where the mischief may be prevented, or cease of itself, without it." (4)
If we believe that Dobbert might turn against his surviving children (or manifest his rage against anyone), we can satisfy Bentham's concern that we are excluding a greater evil by keeping him in prison for the rest of his life. To be certain that he will not torture and kill more children renders his incapacitation far from needless. Indeed, though the death penalty is once again controversial in the United States, if Dobbert might escape or be paroled from prison, one might make a convincing argument in favor of the death penalty on the consequentialist ground of protecting others.
And it is not just Dobbert that the utilitarian worries about. It is all the potential Dobberts out there, others whose rage might turn against their helpless children. Although we know almost nothing about the in terrorem effect of punishment on particular kinds of offenses, and although the kind of offense Dobbert committed seems likely to be extremely difficult to deter in others, the utilitarian is committed to the notion that we must try to discourage crime. Our intuition tells us that, at the margin, fear of punishment ought to have a deterrent effect. Perhaps the parent who feels this level of rage, and has a rational fear of prison or the electric chair, will find help or other ways to deal with the rage. Perhaps.
The utilitarian does not speak of evil. She speaks of balancing harm and benefit. Dobbert might be evil, or he might not. Evil might exist, or it might not. None of these fundamental questions troubles the utilitarian. She is a social engineer. Give her the raw materials, and she will build the most efficient society, one in which the greatest good is bestowed on as many as possible. In that world, Dobbert must be kept away from children, and potential Dobberts must know that severe punishment awaits the acting out of their rage against their children. We need not worry about whether Dobbert chose to do harm. Nor need we worry about whether he deserves life in prison.
The retributivist does not seek to benefit others by punishing Dobbert. Quite the contrary. Kant famously remarked that "one man ought never to be dealt with merely as a means subservient to the purpose of others." (5) He condemned utilitarianism in the strongest of terms: "woe to him who creeps through the serpent-windings of utilitarianism to discover some advantage that may discharge him from the justice of punishment." (6) The only issue the retributivist cares about is whether Dobbert deserves punishment for what he did. Assuming that Dobbert could not have done otherwise, we must at least pause over the question of whether he deserves punishment.
If we are simply the products of genetics and environment, if our actions are caused by forces outside us just as surely as subatomic particles obey the principles of Newtonian physics and quantum mechanics, then how can we speak of "deserved punishment"? We do not blame or praise an electron for an unstable orbit. We do not blame comets for crashing into the earth. We do not blame black holes for sucking matter past the event horizon and rendering its fate unknowable. Why should we blame a person like Ernest John Dobbert, Jr. in the absence of evidence that he could have chosen not to kill his children? If we had perfect knowledge of the relevant facts in the universe, we might discover that Dobbert could not have acted otherwise in any of the decisions that led up to the awful crime--that he was bound to marry, to have children, and to repeat the awful abuse that had been visited upon him. Stated more generally, it might be "that for any given time, a complete statement of the hard facts (7) about the world at that time, together with a complete statement of the laws of nature, entails every truth as to what happens after that time." (8) And if everything that Dobbert did to his children was a truth already entailed in the universe, how can we justify his punishment as a just desert?
There have been basically four responses to the determinism challenge. Kant's is the most complete, simply asserting that punishment is a categorical imperative that society owes itself and offenders. One of Kant's hypotheticals memorably captures this notion. A desert island community has decided to leave the island for good, and the question is what must be done with the murderers in prison awaiting execution. They could be left on the island, posing no further harm to anyone but themselves (Kant assumes). But this would not satisfy Kant's categorical imperative. Instead, "the last murderer lying in the prison ought to be executed" before the community disbands: "This ought to be done in order that every one may realize the desert of his deeds, and that blood-guiltiness may not remain upon the people; for otherwise they might all be regarded as participators in the murder as a public violation of justice." (9)
It is no longer fashionable to speak of punishment as a categorical imperative. An alternative response to the determinism challenge is simply to deny that scientific determinism affects humans at all. Perhaps all of our choices and acts, including those that manifest our obsessions and psychoses, are free in the robust sense that they originated within ourselves. If this dubious principle is true, of course, then there is no difficulty with a value approach based on just deserts, and the Dobberts of the world should be viewed as making robust choices in the same way that I chose to write this essay or that you chose to pick it up and read this far.
Modern developments in psychology, biology, and philosophy, however, make most theorists uncomfortable with denying a role for determinism in our lives. Indeed, I am far from certain that my choice to write this essay was completely free. Pillsbury is a friend of mine, for example, and I read part of his book in manuscript form. Moreover, I sent him a book of mine hoping he would review it. (10) So even what might appear to be an exercise of robust free will can be seen as the result of a complex web of forces, at least some of which are not within my control. Perhaps, then, the ultimate decision to write the essay was not mine in any meaningful sense. Perhaps the hard facts about the world at the point in time when I decided to write the essay entailed the truth that I would write the essay.
Most theorists, however, want to retain the notion of free will for those decisions that are not obviously caused by forces outside of us. This is in no small part because it seems intuitively right. I believe I made a choice to write this essay even if my choice was constrained to some degree. I could have not written it (I believe). Moreover, to deny any role for human will means that we could not speak coherently of moral responsibility. We could not blame or praise. We could not punish because offenders deserved it.
