Criminal culpability is always a function of what the actor believes regarding the nature and consequences of his conduct (and the various probabilities thereof) and what the actor’s reasons are for acting as he does in light of those beliefs. Recklessness, minus its substantiality of risk requirement, perfectly expresses these two dimensions of culpabil- ity. In the last part of this chapter, we turn to more specifi c questions about this culpability calculation.
A. UNDERSTANDINGINSUFFICIENTCONCERN
As defended in the preceding sections, we believe that, as a concep- tual matter, purpose and knowledge are simply species of recklessness.
Culpability, at bottom, is just about risks and reasons. Because of the novelty of our view, we wish to respond to some potential (and not so potential) objections.
30 Th e folding of purpose into recklessness also helps with certain “lit fuse” situations.
Suppose, for example, D lights a fuse with the purpose of burning down V’s business. D is obviously reckless the instant he lights it, as there is some chance he will not be able to snuff it out even if he has a change of heart. He has imposed this risk – which increases as time passes – for no good reason, in this case a purpose to destroy. Now suppose D does have a change of heart in time to snuff out the fuse, but he has an epileptic seizure before he can do so, and V’s business burns down. Did D burn it down purposefully, because he lit the fuse with that purpose, or only recklessly, because he had a change of heart? Both answers have their temptations. On our theory, there is no need to decide.
1. How Many Categories Do We Need?
Th e fi rst concern is that our view “fl attens” culpability.31 Rather than recognizing four distinct mental states, we reduce it all to one. How in the world can all the shades of culpability fi t within one mental state?
Our response – that we actually recognize fi ner gradations – seems to throw us into Charybdis, as other critics may worry about unfettered discretion in the hands of juries and prosecutors.32 Th ese objections are quite diff erent, but both certainly cannot be right.
Th e fi rst question, we believe, should be the conceptual one. How many diff erent mental states are there really? If purpose and knowledge require that we look to both risks and reasons, then they function in exactly the same way as does recklessness. Although acting with pur- pose or knowledge regarding forbidden conduct, results, or circum- stances appears presumptively unjustifi ed, nevertheless, as we argue earlier, justifi cations may be available in such cases.
Moreover, despite the fact that our view appears to reduce all culpa- bility to one level, it does quite the opposite. We can recognize shades of gray that current legal formulations cannot. Our view explains why some instances of extreme indiff erence to human life are just as culpa- ble as some purposeful and knowing killings. In addition, our view can accommodate the intuition that harming someone purposefully can be a very bad reason for acting. Th at is, we may think that a person is more culpable for trying to harm someone than she is if she merely foresees and tolerates it. On the other hand, there seem to be many other reasons that may be just as bad as aiming at the evil. If an actor blows up an airplane and kills the pilot, we might think him no more culpable for killing the pilot to marry the pilot’s wife (purpose) than for blowing up the plane for insurance money and killing the pilot as a side eff ect (knowledge).
Identifying with and aiming at evil are extremely culpable, but so, too, is the indiff erence manifested in acting for weak reasons while risking atro- cious harm.
As noted previously, our ability to accommodate these nuances comes at a price; it may aff ord greater discretion to courts and juries.
31 Joshua Dressler, “Does One Size Fit All? Th oughts on Alexander’s Unifi ed Conception of Criminal Culpability,” 88 Cal. L. Rev. 955 (2000).
32 Alexander, supra note 3, at 953 n.62.
Th is may be true, but we doubt this objection is all that forceful. In terms of legitimate error, there may be more errors under our scheme, but they will likely be smaller than when juries are forced to place the actor’s mental state into one of four (supposedly hierarchical) catego- ries. Moreover, because all subjective mental states are notoriously diffi - cult to prove, and all rely on inferences made by juries, we think there is little reason to believe that juries will misuse our unifi ed conception in ways that they cannot already misuse current mental state categories.
