Faultless mistake of fact: justification or excuse?
According to Lafave and Scott, "No area of the substantive criminal law has traditionally been surrounded by more confusion than that of ignorance or mistake of fact or law." However, the authors go on to say that "[i]n actuality, the basic rule is extremely simple: ignorance or mistake of fact or law is a defense when it negatives the existence of a mental state essential to the crime charged." Typically then, a mistake of fact on the part of a defendant at the time of the offense serves to undergird a failure of proof defense. In a failure of proof defense, the defense attempts to show that an element essential to the definition of the crime is not present and, thus, that the prosecution has not proved its prima facie case. Regina v. Morgan provides an example of the utilization of a defense of this sort. In Morgan, the Lords concluded that since Morgan's co-appellants sincerely believed that the alleged victim consented to intercourse (although in fact she had not), it cannot be said that they intended the act qua prohibited act, regardless of the reasonableness or unreasonableness of each of their beliefs regarding her consent. Thus, according to the Lords, controversial decision, the mens rea requirement is not met in circumstances in which a defendant charged with rape sincerely believed, at the time of the act, that the intercourse was consensual. In effect, the mistake of fact negatives defendant intent.
An affirmative defense may also be founded upon a defendant's mistake of fact. A defense of this sort may concede that the prosecution has made its prima facie case, that is, that the conduct and culpability requirements for the crime are met, but nevertheless contend that the defendant ought to be found not guilty for other reasons, for example, because the defendant's actions were justified or excused. People v. Young serves as an example of the application of such a defense. In this case, Young went to the aid of an eighteen-year-old individual who was struggling against two assailants. According to Young, he intervened because the youth was crying and trying to escape the clutches of the two men who, in the struggle, had nearly removed the young man's pants. As a consequence of Young's intervention, one of the men believed by Young to be an unlawful assailant suffered a broken leg. Unbeknownst to Young, though, the two assailants were in fact plain clothes detectives. So, unfortunately for Young and the detectives, Young was mistaken about the facts of the situation. Specifically, he held the erroneous belief that the two assailants were ordinary civilians, not police officers carrying out their duties. However, although Young knowingly and purposely attacked the two men who turned out to be officers of the law (and, so, mens rea, in addition to the other material elements for assault, was clearly present), Young had an affirmative defense for his conduct. The New York State legislature provides those with reasonable but mistaken beliefs with the following defense: "A person may...use physical force upon another person when and to the extent he reasonably believes such to be necessary to defend himself or a third person from what he reasonably believes to be the use or imminent use of unlawful physical force by such other person."
In this paper, I shall focus upon the affirmative defense of actions similar to Young's, actions "based upon a faultless mistake of fact." I shall advance the following thesis: action based upon a faultless mistake of fact are appropriately deemed excused, not justified, by morality and by the law. In Section II, I explicate the basic distinction between the notions of justification and excuse as well as suggest why the distinction is important. In Section III, I provide an analysis of the notion faultless mistake of fact. Next, in Section IV, I criticize the competing analysis of Kent Greenawalt and provide arguments for the central thesis of my paper. In Section V, I offer what I think is the strongest argument for the claim that actions based upon a faultless mistake of fact ought to be regarded as excused. Finally, in Section VI, I point out perplexing implications of my thesis--problems, centering on issues of responsibility and luck-and offer what I think is an appropriate answer to these problems.
II. Justification and Excuse
Justification and excuse provide two separate categories of affirmative defenses. Although there are hard cases for which it is difficult to determine whether a defendant's action was justified or excused, the basic categories are rather easily distinguished. Kent Greenawalt draws the distinction in the following way: "If A's claim is that what he did was fully warranted-he shot B to stop B from killing other people--A offers a justification; if A acknowledges that he acted wrongfully but claims he was not to blame--he was too disturbed mentally to be responsible for his behavior--he offers an excuse." J. L. Austin similarly distinguishes justification and excuse: "In the one defense, briefly, we accept responsibility but deny that it was bad: in the other, we admit that it was bad but don't accept full, or even any, responsibility."
In effect, when we say that a person's behavior was justified, we are saying that she did the right, or at least, a permissible thing. We might distinguish between the two sorts of justification in the following way. If the individual did what she should have done, that is, the right thing, then we might say that her behavior was strongly justified; if, instead, what she did was merely permissible, we might say of her behavior that it was weakly justified. In contrast to either sort of justified action, if we say that a person's behavior was excused, we are saying, at the very least, that the behavior in which she engaged is properly deemed impermissible but that due to some internal or external factors beyond her control she is not to be held blameworthy for her actions. It is tempting to say something stronger about a person's excused behavior, namely, that the behavior constitutes a case of wrongdoing (that is, a case in which some individual's rights have been violated), but as Joel Feinberg tells us, "Excused . . . wrongdoing is not wrongdoing at all" since the actor is not accountable for the injury caused to the unfortunate victim. Thus, in such cases, it is almost as if the victim's injuries were caused by some natural occurrence, for example, a falling tree in a wind storm, a case we would not be inclined to describe as involving a rights violation. The most we can say of the excused actor is that his behavior was impermissible and would have been a case of "wrongdoing" had the excusing condition not obtained.