A compromise view has thus emerged. One version of the compromise is that forces outside our control influence many of our actions--I was predisposed to write this essay for the reasons set out above--but the choice remains with the individual unless an observer can point to outside factors that deprived her of control. The observer might find evidence of my obsession to please my friends and to keep my word. Lacking that evidence, the choice to write the essay was mine. On this view, there are free acts and unfree acts. Writing essays is (normally) a free act while obsessive hand washing is not. Perhaps the alcoholic can decide whether to take the first drink, even if he cannot resist the temptation to take a second drink. Perhaps Dobbert had no choice but to see his children as targets of his rage and frustration. But surely when Dobbert tortured his children, when he killed them, he could have acted otherwise. Surely.
Pillsbury labels this argument "selective causation" and rejects it. If we assume that genetic and environmental factors cause us to commit certain acts, how can we reject their role in causing us to commit every act? One cannot draw an arbitrary line at particular acts and say that they, but not others, are caused by forces outside our control. Presumably, the same genetic and environmental forces that lead us to engage in obsessive hand washing also cause us to pay our taxes and vote Republican. To use Pillsbury's example, if genetic and environmental factors deprive the alcoholic of choice, it must also be true that genetic and environmental factors cause someone else to be a moderate drink or an abstainer. "The only difference in causation between moderate and excessive drinkers is what genetics and environment predispose for." 
Nor is free will demonstrated by the success some alcoholics have in learning to avoid use of alcohol. Those successes, just as surely as the failures, could be the product of one's genes and environment. Indeed, if one knew enough about the family and upbringing of a particular person who is attending Alcoholics Anonymous for the first time, one might very well be able to predict with certainty whether she will be one of the program's successes or one of its failures. This is what criminologists try to do when they seek to predict future criminality. Though these predictions are far from perfect, research indicates that they do considerably better than random chance. (11)
To reject the easy compromise of "selective causation" is a powerful insight that makes the problem of identifying free acts more difficult to solve. If "selective causation" is incoherent, theorists must choose between determinism and free actions. (12) Pillsbury does not shy away from the challenge posed by determinism: he wants to make a case for first-degree murder as Dobbert's deserved punishment all the while assuming determinism to be true. This is a difficult challenge. Does he succeed? Robert Nozick has characterized the free will/determinism problem as "the most frustrating and unyielding of problems," adding that perhaps the best we can do is to "surround" the problem." (13)
One way to "surround" the problem is to seek an account in which freedom and responsibility are not necessarily incompatible with causal determinism. Showing a theoretical compatibility between acting freely and determinism does not, of course, prove that we act freely, but it keeps us from having to concede the impossibility of free acts in a causally determined world. It permits us to argue that even if causal determinism is true, we can ascribe moral responsibility to human acts. Harry Frankfurt and Gary Watson have offered compatibilist accounts that depend, roughly, on the actor's attitude toward his acts. Frankfurt claims that volition is a higher order desire and that one acts unfreely when one acts against a higher order volition. (14) How might this happen? That one has a volition not to do X does not foreclose a more powerful first order desire to do X. Thus, on Frankfurt's account, one acts unfreely when acting on a lower order desire that is inconsistent with a higher order volition. An example he gives is of two drug addicts who cannot resist taking the drug. One is happy to be an addict while the other wants desperately not to be addicted. Both have the same first order desire for the drug, but the happy addict acts freely, because acting in accord with his second order volition, while the unhappy addict acts unfreely.
Both drug addicts are, of course, equally powerless to resist using the drug. The outcome that they will use drugs is, therefore, determined. But determinism is beside the point in Frankfurt's account. One acts freely when one acts in accord with a second order volition even if one could not have done otherwise. To deal with the problem of how desire turns into volition, of how the hierarchy gets created, Frankfurt supposes that volition comes into being when one decisively identifies with a first order desire. (15)
Watson's account is, for my purposes, quite similar. He argues that an action is unfree when the actor is unable to get what he most values. (16) He draws a distinction between desiring actions and valuing actions. (17) Suppose C values celibacy more highly than sexual satisfaction. Yet C may nonetheless have sex because her desire for sex is stronger than her want for what she values. In that case, Watson argues, C acts unfreely. If, however, she refuses to have sex because her want for what she values is greater than her desire for sex, she acts freely. Notice that whether the person acts freely or unfreely in Watson's account, as in Frankfurt's account, is completely independent of causal determinism. Given a complete statement of the facts about the world at time x, and a complete statement of the laws of nature, it might be that at time x it was already determined that C would, or would not, have sex. But we could say that she acted freely if she acted in harmony with what she valued.
We do not know enough about Dobbert (or any real life person, with the possible exception of ourselves) to know whether he valued the health and lives of his children more than releasing his rage, or whether his wish to protect his children was a higher order of volition than his desire to manifest rage. But we are told that he loved his children, albeit in a strange way. Thus, it is plausible to conclude that he valued his children more than releasing his rage and that his wish to protect them was a higher order volition than his desire to manifest rage. If so, on Watson's and Frankfurt's accounts, Dobbert acted unfreely when he tortured and killed his children. One of the attributes of these accounts is that they accord with our intuitions. We want to believe, I think, that when Dobbert did those acts, he was, in some sense, acting unfreely. It makes us feel better about the human condition.
