Because purposeful conduct is not always unjustifi ed, we argue that it is
the price he pays, for he will assess his chances of winning $1 million as 1 in 1 million. But the offi cials will assess his chances as 1, and everyone else’s chances as 0. Th eir God’s-eye perspective provides them with an information base and epistemic position diff erent from that of the ticket buyers.
Th e same point can be illustrated by a coin fl ip. If one of us fl ips a coin, the others will estimate the chance that it lands heads as 50 percent, and they will do so even if the coin fl ipper already knows the results of the fl ip but has not revealed it. From the fl ipper’s per- spective, if the coin landed heads, the “chance” of its doing so is 1. If not, its “chance” is 0.
12 Th is section draws from Alexander, supra note 3.
not necessarily more culpable than knowledge or recklessness. Finally, we address how our conception resolves questions about how one can act purposefully as to an attendant circumstance.
A. KNOWLEDGE
A person acts with criminal “knowledge” with respect to an element of a crime when he believes to a practical certainty that his conduct is of a particular nature, that an attendant circumstance exists, or that his conduct will bring about a particular result.13 Unlike recklessness, which focuses on both the risk imposed and the actor’s justifi cation for imposing it, at fi rst glance knowledge appears to focus only on risk – of forbidden conduct, resulting harm, or circumstance – and the actor must be practically certain it will result.
One might think from a superfi cial comparison between the formu- lation of knowledge and the formulation of recklessness that when an actor is practically certain that his act will be harmful to others, he is forbidden from undertaking it regardless of his reasons for doing so, whereas if he believes the risk is less than that of practical certainty, he may act if his reasons are suffi ciently weighty. But it is not the case that knowingly imposing harm is always unjustifi ed. For example, one may justifi ably impose a practically certain harm on another to defend one- self or others or, in some cases, where worse harm to another or a simi- lar harm to many others can only be averted thereby.14
“Knowledge” – acting with the belief that it is practically certain that one’s act will be harmful to others or turn out to be of a forbid- den nature – is merely a limiting case of recklessness along the axis of degree of perceived risk of harm. Th e other recklessness axis, the axis of the actor’s reasons, remains fully operative. Th e only real distinction between knowledge and recklessness in the criminal law is that at some point as the risk of harm (or forbidden conduct) approaches a practical certainty, the burden of proof (or production) on the issue of whether the defendant-actor’s reasons for acting justifi ed the risk of harm he created shift s from the prosecution to the defendant-actor. Th at shift in
13 Model Penal Code § 2.02(2)(b) (1985).
14 Id. §§ 3.02–3.05.
the burden of proof (or production) makes sense in terms of both the probability that the act is or is not justifi ed and the relative access to the justifying circumstances, if any. But the diff erence in the allocation of burdens does not refl ect any substantive distinction between knowledge and recklessness.15 Moreover, in our view, the Model Penal Code is cor- rect in its assessment that lack of justifi cation is something the prosecu- tor must prove beyond a reasonable doubt, though this burden may be aided by a permissive presumption placing the burden of production on the defendant.
Seeing knowledge as just a species of recklessness enables us to avoid two errors. First, it allows us to avoid the error of deeming all cases of knowledge to be more culpable than all cases of recklessness, even where the harm risked is the same. For example, someone who imposes a very high risk of harm on another – a risk just short of practical certainty – for a very frivolous reason, such as a thrill, is surely more culpable than one who imposes a practically certain harm for a quite weighty, but ultimately insuffi cient, reason. For example, if Albert knowingly infl icts severe bodily injury on another to prevent him from accidentally destroying Albert’s entire life’s work, Albert may be culpable; but he is surely less culpable than if he imposes a slightly lower risk of the same harm on another just to satisfy his urge to drive like a madman. In short, instances of what we now call extreme indiff erence to human life may not only be equal to but also more culpable than some cases of knowledge.
Th e second error averted by seeing knowledge as just a species of recklessness is the misclassifi cation of cases of willful blindness. A typi- cal case of willful blindness is one where the actor is asked by someone at an airport in a drug-producing country to carry a package on the plane and give it to a specifi c person when he arrives in the United States.
