In all cases in which the actor acts culpably but no harm occurs – cases in which the actual (God’s-eye) risk is zero, but the actor estimates it to be suffi ciently high to render his act culpable – the actor’s causing harm is in some sense “impossible.” Th e criminal law has traditionally referred to a particular category of acts as “impossible attempts,” a description that we fi nd unfortunate, given that all attempts that are attempts – that is, that fail to bring about harm – are in a very real sense impossible.
Given that the gun was unloaded, the trigger jammed, the aim awry, the victim suddenly bent over, and so on, the attempt could not help but fail.
Th e Model Penal Code brought some helpful order to the topic of impossible attempts by deeming all “impossible” attempts to be crimi- nal with the exception of those in the category of “pure legal impossibil- ity.” Th e latter are those cases in which the actor believes his act violates a particular criminal law that does not in fact exist. If, for example, we mistakenly believe dancing on Sundays is criminally proscribed, and, to protest this absurd law, on Sunday we take to the dance fl oor, we have attempted to violate a nonexistent law. Although we are in some sense culpable – we are, aft er all, scoffl aws – there is no law in the penal code that we can be charged with attempting to violate. Th is is a case of pure legal impossibility and not punishable. (Th e case would be diff erent – and under the Model Penal Code, clearly a punishable attempt – if there were such a law on the books, but we were dancing, not on Sunday, as we believed, but on Saturday.)35
In our schema, the equivalent of pure legal impossibility might be the actor who believes that creating perceived risk R of harm H for jus- tifying reason JR is culpable, who nonetheless creates R of H for JR, but creating R of H for JR is not culpable. Th e actor will believe he has acted
35 Th e line between cases of pure legal impossibility and those cases of impossibility that are punishable is not as neat as the preceding example suggests. For example, what if the law were “One may not dance when there is a red fl ag over the courthouse, but one may when there is a green fl ag,” and we, being colorblind, mistake a green fl ag for a red one but dance anyway? For a discussion of inculpatory mistakes of this type, see Larry Alexander,
“Inculpatory and Exculpatory Mistakes and the Fact/Law Distinction: An Essay in Memory of Myke Bayles,” 12 Law & Phil. 33 (1993).
wrongly toward others when he has not. And just as the criminal law does not punish the pure legally impossible attempt, so too would we exempt from punishment those who believe themselves to be culpable but who are not. Th ere are, in our schema, no “attempts to be culpable.”
One either acts culpably or does not, but one cannot culpably try (but fail) to act culpably.
On the other hand, to the extent we make unduly risking proxy conduct criminal, we introduce the following possibility: the actor may believe that he is imposing an unjustifi ably high risk of proscribed proxy conduct PC but not an unjustifi ably high risk of the harms (H) that PC is meant to avert. Th e actor thus appears to be culpable with respect to PC but not with respect to H. But what if PC is only a fi gment of the actor’s imagination? (Perhaps PC was proscribed in the past but is not so now.) Th e actor has displayed insuffi cient concern for “the law” but not insuffi cient concern for the harms the law wishes to avert.
Where the law proscribes proxy conduct, do we need a category like pure legal impossibility? Our hypothetical actor displays a sort of culpa- bility – the scoffl aw sort – but it is diffi cult to see what he can be charged with. One possibility, of course, is to say that acting as a scoffl aw is a form of culpable conduct and punishable as such. We discuss this possibility again in Chapter 8. Th at approach would avert the necessity of distin- guishing pure legal impossibility from other, clearly culpable, mistakes.
Results do not matter for blameworthiness or punishability. Th e law seeks to infl uence the reasons for which a person acts, but it cannot infl uence the results of these actions. Th us, when an actor risks harm to others for insuffi cient reasons, the law’s infl uence has failed, and a cul- pable act has fully revealed the actor’s desert. No further information, such as whether the culpable act caused harm (or benefi t), is needed for us to determine the degree of her punishment.
Although many people believe that results matter, there is a distinc- tion to be drawn between results mattering (as they must because they aff ect the world in which we live) and results mattering for the moral blameworthiness of the actor. We can recognize that some results are harmful and, indeed, that we create laws to prevent such harms, with- out at the same time committing ourselves to the view that results inde- pendently aff ect blameworthiness.
Because the law currently gives independent signifi cance to the role of resulting harm, criminal law doctrine is mired with fl aws. Both cause- in-fact and proximate causation present fl awed doctrinal accommoda- tions to approximate culpability when causal chains go awry. Th e better approach is simply to abandon any concern with results completely.
Finally, we can make a principled distinction between moral or result luck and other forms of luck, and that distinction rests on the very choice that is at the root of culpability. No matter what hand we are dealt, moral agents have the capacity to reason and the capacity to choose to violate the law (or not). It is on this choice and this choice alone that responsibil- ity rests and, along with it, culpability and desert.
197 Having argued that results are neither necessary nor suffi cient for blameworthiness or punishability, we must still answer the question of what type of action is necessary for the actor to be said to have acted culpably. Although this problem is traditionally addressed within the doctrinal rubric of the actus reus for incomplete attempts, the problem for us applies more generally. We must specify an actus reus formula- tion for all crimes.
Th ere are various potential actus reus formulations, drawn along the continuum from the time the actor forms an intention to impose a risk of harm to the time when he believes he is unleashing that risk and it is no longer within his (complete) control. We contend that it is only at the time the actor engages in the act that unleashes a risk of harm that he believes he can no longer control (through exercise of reason and will alone) that he has performed a culpable action.
V I
When Are Inchoate Crimes Culpable and Why?
In this chapter, we begin by setting forth the principles that underlie our adoption of the “last act” formula.1 We then survey the various points along the inchoate crime continuum, from the formation of the intention to impose the risk, to the Model Penal Code’s intention plus “substantial step,” to the common law’s intention plus “dangerous proximity,” to the last act. In our view, it is only the last act – the act through which the actor believes he has relinquished (complete) control over whether he has created an undue risk of harm (or proxy conduct) – that is a culpable act.
(In the next chapter, we further refi ne this view by clarifying that it is the actor’s volition to move his body in such a way as to engage in the last act necessary for the unleashing of the risk that is the culpable “act.”)