The core idea of Watson's and Frankfurt's compatibilist accounts is that acting freely is not exhausted by the notion of being able to have acted differently. Frankfurt asks: "Why should a person's freedom with respect to the performance of one action [the act he does] be thought to have anything to do with his freedom with respect to another [that he was not capable of doing]?" (18) This insight is consistent with Pillsbury's enterprise. What Pillsbury wants is an account of free choice that explains why choice is meaningful. Both Watson and Frankfurt make choice meaningful even if it is not the robust choice to do or not do an act. The choice that lies at the heart of the Frankfurt-Watson approach is the choice to act in harmony with one's values or one's second order volition. That the result of the choice is already fixed by causal determinism, that the chooser may not succeed in getting what she values, does not diminish the sense in which the actor perceives that she is choosing. Indeed, Frankfurt argues that a person can be morally responsible for "choices" that have been predetermined by a neurological device implanted in the brain. This person can "choose" to act in conformity with her second order volition (or, on Watson's account, her values). Even if she fails to effectuate that choice, and was destined to fail, she could still perceive her act of choosing as meaningful.
Pillsbury makes a similar argument. He draws on Peter Strawson's observation that the classic free will/determinism debate simply asks the wrong question: "Strawson maintained that we should not worry about proving the existence of free will ... because this assumes the possibility of its negation. In fact, we cannot live without it."  According to Strawson, we form "reactive attitudes" toward the actions of others that play "a critical part in our moral judgments" and are "fundamental to human nature." [Id.] Strawson "noted that these reactive attitudes are so basic to our condition that they stand independent of the truth or falsity of determinism or free will." [Id.]
But why, Pillsbury asks, is choice so fundamental? He argues that the "answer is strikingly simple. Choice gives meaning to our lives." [Id.] This argument
shows how we can, and should, construct moral value in a morally indifferent universe.... We value choice for our own, deeply important reasons. We value choice by moral responsibility--by caring for all choosers and by taking their choices seriously. In rewarding good choices and punishing bad ones, we reach beyond physical self-interest to achieve greater meaning in life. In that process we may find this meaning or we may not, but we have no hope of finding it unless we value choice. 
This is not a proof that we act freely but a moral theory that explains why compatibilism is an attractive solution to the free action problem. We assign praise and blame to non-coerced choices because it is the way humans construct meaning. It permits us to connect ourselves with a larger "something" outside ourselves--the values that give our lives meaning. It might be that no human could do other than she does, just like no electron could do other than to obey quantum mechanics, but the experience of "choosing" and praising and blaming others for the "choices" they make permits us to experience life as humans rather than electrons. Though Frankfurt and Watson do not explicitly adopt this experiential viewpoint, (19) and Pillsbury does not mention their work, an account of choosing based on one's values or second order volition is fully consistent with Pillsbury's experiential explanation of the role of choice.
The experiential side of being human can neither be proven nor denied. A wedding, to use Pillsbury's example, can be explained scientifically by our need to reproduce ourselves, to keep the family together, and to bring as many children as possible to adulthood so that they can reproduce and continue the cycle. But Pillsbury argues that this
hardly exhausts what we know of weddings and marriage.... Most importantly, it says nothing of love. Love is the value on which the marriage ceremony is based. Love is enormously meaningful to humans, but understanding its meaning requires adopting the human perspective. It means taking the subjective experience of human life seriously. 
He draws a distinction between predicting human acts--the scientific side--and the moral expectations we create by making judgments about choices: "The fact that we can predict human behavior of a particular kind--for example, that psychopaths will hurt others unless stopped--does not mean that the behavior is nonresponsible. Prediction tracks physical causation; expectation constitutes our creative efforts to make the world a better place." [42-43]
To keep true to Pillsbury's premise that accepts determinism, the "creative efforts to make the world a better place" must refer to the way we experience choice. We experience choice, on Pillsbury's account, as creating meaning, as making us human. I think this a nice insight. It tells us a lot about why we have the moral philosophy that we (generally) share today and why we speak so easily in terms of praise and blame. It tells us why the death penalty is difficult to abolish. If we are certain that Dobbert is the one who tortured and killed his children, and that he was not coerced or insane, there is a sense in which even abolitionists understand that Dobbert deserves to die, though we might have other good and sufficient reasons not to impose the penalty.
Pillsbury's insight also helps explain why criminal laws vary across cultures. Though every human on the planet presumably shares the perspective in which choice is experienced as meaningful, different cultures might understand differently what constitutes choice. In some parts of Africa, young women "choose" to have circumcision. In the West, we have difficulty believing that this is a choice, but in the particular culture, it might be seen as a freely chosen act, just as freely chosen as our choices to change jobs or retire. In the United States, we tend to see the choice of a drug addict to continue to take drugs as a robust free will choice (despite what our medical experts might tell us). In some other countries, drug addiction seems to be viewed more as a medical problem. Indeed, Pillsbury's very example of marriage entailing choice based on love is itself culturally contingent. In many countries, marriages are arranged, choice is limited at best, and love is peripheral.