Where the applicable statute punishes only “knowing” drug smuggling, courts have struggled to fi nd the actor – whom they rightly regard as culpable – guilty by deeming his willful blindness to be tantamount to knowledge.16 Yet willful blindness is not knowledge; it is an attempt to
15 Th e Model Penal Code regards (the absence of) justifi cation as part of the prosecutor’s burden of proof. See Model Penal Code § 1.12(2) (1985). It is constitutional, however, to place this burden on the defendant. See, e.g., Martin v. Ohio, 480 U.S. 228 (1987).
16 See, e.g., Mattingly v. United States, 924 F.2d 785, 792 (8th Cir. 1991) (“[T]he element of knowledge may be inferred from deliberate acts amounting to willful blindness. . . .”);
avoid knowledge, and this hiding of one’s head in the sand frequently achieves that goal in that actors who display willful blindness do not in fact believe to a practical certainty that their conduct is harmful.
Th e prototypical willfully blind actor is, of course, reckless. Th e risk he is taking – of, say, smuggling drugs – is an unjustifi able one. Th e unjustifi ability of the risk cannot, however, convert his recklessness into knowledge without absurd results. For example, suppose a drug smug- gler employs 100 “mules” to do his smuggling.17 His modus operandi is to give each mule a suitcase to carry into the target country. He tells the mules that in ninety-nine of the suitcases he places only innocent items, such as clothes, and in one he places drugs. Th e mules are not told which suitcase holds the drugs, and they do not open the suitcases. In such a situation, it would be absurd to deem the mule with the drugs to have
“knowingly” smuggled them. What he has done is take a one in a hun- dred chance of smuggling drugs for no legally suffi cient reason. If a sim- ilar risk imposition would not be deemed “knowing” if undertaken for good reasons – say, for example, you are visiting a foreign country and a person whom you greatly admire asks you to carry a gift to a friend in the United States, and you entertain the thought that there is a tiny but real chance that the person you admire is a drug smuggler – then it is misguided to deem the risk imposition “knowing” merely because one disapproves of the reasons for undertaking it.
In some sense, all risk imposition is willful blindness in that the actor could always seek more information about the risk before acting, although the act thereaft er undertaken would be diff erent temporally and circumstantially from the act undertaken without gathering more information. Moreover, in ordinary recklessness cases, when we assess the actor’s reasons for imposing the risk, we are also implicitly assessing his reasons for not investigating the danger further. Viewing reckless- ness as the all-encompassing conception of criminal culpability, rather than as diff erent from and less culpable than knowledge, allows us to
State v. LaFreniere, 481 N.W.2d 412 (Neb. 1992); see also United States v. Whittington, 26 F.3d 456 (4th Cir. 1994); United States v. Mancuso, 42 F.3d 836 (4th Cir. 1994); United States v. Jewell, 532 F.2d 697 (9th Cir. 1976); United States v. Incorporated Village of Island Park, 888 F. Supp. 419 (E.D.N.Y. 1995).
17 See Douglas N. Husak and Craig A. Callender, “Willful Ignorance, Knowledge, and the
‘Equal Culpability’ Th esis: A Study of the Deeper Signifi cance of the Principle of Legality,”
1994 Wis. L. Rev. 29, 37 (1994).
evaluate correctly the willful blindness cases and does not tempt us to distort them into cases of knowledge.18
B. PURPOSE
A person acts with criminal purpose with respect to the conduct or result elements of an off ense if “it is his conscious object to engage in conduct of [the required] . . . nature or to cause [the required] . . . result.” 19 Unlike knowledge, which at fi rst glance appears to focus only on the actor’s assessment of the risk of harm, purpose at fi rst glance appears to focus only on the actor’s desires regarding the possibility that his con- duct will prove harmful, while ignoring his beliefs about this possibility.
But as with knowledge, this initial appearance proves to be misleading.