Finally, we contend that the law should continue to require that the imbalance of risks and reasons represents a gross deviation from the standard of care that a law-abiding citizen would observe in the actor’s situation.33 Th e criminal law should not be concerned with those actors who, although they impose risks that are not justifi ed by their reasons, are only minimally culpable (because their reasons almost justify the risks they perceive). Moreover, by requiring a gross deviation, the risk of juror error is reduced.
2. Indifference Compared
Some criminal law theorists argue that we should add indiff erence as a distinct type of culpable mental state.34 To us, such proposals turn on what one means by “indiff erence.” 35 Consider the following example.
Danielle decides to play Russian roulette with Andrew. Danielle does not particularly care for Andrew and is wholly equivocal as to whether Andrew is killed. Danielle pulls the trigger, and Andrew dies.
Contrast Darla who plays Russian roulette with Abe. Darla is very fond of Abe; indeed, the two are dating. Yet, Darla and Abe enjoy the rush that comes from playing Russian roulette. Darla pulls the trigger, and Abe dies.
In the fi rst example, Danielle can be said to be indiff erent in two diff erent respects. First, her desire state about Andrew’s fate is one of
33 As suggested by Dressler, supra note 31.
34 See, e.g., Simons, supra note 2; Kenneth W. Simons, “Culpability and Retributive Th eory:
Th e Problem of Criminal Negligence,” 5 J. Contemp. Legal Issues 365, 371 (1994); R. A. Duff , Intention, Agency, and Criminal Liability: Philosophy of Action and the Criminal Law 162–163 (1990).
35 Th is section draws from Kimberly Kessler Ferzan, “Opaque Recklessness,” 91 J. Crim. L.
& Criminology 597 (2001) and Kimberly Kessler Ferzan, “Don’t Abandon the Model Penal Code Yet! Th inking Th rough Simons’s Rethinking,” 6 Buff . Crim. L. Rev. 185 (2002).
indiff erence. She cares not whether he lives or dies. Second, she may be said to be indiff erent because, although faced with imposing the risk of death, she chooses to pull the trigger.
Darla, on the other hand, is indiff erent only in the latter sense.
Th at is, Darla does care, indeed she cares deeply, about Abe. She is not indiff erent toward his death. Nevertheless, we may still say that her choice, to play Russian roulette, manifests culpable indiff erence to human life.
Th e fi rst meaning of indiff erence refl ects attitudes that the actor has irrespective of how she chooses. Th e latter sense represents a normative evaluation of the actor’s choice. For the normative sense of indiff erence, we do not care about how the actor feels when we label her indiff erent.
Rather, no matter what she may wish, hope, or desire, we may decide that she does not care enough.
We are not opposed to the term “indiff erence” in the latter sense.
Whether we employ the term “recklessness,” “indiff erence,” or “insuf- fi cient concern,” all three seem to encompass the disrespect for oth- ers that makes reckless conduct culpable. To us, culpable indiff erence is exhibited by the choice to engage in reckless conduct, the willing- ness to risk the bad side eff ects of one’s action for insuffi cient reasons.
It is the outcome of the actor’s practical reasoning that is problematic.
For us, all recklessness exhibits culpable indiff erence, and some reckless choices, because of the degree of their unjustifi ability, exhibit extreme indiff erence.
On the other hand, we deny that indiff erence as an attitude is itself suffi cient for culpability. First, we are opposed to this conception of indiff erence insofar as it might license punishment in the absence of conscious choice, as we discuss in our argument against punishment for negligence in the next chapter. But we are likewise unconvinced that this approach places responsibility on the correct aspect of the actor’s practi- cal reasoning. For example, assume David runs the red light to get to the Knicks game, and he recognizes that there is a substantial risk of harm to others. But his desire to go to the game is so great that he decides to run the light anyway. We believe that David is reckless because, although he appreciates the substantiality and unjustifi ability of the risk, he chooses to disregard it. But in deciding to disregard the risk of harm to others, David’s practical reasoning might go a variety of ways: David might value
his desire to go the Knicks game at 100 and the potential harm to others at 10 (in terms of his desire to avoid it). Or he might value the Knicks game at 10, but the harm to others at 9. We contend that it is the choice, to pick the game over others’ interests, whatever values David gives these variables, that makes David’s conduct culpable.