What does all this have to do with whether we should punish Dobbert because he deserves it? Pillsbury says that in punishing Dobbert the "state, acting for society generally, sought to defend the possibility of moral meaning based on autonomy."  This claim makes me nervous. Given determinism, how can there be any possibility of "moral meaning based on autonomy"? Perhaps Pillsbury is saying that we simply delude ourselves, that we think choice important but in each case we are engaged in a kind of elaborate philosophical charade because nothing we do or say or think is chosen by us. If our moral values, like a hydrogen atom, are constructed without assistance from us, if our entire world and all of its cultures are, like a black hole in the center of a nebula, an unchosen part of the universe, in what sense can we seek to defend the "possibility of moral meaning based on autonomy"?
But Pillsbury is not unmasking a charade in which we pretend that choice creates moral meaning. Instead, I think he is saying that the punishment of Dobbert was not futile because it served the purpose of allowing us to connect choice to values and to experience the world as a place in which choice is central. (20) The ultimate question is not causation at all but, rather, what kind of society we wish to inhabit. Do we wish to inhabit one in which acts done by Dobbert "would not be the wrongful actions of responsible persons, but akin to the dangerous actions of the crazy, infants, or animals"?  Or do we wish to inhabit a world in which Dobbert's acts are described as wrongs for which he was morally responsible, however little capacity he had to do otherwise?
So when Pillsbury talks about the state punishing Dobbert as a way of defending the "possibility of moral meaning based on autonomy," I take him to be making the following claim. It is wrong to torture and murder children; the moral meaning underlying that value is constructed when we punish the act of torturing and murdering children, just as surely as it is constructed when we praise parents for taking care of their children. We do not have to show that Dobbert could have done otherwise to hold him responsible for the horrible acts he committed. (21)
This is both profound and, upon reflection, obvious. Think about the value of choice in our lives. If my best friend does not return my phone call or fails to meet me as arranged, I will assume that she chose not to call or meet me. I make a judgment about her based on that assumption. Of course, my assumption might be wrong. I might learn that the phone message never got through or that she was in an automobile accident on the way to the meeting place. On these new facts, I would reevaluate my judgment of her "choice" not to phone or meet. But a later message that she was just too busy to phone or meet me is unlikely to make much difference. I would still view her as having made a choice about how best to spend her time, and if I am not important enough to be placed somewhere near the top of her list, I might rethink whether she is really my best friend.
We make these judgments every day, all the time, in a wide variety of social and moral situations. We made a judgment about Bill Clinton's choice to permit Monica Lewinsky to perform oral sex on him in the Oval Office. Indeed, many believe that the principal reason Gore chose Joe Lieberman for his vice presidential running mate was that he stood in the well of the Senate and condemned Clinton for making that choice. If we are constantly praising and blaming choices in our everyday lives, it should come as no surprise that we do it in making decisions about imposing criminal punishment. If determinism is correct, however, my best friend had no more choice in not calling or meeting me than Dobbert did in torturing his children.
That world is just too cold, too indifferent. So, faced with scientific determinism, we retreat to compatibilism. That means, of course, that we can imprison for life or execute actors without knowing whether they, or any of us, could do otherwise. But I am persuaded by Pillsbury, Watson, and Frankfurt that we can ascribe moral responsibility to an actor without knowing that he could have done otherwise. I am far from certain that Dobbert was acting freely in the sense that Watson and Frankfurt use that concept, but that is merely quibbling around the edges. If we can punish actors who could not have done otherwise, deciding which ones were acting freely is merely a sorting problem.
If Dobbert was not acting freely, then Watson and Frankfurt would conclude that he was not morally responsible. This seems intuitively right. Of course, we can still punish Dobbert on a utilitarian rationale. What we cannot do, I believe, is to claim a retributive basis for saying that he deserved the punishment. Perhaps Pillsbury wishes to claim a retributive justification even if he agreed that Dobbert was not acting freely in the Frankfurt-Watson sense. Perhaps Pillsbury wishes to show us another way to justify punishing the Dobberts of the world. To punish them permits us to experience a life that includes choice and thus moral meaning.
But if this is the best understanding of Pillsbury's claim, it is oddly connected to just deserts. Indeed, it seems utilitarian. We will punish Dobbert even if he was not acting freely, even if he did not perceive himself as acting freely, so that the rest of us can construct moral meaning. That understanding is, of course, antithetical to the Kantian view that we can never use one person solely for the benefit of others. But however that dimension of Pillsbury's account is understood, the fundamental point is that Pillsbury, Watson, and Frankfurt have constructed plausible accounts where we can say that Dobbert deserved punishment even if he could not have done otherwise. We can be retributivists in a causally determined world.
II Choice and Moral Value: The Law of Homicide
The second half of Judging Evil applies Pillsbury's value approach to discrete questions in the law of criminal homicide. His value approach has a beneficial effect in understanding some of the stresses that homicide law has endured. It suggests a few changes that are novel and makes a good case for these changes. When all is said and done, Pillsbury's model homicide code is an improvement over the typical American homicide statute based on common law doctrine.