I want to suggest that the expressive roles played by the notions of justification and excuse provide significant reasons to distinguish sharply between the two sorts of defenses. Here I appeal to Joel Feinberg's seminal article, "The Expressive Function of Punishment," in which he convincingly argues that one distinctive feature of punishment is that it "expresses the judgment (as distinct from any emotion) of the community that what the criminal did was wrong." Similarly, I contend that the justification/excuse distinction provides a framework within which the community can express judgments regarding the defendant's actions. Admittedly, the distinction itself does not allow for more exacting discriminations than we might otherwise make regarding the defendant's level of blameworthiness for her actions; whether a defendant's conduct is deemed justified as opposed to excused or excused as opposed to justified tells us nothing different about the level of blameworthiness that should be attributed to the defendant. In either case, the defendant is undeserving of blame: if justified, the defendant did nothing wrong for which she might be blamed; if excused, her lack of responsibility precludes our blaming her. Nevertheless, the acceptance of one defense as opposed to the other tells us a great deal about the reasons underlying the defendant's blamelessness and, consequently, very much about the community's view of the sort of action in which the defendant engaged. When the defendant's behavior is justified, the community expresses its judgment that this sort of behavior is appropriate and that it merits praise, or else that this sort of behavior is at least acceptable according to the community's standards. On the other hand, when the defendant's conduct is merely excused, the community expresses its judgment that others should not engage in behavior of this kind and, more importantly, that those who do so voluntarily will be sanctioned accordingly.
Greenawalt defends the basic distinction on related grounds: "because it reflects and reinforces moral judgments, criminal law should illuminate the moral status of various courses of action, and the community should be concerned with the reason a particular person goes unpunished." Greenawalt valuably supplements the considerations raised above by drawing attention to the necessarily reciprocal relationship between legal judgments and community standards. If our standards are to inform our legal judgments (and, thus, if the latter are to express the former), it is necessary that our legal judgments provide the community with relevant feedback. We can be sure that legal judgments actually express our societal standards only if the reasoning behind the judgments is made sufficiently explicit for potential community criticism. It appears that Greenawalt is acutely aware of this fact.
Lastly, we might add, the distinction allows the community to express its standards to the defendant himself. For example, as Greenawalt puts it, "actors who are genuinely justified should have legal confirmation that they have acted appropriately ... " In summary then, the distinction between justification and excuse is important because we are concerned not only with whether a given defendant is blameworthy for her actions. In addition to our concerns about blameworthiness, we have reason to express societal judgments about particular forms of conduct and to provide those in the community as well as those most affected by the legal judgments, that is, the defendants themselves, with the reasons behind our legal judgments. The disparate categories of justification and excuse, in typical cases, clearly advance these objectives.
III. Faultless Mistake of Fact
Before we can determine whether an action based upon a faultless mistake of fact is appropriately deemed justified or excused, we must first understand exactly what it means to say of someone that he was "faultlessly mistaken about some matter of fact." First, it is clear that the mistake of fact refers to some aspect of the defendant's belief state at the time of the illegal conduct. Specifically, the defendant's belief(s) did not correspond to or reflect the facts as they actually were. The difficult part of the analysis lies in ascertaining what it means to say of a defendant that he was faultless in the acquisition and maintenance of the mistaken belief, that is, in coming to be and remaining in the erroneous belief state. I want to suggest that an account of doxastic fault (that is, fault regarding belief acquisition and belief maintenance) will appeal most plausibly to some theory of epistemic justification. (Very roughly, theories of epistemic justification provide conditions under which it is permissible to acquire and maintain particular beliefs.) On such an account, we would consider a belief to be faultless to the extent that the belief is epistemically justified. In order to determine which sort of theory of justification best serves our present purposes, it is necessary that we look at the two basic types of theories of epistemic justification.
Internalist theories of justification hold that whether or not a belief is justified is solely a function of the internal states of the epistemic agent, whereas externalist theories hold that other features, features external to the epistemic agent, are relevant to justification. Coherence theories of justification provide an example of the former type of epistemic theory. Roughly, according to coherence theories of justification, the justification for any belief depends solely upon the relationship of the belief in question to the other beliefs held by the epistemic agent. In contrast, process reliabilism is an i externalist theory of justification. This theory of epistemic justification holds that a belief's justification depends upon the reliability of the cognitive process which produced the belief. Simply put, a cognitive process is reliable to the extent that there is a high indefinite probability that it will produce true beliefs. So, for instance, if 99 out of 100 of one's perceptual beliefs are true, since there is a high indefinite probability--.99--that a belief formed by this cognitive process will be true, one is justified in acquiring and maintaining one's perceptual beliefs.
Internalist theories of justification best capture faultlessness in belief acquisition and belief maintenance. Although the details of internalist theories of epistemic justification differ, internalists agree that a person is justified in acquiring and maintaining a belief just in case she is doing the best she can with the beliefs she has. What they disagree about is what exactly constitutes doing the best one can with the beliefs one has. Thus, on the internalist account, one cannot be unjustified in the acquisition and maintenance of a belief unless one is somehow epistemically at fault, and if one is epistemically at fault, it follows that one is epistemically unjustified. So, epistemic faultlessness is necessary and sufficient for internalist epistemic justification.