Th e crucial point is that in order to act with criminal purpose, the actor must believe that his conduct increases the risk of harm, even if the increase is very slight. Consider the Jackal, who fi res a shot at de Gaulle from the Eiff el Tower believing his chance of success to be one in a mil- lion. If he does succeed, he will have shot de Gaulle “purposely,” given that it is his “conscious object” in shooting that de Gaulle be killed (that is how he collects his assassin’s reward). But characterizing this act as pur- poseful assumes that the Jackal believes that, although fi ring the gun pro- duces only a one in a million chance of killing de Gaulle, fi ring the gun does increase the probability of de Gaulle’s being killed over the prob- ability of de Gaulle’s being killed if the Jackal refrains from fi ring the gun.
If the Jackal does not believe his fi ring the gun increases the probability of killing de Gaulle – if he believes de Gaulle has no worse chance of sur- viving if he fi res the gun than if he does not – then were the shot to kill de Gaulle, it would not be a purposeful killing. For in such a case, the Jackal would not have fi red the gun for the purpose of killing, given that he did not believe fi ring the gun produced any increased chance of killing de Gaulle.
18 Recklessness, because it is sensitive to both risks and reasons, allows us to distinguish between those whom David Luban calls ostriches, who merely do not want to know, and those whom he calls foxes, who contrive deniability. See David Luban, “Contrived Ignorance,” 87 Geo. L.J. 957, 968–975 (1999). Ostriches and foxes may take the same risks, but they have diff erent reasons for doing so. Both may be culpably reckless, but foxes are more culpable than ostriches given the same level of risk and the same harm.
19 Model Penal Code § 2.02(2)(a) (1985).
Now, there are outlier cases where, even though the actor believes that his action has some chance of success, he may also believe that he is decreasing the overall chance of harm to the victim. Imagine that David wants to kill the president but knows that he is a terrible shot.20 He also knows that a hired hit man (a far better shot) intends to kill the president later in the day. David also believes that if he takes his shot and misses, then security will be improved and it will decrease the chances that the later hit man will be successful. Th us, David may come to believe that his taking the shot will actually decrease the chance of the president’s being killed. If David proceeds to shoot at the president and, against all odds, the shot fi nds its mark, is it correct to characterize David’s act of homicide as “purposeful”?
In our view, David has purposefully killed the president. To do something on purpose requires the actor to understand that his action can have some eff ect in the world – some chance of success. Indeed, it would be irrational for David to proceed if he believed that he could not kill the president. Th us, it may be said that David is acting purposefully even if he knows that if he fails he will then decrease the probability of the president’s being killed, and to a greater extent than his shooting by itself increases that probability.21
Some may argue that, as a conceptual matter, our claim that purpose entails both a reason to cause the harm and a belief that one may succeed runs counter to the standard position within the philosophy of mind and
20 Private correspondence of David Dolinko, Professor of Law, UCLA School of Law, and Larry Alexander.
21 Th is hypothetical also raises a second question about the relevance of David’s knowledge that his action might actually decrease the overall probability that the president will be killed. Th is belief may have the eff ect of justifying David’s action. Th e question is complex, as it requires us to ascertain whether David must be motivated by, or simply aware of, the justifying circumstances and whether David may avail himself of a lesser-evils defense when he chooses the lesser, but perhaps not the least, evil. We attend to all of these ques- tions in Chapter 4. Th e important point for now is that purpose does require both that the actor act on his desire to cause a harm and that he believe that his action has a chance of success.
Another point that the hypothetical raises, again relevant to the question of justifi ca- tion, is whether imposing a risk of dying sooner can be justifi ed by averting a greater risk of dying later. Obviously, we believe it can be, which is why we consent to risky surgery to prevent a later but riskier disease. But because we are mortal and will die no matter what precautions we take, not every present risk can be justifi ed by averting a later and greater risk.
to a common view that the degree of risk is irrelevant to the culpability of purposeful actions. Let us discuss each of these objections.