Pillsbury's retributivist principles suggest that we should blame a person for choosing to do evil and that doing evil is an attack on the value of another person. (22) Seeing homicide law through this lens clarifies much. Pillsbury joins a long line of theorists who reject the felony murder rule (which its country of origin, England, rejected by statute almost half a century ago).  Under the classic version of the felony murder rule, an offender who accidentally kills during the perpetration of a dangerous felony is automatically guilty of first-degree murder. What is wrong with this rule, using Pillsbury's lens, is that there is no reason to blame the offender for accidentally killing someone, nor can an accidental killing be an attack on the value of another person. To be sure, the offender engaged in a dangerous felony, creating a risk of death. But armed robbery is punished more severely than robbery which is punished more severely than larceny precisely because of the escalating risk of death. To upgrade an accidental killing into first-degree murder is seriously overcounting under a value approach because there is no blame for the killing that exists independently of the underlying felony. If the felon negligently, recklessly, or purposely kills during the course of committing a felony, the proper punishment would be whatever grade of criminal homicide is manifested by the particular mens rea.
Those who hold a utilitarian view of criminal law are probably not as offended by felony murder as are the retributivists. If most killings that occur during a dangerous felony are at least reckless, then prosecutors can typically charge the killing as murder based on "extreme indifference to the value of human life." (23) It would be second degree murder, in all likelihood, but that is a very serous offense. Thus, leaving aside the death penalty issue (which Pillsbury also explicitly leaves aside), most dangerous felons who kill can be convicted of murder with or without the felony murder rule. The felony murder rule, of course, is usually an easier route to a murder conviction because the prosecutor does not have to prove the reckless mens tea and the extreme indifference to the value of human life. For the utilitarian, the use of the felony murder rule to achieve roughly the same results achievable by a more discerning inquiry is probably good enough reason to leave the felony murder rule in place.
Some killings during the commission of a felony will be accidental, but one who robs and kills, even accidentally, could be deserving of a longer period of incapacitation under some forms of utilitarian analysis than one who robs and does not kill. Moreover, we want to deter dangerous felonies, and an occasional murder conviction might deter better than long sentences for armed robbery. Of course, the felony murder rule is not necessary to achieve this presumed marginal deterrence. Why not simply execute, at random, one of every fifty or one hundred armed robbers as a way of achieving greater deterrence of robbery? That is roughly the effect of the felony murder rule and with roughly as much justification. But whether based on the perceived need for additional deterrence or the perception that those who kill even accidentally during felonies are more dangerous, utilitarianism might permit punishing an occasional accidental killing as felony murder. (24) That the United States continues to use felony murder suggests that, on this issue at least, legislators are utilitarians.
Related to his critique of felony murder grading, Pillsbury points out that the use of premeditation as a marker for the most serious murders also misses the retributive mark some of the time. Premeditation is sometimes a good marker for the most blameworthy killings--as in contract killings or killing a judge or district attorney--but it is not inevitably true, nor is it true that unpremeditated killings always lack the most serious blameworthiness. Consider a mercy killing in which the victim was going to die within days, days that were going to be filled with unendurable pain.[105-06] (25) This is first-degree murder in jurisdictions that use premeditation as one marker for the most culpable killings. Though a mercy killing is criminal homicide in this country, making it first degree murder seems a bad result when the choice to kill does not attack, but rather affirms, the value of the victim. Pillsbury contrasts mercy killing with a California case in which the defendant stabbed a child over sixty times in a drunken sexual frenzy that did not suggest premeditation. [104-05] (26) But surely this killing should be first-degree murder even if the prosecutor cannot prove premeditation.
He solves both of these grading problems with a revised statutory definition of the most serious type of murder: "the purposeful killing of another human being for profit, to further a criminal endeavor, to affect public policy or legal processes, because of animosity toward the victim's race, religion, ethnicity, sex or sexual orientation, or to assert cruel power over another."  Though I would delete the last phrase because of vagueness problems, I think this statute a decided improvement over the typical murder statute that uses premeditation and felony murder as the major devices to identify the most blameworthy killings. Notice how Pillsbury has preserved the core of premeditation and felony murder without all the conceptually unsatisfying bag and baggage. We would still grade a contract killing, a purposeful killing while committing another crime, or a purposeful killing of a judge or district attorney as the most serious form of murder. The purposeful killing of the child in the California case would be included because there was evidence that the defendant killed her while furthering a sexual crime (or if the "cruel power" phrase remains, under that head as well). But the coolly premeditated mercy killing would not be first-degree murder. This is an intuitively satisfying reversal of the actual outcomes in these cases.