On the other hand, externalist accounts of epistemic justification hold that the justification for any belief depends upon factors external to the agent herself and, thus, upon factors necessarily beyond her control. This feature of the externalist account allows for the following possibility: one can unjustified in one's belief even though one is doing the best one can with the beliefs one has. So, on the externalist account, the epistemic agent can be unjustified yet epistemically faultless. Two sorts of examples vividly bring out the sense in which this result is possible. The first sort of example appeals to very skeptical worries regarding the connection between our beliefs and the external world. The original, Cartesian form of the skeptical worry asks that we consider the possibility that our beliefs are caused not by the external world as we (think we) know it but instead by an all powerful evil spirit bent on deceiving us. In the contemporary analogue of Descartes' skeptical problem, the epistemic agent is no longer deceived by an evil demon; rather, she is a brain in a vat of nutrient and receiving (deceptive) sensory input via a connection to a very sophisticated and powerful computer. The skeptical worries are relevant to the analysis of epistemic faultlessness in the following way. On the process reliabilist account, if it turns out that we are in fact deceived by a Cartesian evil demon or else that we are brains in a vat, in which case our cognitive processes, for example, perception, are very unreliable, then we are unjustified in holding the vast majority of our beliefs. More importantly for the present analysis, since the ultimate reliability of our cognitive processes is something over which we have no control, we are unjustified through no fault of our own.
Similar difficulties for an externalist account of epistemic faultlessness arise in the context of a second sort of example. Examples of this sort appeal to what William Alston refers to as "cultural isolation." Consider Alston's particular example of cultural isolation:
If I have grown up in an isolated community in which everyone unhesitatingly accepts the traditions of the tribe as authoritative, then if I have never encountered anything that seems to cast doubt on the traditions and have never thought to question them, I can hardly be blamed for taking them as authoritative. There is nothing I could reasonably be expected to do that would alter that belief-forming tendency. And there is nothing I could be expected to do that would render me more exposed to counterevidence.
However, insofar as the unthinking acceptance of tradition is not a reliable cognitive process, the tribesman turns out to be epistemically unjustified on the externalist theory of epistemic justification. So, once again we see that one can be epistemically unjustified yet epistemically faultless on the externalist account of epistemic justification.
We should also recognize that one can be epistemically at fault yet epistemically justified on the externalist account of epistemic justification. As Laurence Bonjour points out,
The intuitive difficulty with externalism ... is this: according to the externalist view, a person may be highly irrational and irresponsible in accepting a belief, when judged in light of his own subjective conception of the situation, and may still turn out to be epistemically justified .... His belief may in fact be reliable, even though he has no reason for thinking it is reliable--or even has good reason to think that it is not reliable. But such a person seems nonetheless to be thoroughly irresponsible from an epistemic standpoint in accepting such a belief.
Bonjour applies the reliabilist theory of epistemic justification to a hypothetical example in order to support his point: an individual who is in fact reliably clairvoyant and acquires beliefs in this manner turns out to be epistemically justified on the externalist account of justification, regardless of whether he has no reasons to support his trust in his special powers and regardless of whether he in fact has reasons from his own point of view that speak against such trust. Clearly, however, it appears that such an epistemic agent is epistemically at fault for acquiring beliefs in this fashion and for maintaining them without reasons or in the face of (apparently) disconfirming reasons. Thus, on the externalist account of epistemic justification, one can be epistemically at fault yet epistemically justified.
These considerations support the following conclusion. Since on the externalist account of epistemic justification it is both the case that one can be epistemically faultless but epistemically unjustified and that one can be epistemically at fault yet epistemically justified (though obviously not at the same time for the same belief), it appears that if we are to understand epistemic faultlessness in terms of epistemic justification, then epistemic faultlessness must be understood in terms of an internalist theory of epistemic justification.
IV. Justified Belief and Justified Action
Now that we have a very basic analysis of faultless mistakes of fact, an analysis under which they are best understood as (internally) justified, but mistaken, beliefs, we are in a position to discern the connection between the faultless mistake of fact and justified action. For purposes of clarity, let us refer to the sort of justification which renders a mistake of fact faultless as e-justification (for epistemic justification) and to the sort of justification regarding conduct as a-justification (for action justification). It is important to note that these are very different sorts of justification, and, as I shall argue in this section, the connection between the two is not as strong as one might at first assume.
In both "The Perplexing Borders of Justification and Excuse" and "Distinguishing Justifications from Excuses," Kent Greenawalt provides, among other things, an analysis of the connection between e-justification and a-justification. According to Greenawalt, the connection is as follows: an action based on a faultless mistake of fact that would be a-justified, if the belief were not only e-justified but also true, should be regarded as a-justified, not as excused, by the law and morality. This contention is an important one within the overall structure of Greenawalt's articles. First of all, it serves as an important component within his argument for the superiority of the "actor's appraisal approach" over the "objective approach"
and the "actual facts approach" to legal justification. The "actor's appraisal approach" to legal justification holds that an actor's faultless belief that (what would in fact be) justificatory circumstances obtain is necessary and sufficient for a-justification. In contrast, George Fletcher advocates the "objective approach" to legal justification, which holds that an action is a-justified only if justificatory circumstances actually obtain. The "actual facts approach," as I understand it, is simply a strong version of the "objective approach" to legal justification. It holds that an action is justified if and only if justificatory circumstances actually obtain. Secondly, Greenawalt's contention serves as a premise in his argument against the claim that a-justification and excuse can be distinguished by looking at the appropriateness of defense, intervention, and support on the part of other actors. In this section, I shall present arguments which, I think, undercut Greenawalt's claim that actions based upon faultless mistakes of fact are a-justified, and I shall go on to point out some of the implications of these arguments for Greenawalt's other central theses.