However, we believe that an “indiff erence as attitude” proponent could be placing blame earlier in the calculation – to the precise amount of weight given to the interests of others. David is indiff erent because he is not giving the appropriate weight to the interests of others. But let us assume that David decides not to run the light, still valuing others’
interest at 9, but not being a basketball fan, weighing the interest of the game at 1. Here, although David does not run the light, his value system is still such that were he to become a fan, he would be willing to impose great risks on others to get to a Knicks game. Do these theorists wish to punish David for stopping at the light because he in fact gives the inter- ests of others too little weight? Alternatively, is David indiff erent if he correctly assesses the value of others’ lives at 100 but grossly overvalues the Knicks game at 1,000 and thus runs the light? If so, his indiff erence is being manifested in his choice, not merely in the weights of others’
interests that inform the choice.
3. Bizarre Metaphysical Beliefs and Culpability
In Bad Acts and Guilty Minds,36 Leo Katz collects some cases from the colonial period in Africa that, rendered schematically, look like this: Th e actor kills someone and claims that the victim was an evil spirit or a witch, not a human being. Th e law proscribed knowingly killing human beings, not evil spirits or witches. But the actor’s mistake is not of the ordinary factual kind. Th e actor, even if shown the body, the organs, the DNA, and so forth, would probably have said, “Yes, it looks like a human being, but see that mole: that mole proves that this is really a witch. Witches look like human beings in all respects, right down to their DNA, but they are not.”
Assuming these actors were otherwise sane, what should we say regarding whether they acted culpably? Th eir mistakes were, unlike the usual mistakes that negate mens rea in criminal law, not factual
36 Leo Katz, Bad Acts and Guilty Minds: Conundrums of the Criminal Law (1987).
but metaphysical. Th ese actors believed in a richer ontology than the criminal law contemplates.
Were they nonetheless reckless under our scheme? Or do bizarre metaphysical mistakes exculpate just as do factual mistakes, no matter how bizarre? If one accepted their ontological views the way we argue one must accept actors’ factual views when assessing culpability, then it would be diffi cult to deem them culpable. Believing one’s victim is a
“witch” would be no diff erent in its exculpatory power from believing one’s victim is a scarecrow, not a live human being. Of course, to the extent that the actor is aware of some risk that he may be wrong (she might not be a “witch” despite the fact that “she looks like one”),37 the actor might still be reckless for taking the risk of killing a nonwitch.
Whether he would be will depend in how he estimates the competing risks, and what dangers he believes witches present.
4. Deontological Norms and Consequentialist Justifi cations
In balancing risks and reasons, we must also take into account deon- tological constraints on when actions may be consequentially justifi ed.
Th is leads to the question of whether we deem an actor who violates a deontological norm in order to do good – for example, he draws blood from nonconsenting X in order to save the lives of Y and Z – to have acted with insuffi cient concern. We are inclined to say that appropriating another for one’s own purposes, however benign, simply is manifesting insuffi cient concern toward that person. In such cases, the risks cannot be outweighed by the actor’s reasons. We take up this issue much more fully in Chapter 4, and we raise it again in Chapter 8 when we look at the retributive desert of those who act in ways that are justifi ed by a conse- quentialist calculus but who violate deontological side constraints.