The application of Pillsbury's value approach to the hoary doctrine of provocation also produces insights. At the common law, courts sometimes held that a murder was sufficiently a crime of "hot blood" or passion that it should be reduced to voluntary manslaughter. This doctrine seems to have originated in killings during mutual combat that followed a quarrel. This grew into a doctrine that, for many years, permitted a defendant to avoid a murder conviction when the killing followed his discovery of his wife having sex with another man. At least in its early form, a provocation defense was not available to women who killed their husbands.  Nor was it available to a man who discovered his lover in the arms of another. (27)
Provocation has recently received considerable scholarly attention. (28) Feminists claim that even ill its modern form (now permitting women also to use it as a defense) it manifests gender bias. Other critics have claimed that it promotes and rewards violence; after all, as Stephen Morse has noted, it is not that hard to refrain from killing even when one has been provoked. (29) Pillsbury notes the additional problem that class and race bias may inform how courts decide what constitutes provocation. Other scholars have worried about how best to state the rationale for the doctrine. Is a provoked killing partially justified because of the wrongdoing on the part of the victim? Or is a provoked killing partially excused because of the passion that has overtaken the killer? (30)
Pillsbury joins the scholars who conclude that provocation is based both on justification and excuse. He also agrees with the gender, class, and race bias criticism. He agrees to some extent with the abolitionists that it rewards and promotes violence. But he would retain the core of the provocation doctrine by returning to a version of its original form, a response to serious physical assault. At least this is the way I read the section in which he suggests that the paradigm case of voluntary manslaughter should be a killing provoked by serious physical assault. [159-60] This retention and limitation of provocation makes a good deal of sense. It avoids much of the gender, class, and race bias. What constitutes a serious physical assault would not seem to vary much by sex, class, or race. Limiting the claim to cases of a serious physical assault limits the category of cases in which the claim can be raised and thus limits the extent to which the defense weakens the law's preference for non-violence. Finally, remember that Pillsbury's project is about blaming the actor who chooses evil and who attacks the value of another person. It is easier to understand how someone might lose control and kill after being the victim of a serious physical assault. This person consequently should be blamed less and has made less of an attack on the partially responsible victim. The killing is still criminal homicide, reflecting Stephen Morse's view that even severely provoked individuals ought to be able to resist the urge to kill, but it is manslaughter rather than murder.
Unfortunately, after asserting that serious physical assault is a paradigm for what counts as sufficient provocation, Pillsbury weakens his statement of the rule to include good reason to believe that the victim "committed a serious wrong against the defendant or a loved one." ]192] This undermines the salutary reform implicit in his discussion of the paradigm case. To be sure, his proposed instructions cite examples of "serious harm such as a sexual or other serious, wrongful assault," but even here he goes on to include "[s]ignificant threats to the safety of the defendant or a loved one."  As provocation does not now include threats, and as threats are likely to be quite common in cases in which a defendant would respond with deadly force, Pillsbury actually broadens the provocation defense here, undermining much of what he gains by his physical assault paradigm. But a legislature that adopted his proposal without the "serious wrong" part would achieve a meaningful reform of the law of provocation.
Pillsbury completes his account of homicide law by considering indifferent homicides. One can be indifferent and cause homicide in two ways. The most obvious is when an actor perceives and disregards a risk to the lives of others. This mental state, a form of recklessness, is the minimum mental state required by the Model Penal Code for murder and manslaughter. It requires the state to prove a subjective awareness of the risk to human life. Its disregard is what demonstrates indifference to the value of others. The driver who kills another after taking extreme risks while driving in a rainstorm is an example of this type of indifferent killer.
The second way in which one can be indifferent and cause homicide is when one is not aware of the risk. Under the Model Penal Code, this actor is merely negligent and cannot be guilty of murder or manslaughter no matter how gross a deviation is his failure to perceive the risk. The Code defends this judgment on the ground that murder and manslaughter should be reserved for actors who are subjectively aware of the risk their conduct poses. To fill the gap left by this judgment, the Code creates the lesser offense of "negligent homicide."
Although Pillsbury's account is based on "evil," he does not find subjective awareness of risk to be the proper dividing line between murder and manslaughter, on the one hand, and lesser homicide crimes or torts on the other hand. Indeed, his account of evil includes moral indifference in situations that require accurate perceptions of the risk. Take, for example, the dentist who treats a frail and sick young patient. [173-76] (31) The dentist ignored the advice of the patient's physician not to put her under general anesthesia, he administered a dose far too large for someone her size, and he ignored clear signs of respiratory difficulty during and after the extensive procedure. Under the Model Penal Code, to find the dentist guilty of murder or manslaughter, the jury must find that he was reckless, that he disregarded a substantial and unjustifiable risk that she would die from the procedure.
But why would a dentist proceed with the operation if he perceived the risk? It is professional suicide to have patients die because of a failed medical procedure. Yet Pillsbury's intuition, and mine, is that this sort of professional gross negligence is as culpable as many examples of reckless behavior. To convict of murder or manslaughter under the Model Penal Code, the jury would have to engage in the following fiction--because most dentists would have perceived the risk, he must have perceived it even though it makes no sense to think that he actually did. Pillsbury would replace that fiction with the following assessment of culpability. The dentist was in a position to perceive the risk; he knew of the patient's weakened condition and of her physician's warning not to use a general anesthetic. When he failed to exercise professional judgment to inform himself of the substantial risk to her life, that indifference is not only morally culpable, it is a form of evil because it does not place a sufficiently high value on the life of his patient.
For Pillsbury, indifference to risk can be evil and thus can give rise to serious criminal liability without having to twist the notion of recklessness. Perhaps the best critique one can make of the Model Penal Code here is its approach to intoxication as a defense to criminal culpability. Pillsbury is kind to the Model Penal Code on this one, accusing it only of "awkward moves."  I would accuse the Model Penal Code of betraying its principles to accommodate public perceptions about how we should treat drunks who cause harm.