Greenawalt's claim, explicitly stated, is as follows: "the actor's blameless perception of the facts ought to be sufficient to support a justification [of his action]." Greenawalt explains the reasoning behind this conclusion:
[T]hat the law should treat [such a] defense as one of justification seems plain, at least if the law's crucial distinction is to be between warranted and wrongful action. In respect to [such behavior], society would expect and hope that a similar actor with a similar set of available facts would make the same choice. And society's view of the morality of the manner in which he acted, its moral judgment about him, should not be affected by the unfortunate outcome .... [H]e has acted as the most competent practitioner in his field would have acted.
Greenawalt applies this analysis to Young, whose case was described in Section I. According to Greenawalt, Young is to be "praised, not blamed, for what he did, and members of society would wish that others faced with similar situations requiring instant judgment would act as Young did." Importantly, Greenawalt's argument rests upon the acceptable contention that "justification in law should correlate...closely with moral appraisal." So, initially at least, Greenawalt's remarks appear intuitively plausible. If an actor reasons as best she can given the facts as she sees them, we might find that we are inclined to say that the behavior based upon her faultless mistake of fact is justified. However, I want to suggest that this conclusion fails to take seriously the distinction between the two notions of justification noted above, namely, e-justification and a-justification.
First, it is important to note that Greenawalt, despite his claim above that "the actor's blameless perception of the facts ought to be sufficient to support a justification [of his action]," is not advocating that e-justification simpliciter entails a-justification. Clearly, to do so would be a mistake: it would mean, among other things, that unjustified action necessarily involves some epistemic error! Rather, Greenawalt holds that if an action would be justified were the agent's belief true, then e-justification is sufficient for a-justification. For simplicity's sake, let us assume for the remainder of this paper that the behavior of those whose actions are based upon faultless mistakes of fact would be a-justified if their beliefs were not only e-justified but true as well.
Nevertheless, Greenawalt's analysis lends itself to criticism even when his claim that "the actor's blameless perception of the facts ought to be sufficient to support a justification [of his action]" is restricted to those cases in which the actor faultlessly believes that (what would in fact be) justificatory circumstances obtain. In short, the inference from e-justification to a-justification proves unacceptable for this class of cases as well. First, the inference ignores the fact that the morally relevant costs of accepting a belief are typically significantly lower than the morally relevant costs of acting upon that same belief. That is, although one may be justified in holding some belief that p, the morally relevant costs of acting on that p might be unacceptably high when that p is false and, thus, render the act unjustified.
For instance, suppose that Fred is e-justified in believing that p:
Charlene is cheating on Fred's friend Charlie. Fred's epistemic justification, let us say, rests upon the fact that this belief coheres with the other beliefs in Fred's system of beliefs. We might assume that Fred holds himself to high epistemic standards and believes that he has good evidence for this conclusion, that he has examined competing explanations for the evidence, and that, all in all, he has been a flawless reasoner. It does not follow from these facts that Fred is a-justified in conveying his belief to Charlie. If Fred is mistaken, little harm can come from his belief so long as it remains simply his belief. However, the stakes are raised considerably if Fred conveys this information to others and, in particular, if he conveys it to Charlie. For, if Fred acts upon his belief and it turns out that he is in fact mistaken, he is likely to have caused considerable harm to Charlene and possibly to others as well. We might partially, or even totally, excuse Fred's action, given that his belief was e-justified. However, we are unlikely to say that Fred's action, in this situation, was an a-justified one, despite the fact that he was e-justified in his belief that p. So, if legal analysis is to correspond closely to moral analysis as Greenawalt suggests it should, there is reason to deny any strong, general connection between e-justification and a-justification. Thus, on its most plausible interpretation, Greenawalt's contention that "the actor's blameless perception of the facts ought to be sufficient to support a justification [of his action]" is false.