B. ASSESSINGTHERISK
1. The Holism of Risk Assessment
In determining whether the actor is reckless, risk is assessed holisti- cally. Th at is, we aggregate risks. Suppose, as will ordinarily be the case,
37 Th is is the test according to Monty Python and the Holy Grail (EMI Films, 1975): “How do you know she’s a witch?” “Because she looks like one.”
that a given act increases by varying amounts the risks to various legally protected interests. So the actor might believe that act A increases the risk to legally protected interest I1 by R1, increases the risk to legally protected interest I2 by R2, and so on. His culpability for A is a func- tion of the sum of the risks he imposes on those interests. It is not just based on, say, the highest risk A imposes, or on the risk to the most important interest. An armed robbery increases the risk to the bank’s money by a signifi cant amount, to the physical safety of employees and customers by a diff erent amount, to their emotional states by a still dif- ferent amount, and so on. Th e risks created are to diff erent interests and are of diff erent magnitudes. Similarly, a speeding car increases risks of death, of bodily injury, and of property damage, and again the risks are of diff erent magnitudes. Even if no one of the risks, viewed in isolation, would render the speeding reckless, the sum of them might. Th is is what we mean by the “holism” of risk assessment.
Th is raises the concern that if one were accidentally to kill in the commission of what is, under current law, a petty crime, our analysis would treat the accidental homicide as a reckless homicide so long as the commission of the petty crime entailed an ever-so-slight increase in the risk of an accidental killing (and the actor disregarded this risk).
For the petty crime is unjustifi able, making every risk it imposes an unjustifi able risk, no matter how slight. We accept this result. First, as we argue in Chapter 5, results should not matter to desert and pun- ishability. But, second, the holism of risk assessment means that com- mitting a petty crime is reckless, not with respect to any discrete legal interest such as death but with respect to the whole array of legal inter- ests the risks to which it increases. Because the actor believed that the increased risk of accidental death was very small, whereas the increased risk to whatever interest the petty crime is concerned with was much greater, his overall culpability for his recklessness will be pretty low, as compared with one who adverts to a high risk of death and a low risk to a petty interest.38 One is not discretely reckless as to death, discretely
38 Th is means, of course, that the recklessness involved in committing what is – again, under current law – a petty crime will vary according to the circumstances. One who attempts to kill a deer out of season will be reckless with respect to, inter alia, killing or wounding a human being. But he will be more or less reckless depending on whether he shoots at the deer in a sparsely or heavily populated area, or uses a bow and arrow or a bazooka.
One would expect that the culpability for committing and attempting to commit what are
reckless as to injury, and so forth for a given act; one is just reckless, as determined by the sum of risks and interests, with the level of culpabil- ity determined by which interests are subjected to which risks (and, of course, why).
Th ere is a second nettlesome problem lurking in this neighborhood.
Consider Frankie, who is driving quite carefully to Johnny’s house, intending to kill him. She is aware that even careful driving increases the risks of death, injury, property damage, and so on. And her reason for driving – to kill Johnny – is surely not a reason that would justify imposing even very tiny risks on anyone. So should we conclude that her driving – remember, it is careful driving – is reckless?
Th ere are two possible solutions that we reject. One is to resur- rect the substantiality prong of recklessness; the other is to measure risk by comparison to a counterfactual baseline. Ultimately, it is our view that we should accept that Frankie is reckless, even if she is only slightly so.
First, as we stated previously, we believe there are good reasons to reject the substantiality prong. Actors who play low-risk Russian rou- lette impose risks (albeit minuscule ones) for bad reasons. We believe these actors are culpable. Moreover, when an actor purposefully tries to harm another, it is immaterial how unlikely the risk of success is; if the actor believes he has even the slightest chance of success, he still acts purposefully and may, under current law, be punished quite severely despite his estimated low probability of success. We see no reason why it is normatively or conceptually desirable to view low-level risk imposi- tion as immune from criminal liability.
Th e other possibility involves the baseline by which increases or decreases in risk are measured. For example, if Frankie were not driving to kill Johnny, would she be sitting at home watching TV, imposing vir- tually no risk to anyone of anything untoward, or would she instead be driving to the mall to shop, a route that is more heavily traffi cked than the route to Johnny’s house and thus entails higher risks to others from
crimes under current law already includes culpability for the average amount of risk such crimes and attempts create regarding interests other than the interest that is the crime’s primary concern. But if the crime is committed in a way that imposes above-average risks on these other interests, the actor could justifi ably be punished for the basic crime or attempt and for the additional amount of recklessness.