The Model Penal Code creates a special rule for intoxicated defendants who argue that they lacked the awareness necessary to be guilty of a crime requiring recklessness. (32) The rule is that the jury should consider not the risk of which the intoxicated person was actually aware but, instead, make the counterfactual determination of what risk he would have perceived had he been sober. By substituting the risk perception of the sober defendant, the Model Penal Code effectively makes the decision to get drunk bear the weight of the heightened criminal culpability. A defendant who did not perceive the risk because of any other condition or combination of conditions--the dentist whose rush to treat as many patients as possible blinded him to the risk--would not be guilty of a crime requiring recklessness. The drunk, however, is guilty if he would have perceived the risk while sober. While there is nothing inherently wrong with assigning this much weight to the decision to get drunk, it is an ad hoc exception to the awareness principle upon which the Model Penal Code is otherwise based.
Pillsbury reaches the same result as the Model Penal Code on the intoxicated driver who kills. [176-79] But his account requires no special rule because the indifference that driving while intoxicated manifests is precisely the kind of indifference that can give rise to murder or manslaughter liability, using Pillsbury's account, because it insufficiently values the lives of others. For Pillsbury, the key is indifference to the value of others, whether or not demonstrated by awareness. Indifference is often, but not always, manifested by proceeding in the face of awareness of the risk. In an insightful passage, Pillsbury writes:
Awareness of risk does yeoman service as a culpability test in most cases, because awareness usually serves as a proxy for indifference. In the usual case, facts that indicate awareness of risk also indicate indifference, while facts that negate awareness tend to negate indifference. Intoxication presents the unusual situation where the same facts that negate awareness (the defendant's intoxication) actually support a finding of indifference. Here the law must choose between the two approaches. We should not be surprised that the law plumps for indifference, for indifference should be the moral principle underlying all punishment of careless conduct causing death. 
That seems exactly right to me and is a nice summary of Pillsbury's entire account. The evil about which he writes is the failure to value sufficiently the lives of others. Indifference to the value of others is potentially just as blameworthy as the disregard of a specific risk. The dentist who ignored all he knew about his frail patient while he overmedicated and killed her is just as blameworthy as the actor who realizes that speeding and blindly changing lanes during a rainstorm might result in the death of another driver. Both the dentist and the speeder made a choice (or at least we treat them as if they had made a choice). It is the choice not to value sufficiently the lives of others that gives rise to criminal liability in Pillsbury's account, and the relevant choice can be made with, or without, awareness of the risk.
Evil, on Pillsbury's view, is the failure to value others sufficiently. It is a satisfying account. It tells us much about why we make certain conduct criminal in the first place. It illuminates much about blame and praise. It ultimately reveals some fundamental truths about who we are as a legal community and as a culture. Integrating these insights into a single work, and one that is almost as good a "read" as Pillsbury's novel, (33) is quite an accomplishment.
Thanks to John Douard, Joshua Dressler, Bill Heffernan, and John Kleinig for helpful comments, on this paper.
(1) I wrote this essay early in 2001 when identifying "evil" seemed a nice metaphysical problem rather than an international crisis.
(2) The Dobbert case, including his execution following an appeal to the United States Supreme Court, begins the book, S.H. Pillsbury, Judging Evil vii-viii. [Subsequent bracketed references in the text refer to Pillsbury's book.] One of the delights of Pillsbury's book is his frequent use of real cases to raise difficult conceptual and moral issues. See also Dobbert v. Florida, 432 U.S. 282 (1977).
(3) Arenella, Convicting the Morally Blameless: Reassessing the Relationship Between Legal and Moral Accountability, 39 U.C.L.A. L. REV. 1511 (1992).
(4) J. BENTHAM, AN INTRODUCTION TO THE PRINCIPLES OF MORALS AND LEGISLATION 170 & 171 (1823 edition).
(5) I. KANT, THE PHILOSOPHY OF LAW 195 (W. Hastie trans. 1887).
(7) In discussing free will, philosophers sometimes draw a distinction between "hard" and "soft" facts. This distinction is not critical to my essay and I will offer only a brief account for readers who might be interested. Soft facts are relational to the hard facts that actually happen. Soft facts are thus non-genuine. To illustrate, you go to the movies on Thursday. It must, then, have been true on Wednesday that you would go to the movies on Thursday. But if that is true, then the fatalist argues that you had no choice on Wednesday about whether to go to the movies on Thursday. A response, derived from William of Ockham, is that the statement "on Wednesday it was true that you would go to the movies on Thursday" is a relational (soft) fact about Wednesday and not a hard fact about Thursday. It is true only in a rear-view mirror sense that if you go to the movies on Thursday it will have been true on Wednesday that you would go to the movies on Thursday. Soft facts about Wednesday do not deprive us of choice on Thursday. See Fischer, Introduction: Responsibility and Freedom, in MORAL RESPONSIBILITY at 26 (J.M. Fischer ed. 1986) [hereinafter MORAL RESPONSIBILITY]. Of course, not all facts about Wednesday are soft facts. Suppose, for example, that on Wednesday you read a review of the movie that opens on Thursday. There is nothing that you can do later to change this fact about the universe but, notice, that this fact does not entail any statement that predicts your behavior on Thursday. While the Ockham distinction between hard and soft facts serves to defeat the fatalist, it does not defeat causal determinism. Determinism makes no claim at all about the inevitability of soft facts and, instead, depends on the more elaborate premise set out in the text.
(8) Id. at 33.
(9) I. KANT, supra note 5 at 198.