In addition to these considerations, there is a strong conceptual argument in favor of regarding actions based upon a faultless mistake of fact as excused, not as justified. Let us say that Fred was not e-justified in believing that p. For instance, let us assume that it was unreasonable for him to believe that Charlene was having an affair given the other beliefs that he held and, so, that he was unjustified on our internalist account of epistemic justification. In addition, let us also assume that he sincerely held the belief and proceeded to share his belief with Charlie.(38) In these circumstances, we would want to say that Fred's behavior was blameworthy, to some extent or other. Clearly, he is not to be blamed to the extent to which we would blame him were he deliberately to lie to Charlie in order that he might have a chance to be with Charlene. Instead, we partially excuse his behavior because his intentions were good (let's suppose) and because he truly believed, although unjustifiably, that Charlene was having an affair. As Greenawalt puts it, "negligent ignorance is better characterized as a partial excuse."(39)
Greenawalt's proposal that we construe negligent ignorance as a partial excuse rests upon the acceptable premise that the person who harms another due to her negligent ignorance is less responsible for what she has done than is the deliberate actor who brings about a comparable harm. In addition to this premise, I want to suggest, we ought also to accept the following claim: for cases in which one harms another due to one's negligent ignorance, the less negligent one's ignorance, the less responsible one is for that harm (other things being equal, of course) since responsibility for the harm caused decreases with decreases in the extent to which the actor is at fault, and, in particular, with decreases in the extent to which the actor is epistemically at fault. Recall that on the account I have presented in Section III above, epistemic faultlessness is best understood in terms of epistemic justification. So, if we accept my analysis of epistemic faultlessness as well as the above stated connection between epistemic fault and responsibility, it follows that the more one is e-justified in one's belief, the less responsible one is for any harm caused by those actions based upon the belief in question (again, other things being equal). If we also accept Greenawalt's claim that the responsibility-reducing factor negligent ignorance ought to be understood as a partial excuse, and I think that we should, then we ought also to accept the following claim: for cases in which one's actions are based upon some mistake of fact, the more one is e-justified in one's belief, the greater the extent to which one's wrongful actions based upon that belief are excused.
Although epistemic justification is typically construed as an all-or-nothing state, that is, either one is or one is not epistemically justified,(40) I want to propose that since we are using the notion to understand epistemic fault (and since clearly one can be more or less at fault for what one believes), it makes sense to speak of an agent as being more or less epistemically justified in his belief. Now, in the situation at hand, namely, that p is false and Fred is not e-justified in believing that p, if we imagine cases in which Fred's belief that p is increasingly e-justified, we would want to say that the action based upon the belief, namely, telling Charlie that Charlene has been unfaithful, is increasingly excused at each point along the continuum of epistemic justification. Thus, for example, at the point at which Fred's only fault was that he overlooked one piece of disconfirming evidence to which he had epistemic access (that is, he overlooked some other belief within his belief system that made it unreasonable for him to believe that p), his behavior is excused to a greater extent than it is at the lower point on the continuum of epistemic justification at which he failed altogether to subject his incoming belief to potentially disconfirming beliefs. As we move along the continuum of epistemic justification for Fred's belief, we get to a point at which Fred's belief is almost totally e-justified. Correspondingly, at this point, Fred's action is almost totally excused (given, of course, our assumption that his intentions are good). What do we say, then, about Fred's action at the point at which his belief is totally e-justified? It makes sense, I think, to say not that his action, namely, telling Charlie of Charlene's (alleged) unfaithfulness, is a-justified, but, rather, that the action is totally excused.
Greenawalt's analysis surprisingly precludes this possibility, the possibility that one's action can be totally excused due to a mistake of fact. Although there is nothing logically contradictory about Greenawalt's conclusion, it makes for a conceptually awkward result. He appears to accept, in all cases but one, namely, the case in which one's belief is totally e-justified, that if one's belief is false, the more one is e-justified in holding the belief the greater the degree to which one's actions based upon that belief are excused. However, Greenawalt refuses to accept the natural conclusion to which this analysis points: If one has a false belief and one's belief is totally e-justified, then one's actions based upon this belief are totally excused (assuming of course that one's actions would have been justified had one not been mistaken in one's belief). In short then, for purposes of conceptual clarity we ought to regard actions based upon (totally) faultless mistakes of fact as excused, not as justified.
Rejecting Greenawalt's claim that an action based on a faultless mistake of fact should be regarded as justified, not as excused, by law and morality has important implications for other parts of his argument. First of all, it is no longer obvious that the "actor's appraisal approach" to justification is superior to the "objective approach" to justification. Admittedly, however, Greenawalt's argument against the "actual facts approach" still stands.(41) If one's bad intentions lead accidently to good results, we should not say that the act is justified. However, if my arguments are cogent, neither should we say that the actions of the well-intentioned, blameless, but mistaken reasoner are justified, although they may be totally excused. Some combination of the two approaches appears apt. A-justification may require that the action be warranted by the facts and motivated by the facts as well.(42)
Secondly, if my arguments are correct, they significantly weaken Greenawalt's argument against the claim that justification and excuse can be distinguished by looking at the appropriateness of defense, intervention, and support. Greenawalt uses People v. Young in his attempt to show that it is not true "that a justified act may be supported but not stopped, whereas an excused act may be stopped, by its victim or an intervener, but not supported."(43) On the analysis presented here, Young's behavior was totally excused given that his beliefs about the situation were totally e-justified. Since Young's behavior was excused but not justified, the detectives were right to fend off his intervention. Hence, it turns out that Young actually supports distinguishing justification from excuse by looking at the appropriateness of defense, intervention, and support--the line of argument that Greenawalt calls into question.
In summary then, it appears as though the distinction between justification and excuse is blurred needlessly by Greenawalt's claim that a faultless mistake of .fact is sufficient to justify one's behavior. I have argued that his contention should be rejected because (1) it ignores the fact that the morally relevant costs of acting upon a mistaken belief typically exceed the morally relevant costs of holding a mistaken belief and (2) it leads to a conceptually awkward result. Having established a strong case for the untenability of Greenawalt's analysis, we are now in a position to see both that there is reason to think that there is some objective element which is necessary for a-justification (although this element is very unlikely to be sufficient) and that it is possible that justification and excuse can be distinguished based upon the rights of others to intervene with defense or support. However, in addition to the considerations presented above in response to Greenawalt's analysis, I contend that there is an even more forceful argument for holding that an action based upon a faultless mistake of fact is excused, not justified. It is to that I now turn.