First, consider the objection that an intention to A does not entail a belief that one will A.22 To use an example of Michael Bratman’s, if Alex discovers a log blocking his driveway, he may form the intention to move it, but not form the belief that he will move it. Th en, there is noth- ing irrational about Alex’s (1) forming the intention to move the log, and (2) forming the intention to call a tree company if or when he fails.23 As a conceptual matter, it appears that actors can form intentions to p and yet not believe that they will p.
At the outset, we do not argue that acting purposefully requires that one believes one will succeed. Rather, our claim is far more limited – it is that acting purposefully requires a belief that one has some chance whatsoever of success. When Alex tries to lift the log, he may not believe that he will lift it, but he surely does not believe it to be absolutely cer- tain that he will fail.
Second, it seems just as natural to say that what Alex intends is “to try to lift the log.” When we are cognizant of the likelihood of our fail- ure, we sometimes cast our intentions not in terms of results but in terms of “tryings.” Of course, once one does not believe that one will succeed, but only that one will “try,” one is not committed to any degree of success. But now one can see just how fi ne (or nonexistent) the line is between purpose and recklessness. Purposeful riskings – like Russian roulette – are commonly seen as instances of recklessness. If Cowardly Jackal is paid $1,000 to kill de Gaulle, and, afraid of being caught, fi res from a distance from which he estimates his likelihood of success at 1/1,000, he imposes the same risks for the same reasons as Risky Jackal, who is paid $1,000 to create a 1/1,000 chance of de Gaulle’s death.
Also, although we may speculate about the activities that we can engage in without believing that we can succeed, this objection hardly hits the mark when it comes to criminal activity. An actor who purpose- fully engages in conduct to cause harm to another does so in order to succeed (or at least to try to succeed). Indeed, instances of pure factual impossibility (voodoo and the like) are cases in which the defendant
22 See Michael Bratman, Intentions, Plans, and Practical Reason 37–41 (1987).
23 Id. at 39.
irrationally believes in the likelihood that she will succeed despite all science and common sense to the contrary. If there exists an actor who intends to harm another through a particular act but does so without a belief in any likelihood of that act’s success, we will concede that such an actor is not culpable for that act.
Finally, even if one does not agree with our claim that one may con- ceptually reduce purpose to recklessness, we still believe that norma- tively we can. Th at is, in instances of both purpose and recklessness, the actor’s culpability ultimately turns on both his reasons for acting and his belief as to the likelihood of success.
Th is last normative claim brings us to the second possible objec- tion – that the culpability of a purposeful actor does not turn on his estimate of the likelihood of his success. As between two actors, each of whom has the same harm as his purpose and acts for the same reasons, is the one who believes the risk he is imposing is higher more culpable than the other? In other words, if we hold reasons constant, and among those reasons is the purpose to harm, do variances on the “perceived risk” axis aff ect culpability?24
We are inclined to say that culpability does vary with the actor’s estimate of the probability. Our culpability assessment evaluates the defendant’s choice to impose this risk for these reasons. Even if the actor would have imposed a greater risk – something we can presumably assume because he acts purposefully25 – this is not the choice he made.26 Moreover, even though the actor wants the harm to occur, he may be unwilling to impose a greater risk. Th e Cowardly Jackal who wants to kill de Gaulle but by imposing only a 1/1,000 risk is less culpable than the Brave Jackal who holds a gun to de Gaulle’s temple and then pulls
24 For a discussion of this issue, see Itzhak Kugler, Direct and Oblique Intention in the Criminal Law 90–102 (2002). We return to this issue in Chapter 8.
25 As suggested by Kugler, id.
26 Alan Michaels suggests something akin to this argument in “Acceptance: Th e Missing Mental State,” 71 S. Cal. L. Rev. 953 (1998). Michaels’s argument is that some actors who believe their conduct only risks harm feel the same degree of indiff erence as actors who believe their conduct will certainly cause harm, and because both types of actors “accept”
(i.e., psychologically resign themselves to) the harm, both are equally culpable irrespec- tive of their diff erent estimates of the probability of the harm’s occurrence. Although we believe that this equal level of indiff erence shows both actors to have bad characters, still, only one was presented with the opportunity to impose a practically certain harm. Th us, they are not equally bad actors.