(10) Pillsbury not only agreed to review my book, but also completed the review long before I finished my review of his book. See Pillsbury, A Rough Country Guide: Double Jeopardy, Doctrine, and Realism, 20 Crim. Just. Ethics 53 (2001) (reviewing G.C. THOMAS III, DOUBLE JEOPARDY: THE HISTORY, THE LAW (1998)).
(11) See, e.g., Wenk, Robison, & Smith, Can Violence Be? Predicted ? 18 CRIME & DELINQUENCY 393 (1972) (developing index of "violence-proneness" that correctly predicted which of a sample of juvenile offenders would engage in future violent behavior at about a 90 percent rate).
(12) To be sure, there are other positions--sophisticated philosophical attempts to find a role for autonomy in a world of scientific determinism. See, e.g., R. NOZICK, PHILOSOPHICAL EXPLANATIONS 291-397 (1981). But Pillsbury writes for those without an "extensive background in philosophy." [xi] In any event, the key question is whether a retributivist account can be squared with determinism, not whether determinism is the best description of human action.
(13) R. NOZICK, supra note 12, at 293.
(14) Frankfurt, Freedom of the Will and the Concept of a Person, in H.G. FRANKFURT, THE IMPORTANCE OF WHAT WE CARE ABOUT, 11-25 (1988).
(15) The identification itself might of course be causally determined. If so, Frankfurt's account is phenomenological, but this does not diminish its explanatory power.
(16) Watson, Free Agency, in MORAL RESPONSIBILITY, supra note 7 at 81-96.
(17) Again, if the distinction between valuing actions and merely desiring actions is itself causally determined, the account is phenomenonolgical. See supra note 15.
(18) Frankfurt, Three Concepts of Free Action: 11, in MORAL RESPONSIBILITY, supra note 7, at 117.
(19) Their accounts might ultimately be phenomenological, however. See supra notes 15 & 17.
(20) This shows that Pillsbury's account is explicitly phenomenological. Frankfurt's and Watson's accounts may be as well. See supra notes 15 & 17.
(21) Later in the paragraph where Pillsbury talks about the " possibility of moral meaning based on autonomy," he characterizes what Dobbert did as "chosen acts."  One way to understand this claim is that it refers to free acts in Watson's or Frankfurt's sense that Dobbert valued or identified with the acts that he did not have a choice about committing.
(22) Pillsbury's value approach might seem to have difficulty explaining criminal laws that do not protect property and personal integrity. Can we say, for example, that violating the law requiring us to pay income tax is an attack on the value of another person? While the argument is not easy, I think it possible. If you do not pay your income tax, I have to pay more. If most Americans do not pay, I have to pay a lot more. Whether or not this is persuasive, Pillsbury limits his focus to homicide and I shall do the same.
(23) This is the Model Penal Code formulation, see MODEL PENAL CODE [section] 210.2(1)(b), but it roughly tracks the common law notion permitting a murder verdict on a mens rea of abandoned and malignant heart. See CALIFORNIA PENAL CODE [section] 188 (codifying this common law approach). One who kills while committing a dangerous felony, say robbery, seems likely to have possessed extreme indifference to the value of human life or an abandoned and malignant heart.
(24) Some utilitarians would object to this on distributional grounds. Many theorists hold to H.L.A. Hart's model, which has a general justifying aim and a principle to control distribution of punishment in individual cases. Thus, one could have incapacitation as the general justifying aim and retribution as a distribution principle, which would require rejecting the felony murder rule when it is applied to an accidental killing.
(25) Discussing State v. Forrest, 363 S.E.2d 252 (N.C. 1987).
(26) Discussing People v. Anderson, 447 P.2d 942 (Cal. 1968).
(27) See, e.g., Rex v. Greening, 3 K.B. 846 (1913).
(28) See S.H. PILLSBURY, supra note 2, at 146-58, reviewing this scholarship.
(29) Morse, Undiminished Confusion in Diminished Capacity, 75 J. CRIM. L. & CRIMINOLOGY 1, 33-34 (1984).
(30) For some thoughts on this debate, see, e.g., Ashworth, The Doctrine of Provocation, 35 CAMB. L.J. 292 (1976); Dressier, Provocation: Partial Justification or Excuse? 51 MOD. L. REV. 467 (1988); Singer, The Resurgence of Mens Rea: I--Provocation, Emotional Disturbance, and the Model Penal Code, 27 B.C.L. REV. 243 (1986). See also authorities cited in S.H. PILLSBURY, supra note 2, at 237-38 notes 36 & 37.
(31) Discussing People v. Protopappas, 201 Cal. App.3d 152 (1988).
(32) See MODEL PENAL CODE [section] 2.08.
(33) S.H. PILLSBURY, CONVICTION (1992).
George C. Thomas, author of Double Jeopardy: The History, the Law, is Professor of Law at Rutgers University.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Judging Evil: Rethinking the Law of Murder and Manslaughter|
|Author:||Thomas, George C., III|
|Publication:||Criminal Justice Ethics|
|Article Type:||Book Review|
|Date:||Jun 22, 2001|
|Previous Article:||Domestic violence and hate crimes: acknowledging two levels of responsibility.|
|Next Article:||Broadening the scope of criminal law scholarship.|