V. The Argument From Responsibility
The argument I shall advance rests upon the following principle: a person is not responsible for an action to to the extent that the action is not voluntary.(44) Certainly, the principle needs refinement to avoid obvious objections. For example, we must distinguish the cases in which a person engages in an impermissible action through no fault of his own from those in which he purposely, knowingly, recklessly, or negligently puts himself in a situation in which he will "nonvoluntarily" behave in an impermissible manner. Thus, we might amend the principle to read: a person is not responsible for an action to the extent that the action is not voluntary, unless he is somehow culpable for finding himself in the situation at hand. For the purposes of this paper, we need not concern ourselves with the supplementary clause. It is not necessary that we delve into the intricacies of hard cases since the individual actors with whom we are concerned are, by hypothesis, blameless. Thus, we can avoid dealing with whether or not the person is to blame for what Holly Smith has called the "unwitting act" (the act which is the result of prior negligence, recklessness, etc.) in addition to being blamed for the "benighting act" (the initial negligent, reckless, etc., act).(45) The important question then for the present analysis is whether the behavior of our faultless reasoner should be construed as voluntary or as significantly nonvoluntary.
Let us return to Young's case, which is described above in Section I. Young, let us assume, was e-justified in believing that unless he came to the aid of the attacked youth, the pair of muggers would use unlawful force to cause physical harm to their victim. Recall that the attackers were not muggers after all but were instead undercover detectives making a lawful arrest. I contend that there is a sense in which Young's actions were not voluntary. To see that this is the case, it is necessary that we look at the connection between mistakes of fact and agent voluntariness.
In Harm to Self, Feinberg provides an acute analysis of the connection between mistakes of fact and agent voluntariness.(46) He contends, following Aristotle, that mistaken belief or ignorance "tends to diminish or de feat voluntariness...."(47) Thus, to the extent that Young's actions rested upon a (relevant) mistake of fact, we might conclude that his actions were not voluntary. Our reasoning would proceed as follows. Young might have chosen voluntarily to attack the assailants qua muggers; however, he did not choose voluntarily to attack the assailants qua undercover detectives. Young might appropriately say of the latter, "That's not what I meant to do at all!" Feinberg would call these descriptions of Young's actions "a thin act-description" and a "thick act-description," respectively. As Feinberg explains the matter, "acts corresponding to the thinner descriptions tend to be voluntary, whereas those corresponding to thicker descriptions, incorporating consequences into the action itself that were not expected by the actor because of his mistaken factual belief, are a good deal less than voluntary."(48) Thus, when we consider the thicker description of Young's actions, we see that there is in fact a clear sense in which his behavior was nonvoluntary. To use J. L. Austin's phrase, "it isn't fair to say he simply did A--he was really doing something different and A was only incidental, or he was looking at the whole thing quite differently."(49)
What description is relevant to the case at hand? We need to determine the appropriate description since the law cannot view Young's actions as both voluntary and nonvoluntary.(50) Given the circumstances of the case, I contend that the thicker description is the relevant one. Young was prosecuted for assaulting the men qua detectives, not qua muggers. His actions were wrong, the prosecution might argue, because he violated the rights of the detectives making a lawful arrest. But if this is the correct description of what Young did, namely, Young attacked the men qua detectives, Young's behavior cannot be construed as voluntary. (Although we continue to use the locution "Young's behavior," we should recognize that since the behavior in question, under the thicker description, must be regarded as significantly nonvoluntary, it follows that it was not, strictly speaking, Young's behavior after all.) Given that Young's behavior was significantly nonvoluntary, it cannot be the case that he was responsible for what he did. Consequently, Young's actions, which were based upon a faultless mistake of fact, are appropriately deemed excused, not justified. It is not the case that Young is blameless because he did the right thing; instead, although what Young was accused of doing was impermissible and, thus, he cannot even be deemed weakly justified, he is blameless because he was not responsible for his behavior.
VI. Perplexing Implications
The above conclusions, although I think they are correct, lead to perplexing problems. Imagine that, contrary to what actually happened, Young was correct about the men who attacked the youth. So, in this counterfactual world, all of the facts are the same, except that the two assailants were in fact muggers.(51) If Young is prosecuted for assault, he can advance an affirmative defense, namely, defense of another. In all likelihood, he would not even be prosecuted because his behavior was so clearly appropriate and, thus, justified. The issue of concern then is the following. In our counterfactual world, Young's behavior is to be construed as fully voluntary and morally and legally appropriate. In fact, counterfactual Young's behavior merits praise. On the other hand, it appears that the analysis I have presented commits us to holding that Young's behavior in the actual state of affairs (for which he was not responsible) cannot be said to be praiseworthy at all, for that would be like praising the behavior of a man whose hiccup distracted a gunman who was consequently disarmed. However, Young displayed the same intentions, desires, and, at one level, the same actions in each case.(52) Is this initial conclusion--that whether or not a person deserves praise depends upon factors external to his control--an acceptable one?
The situation at hand presents a problem for the moral philosopher because it looks to be the case that certain aspects of Young's moral status depend purely upon luck. It just so happens that in the actual state of affairs Young was particularly unlucky: his justified beliefs did not correspond to the facts as they were. In the counterfactual state of affairs, Young was lucky (in the sense that he was not unlucky in the manner that actual Young was unlucky) enough to have it turn out that his justified beliefs about the relevant features of the case were accurate. On the face of it at least, it appears that nothing for which Young was responsible in one case as opposed to the other can make for a distinction upon which we can justify the disparate evaluations to which my analysis seems to point, namely, that in the actual state of affairs, his behavior was nonvoluntary and excused and, thus, not the sort of thing that merits praise; in the counterfactual state of affairs, his actions were voluntary and strongly justified and, thus, praiseworthy. In short, the analysis seems committed to the claim that factors beyond one's control can be determinate in moral evaluation.(53)
The difficult issues to which my analysis has led are very similar to the type of concerns addressed in Thomas Nagel's article, "Moral Luck."(54) Nagel puts forth several ways in which "the natural objects of moral assessment are disturbingly subject to luck."(55) Most important for our purposes is his discussion of "luck in the way one's actions and projects turn out."(56) Although the luck upon which the above analysis, assessment of Young's actions depends does not correspond exactly to any of Nagel's sub-categories of unlucky results, it corresponds most closely to the type of luck involved in what he calls "cases of decision under uncertainty."(57) Like Young, people are often forced to make decisions under conditions of uncertainty, and the results of acting upon their decisions are highly subject to luck. Nagel argues that for cases of this sort "[o]ne kind of assessment of the choice is possible in advance, but another kind must await the outcome, because the outcome determines what has been done."(55) Nagel's conclusions are relevant to the issue at hand in the following manner. If we accept Nagel's conclusion that factors beyond one's control are relevant to moral assessment, then we should be prepared to accept that in Young's case factors of this sort are determinate. For the only difference between the actual and counterfactual conditions surrounding Young is, roughly speaking, the way things happened, as a matter of luck, to turn out.
Nagel's analysis rests upon an appeal to our moral experience.(59) He correctly points out that we blame ourselves and others more for being negligent when things unluckily turn out badly than we do when there are luckily no bad results of our negligent behavior. Accordingly, as Nagel suggests, there is a conflict between our luck-influenced practices and the conclusions we arrive at from rational reflection about responsibility; for, in our more reflective moments, we readily deny that a person can be to blame for that which is not under his control. What conclusions might we draw from our recognition of the conflict? Nagel infers that "one is morally at the mercy of fate, and it may seem irrational upon reflection, but our ordinary moral attitudes would be unrecognizable without it."(60) Thus, according to Nagel, we must concede that the notion of luck is central to genuine moral assessment.
One way to counter Nagel's conclusions would be to provide an alternative analysis of our everyday blaming behavior. There is, I think, such an analysis. I want to suggest that the connection between luck and morality in our everyday blaming practices has less to do with the nature of morality than it does with the cognitive limitations of human beings. Arguably, one reason that we blame the negligent person for whom things turn out badly to a greater extent than we blame the negligent yet lucky individual is that we can almost never be certain of the latter that he was actually negligent or, better yet, that he was sufficiently negligent to justify our blaming him to the extent that we blame the individual for whom we have evidence of the fact that he was negligent. On this analysis, bad consequences of actions serve merely as prima facie signs of our own negligent behavior and the negligent behavior of others. Thus, our everyday blaming practices are driven to a significant extent by epistemic considerations in addition to metaphysical ones. In cases in which we are negligent but lucky (for example, we leave the bath water running with the baby in it, and the baby is unharmed),(61) I contend that if we only realized how close we came to tragedy, we would blame ourselves as we would if things had turned out unluckily. Since our blaming practices are significantly influenced by epistemic considerations, we are often able to avoid the blame we actually deserve in situations in which we were negligent but things turned out luckily for the better.
In contrast to the cases with which this paper is concerned, most of Nagel's examples deal with persons who are blameworthy to some extent or other. The primary question he attempts to answer is whether a person can be more or less blameworthy due to factors beyond her control. Recall that the analysis presented above is not strictly concerned with issues of blame worthiness. However, it is concerned with issues of praiseworthiness, and, as I think Nagel correctly assumes, the analysis of the two should be quite similar in form.(62) That is, if one holds that praiseworthiness is subject to luck, then one should hold that blameworthiness is subject to luck as well. Thus, unless we can ameliorate the problems implied by the above analysis of actions based upon a faultless mistake of fact, we shall be stuck with an account of praiseworthiness and an account of blameworthiness similar to Nagel's analysis, an analysis I am inclined to reject. Fortunately, I think that there is a way in which we might avoid the problems regarding the relation between luck and Young's praiseworthiness. It is to this solution that I turn now.
Recall that the problem is to avoid the counter intuitive result seemingly implied by the above analysis of actions based upon a faultless mistake of fact, namely, that whether or not a person is deserving of praise can depend upon factors beyond her control. I contend that we can avoid this result. Our intuitions correctly tell us that Young was praiseworthy in both the actual and the counterfactual world. Indeed, few people are willing to risk bodily injury, let alone death, for the sake of a stranger. The difficult part of any such analysis lies in explaining why, or better yet, in virtue of what characteristic, the actual Young is deserving of praise. Most importantly, it appears as though the analysis must provide an explanation without appealing to what actual Young did, unless we are to abandon the above argument from which we concluded that actual Young's behavior, under the relevant description, was nonvoluntary and, thus, not a proper object of praise. However, I want to suggest that there are two types of appeal to Young's behavior, and one of these appeals will not conflict with the conclusions above. Or so I shall argue.
Although we cannot appeal to a direct connection between Young's actions and his praiseworthiness, we can appeal to a very important indirect connection. The fact that Young acted as he did gives us conclusive evidence regarding how he would have acted in the counterfactual situation, that is, in the situation in which the assailants were indeed muggers. Accordingly, actual Young is praiseworthy because in the relevant circumstances (circumstances subjectively indistinguishable from those in which he in fact acted) he would have acted in such a manner that his actions would have been praiseworthy. There is no reason to think that whether Young is praiseworthy depends solely upon whether his actual behavior is praiseworthy. Instead, I want to suggest that Young is praiseworthy for being the type of person who would put himself at risk to save an innocent victim. Thus, although actual Young's behavior is of no substantive import in determining whether or not he is praiseworthy, his behavior has indirect, epistemic significance: by looking at his behavior in actual situations we can infer how he would have acted had the circumstances been different. Hence, we may accept our conclusion that Young was not responsible for what he did and, so, that his actions should be deemed excused, not justified, by law and morality. Nonetheless, we may conclude consistently in addition that he deserves praise to the same extent to which he would have deserved praise had it actually turned out that the assailants were muggers. Thus, we are no longer left with perplexing implications of the above analysis.
I have argued that an action based upon a faultless mistake of fact is appropriately deemed excused, not justified, by law and morality. Just because it is true both that one is e-justified in believing that justificatory circumstances obtain and one's action would be a-justified were one not mistaken, it does not follow that one is a-justified in acting upon such a belief. If my arguments are cogent, there is good reason to think that mistaken, though reasonable, beliefs of this sort under mine agent voluntariness. The fact that actions based upon mistaken beliefs are significantly less than voluntary entails that the actors are not responsible for their behavior when their actions are understood under the relevant descriptions. Since they are not responsible for what they have done, we should deem their actions excused, not justified. Although it appears that we are left with the troubling conclusion that since they are not responsible we cannot see such agents as praiseworthy, I have argued that this conclusion is a mistaken one. It relies upon the unacceptable premise that a person can be praiseworthy only for what he has actually done. In fact, although a person may not be praiseworthy for what he has done, it may be the case that we can infer from his actions that he is the sort of person who deserves praise. Since we know what he would have done had the facts been different, we are in a position to praise him for his disposition to do the right thing. So, in the end, we are left with a coherent analysis of actions based upon a faultless mistake of fact, an analysis that meets our considered judgments regarding the moral praiseworthiness of those actors whose actions would have been strongly justified had the facts been as they were epistemically justified in believing them to be.
I want to thank Joel Feinberg, Christopher Griffin, Jean Hampton, Keith Lehrer, Joel Pust, Linda Radzik, David Silver, Lori Speagle, Christopher Wellman, and especially Thomas Hudson for comments on earlier drafts and for many helpful discussions regarding the subject matter of this paper. I am also grateful to the editors of Criminal Justice Ethics for many useful suggestions and stimulating questions. (1) W.R. LAFAVE & A.W. Scott, JR., Substantive Criminal Law 575 (1986). (2) Id. Surprisingly, this is from their chapter, "Justification and Excuse." They seem unaware of the possibility of an affirmative defense based upon mistake of fact. (3) Regina v. Morgan, A.C. 182 (1976), 2 All E.R. 347 (1975). The facts of the case are as follows. One of the four appellants, Morgan, had invited the others, all younger than he, to his home to have sexual intercourse with his wife. Allegedly, the three other appellants were led to believe that Mrs. Morgan would consent to their doing so. According to them, they were told that any resistance on the part of Mrs. Morgan should be regarded as a sign of her pleasure, not as a sign of her opposition to the sexual intercourse. According to Mrs. Morgan, she was held down while each of the three young appellants forcibly had intercourse with her. Mr. Morgan was not charged with rape but with aiding and abetting the others. The controversy which ensued focused upon the issue of whether a sincere, though unreasonable, belief that an alleged victim had consented was sufficient to negative defendant intent. Lord Hailsham had argued that it was: Once one has accepted, what seems to me abundantly clear, that the prohibited act in rape is non-consensual sexual intercourse, and that the guilty state of mind is an intention to commit it, it seems to me to follow as a matter of inexorable logic that there is no room either for a ,defense, of honest belief or mistake, or of a defense of honest and reasonable belief and mistake. Either the prosecution proves that the accused had the requisite intent, or it does not. In the former case it succeeds, and in the latter it fails. Since honest belief clearly negatives intent, the reasonableness or otherwise of that belief can only be evidence for or against the view that the belief and therefore the intent was actually held .... Morgan, 2 All E.R. at 361. As E. Curley points out, the Morgan ruling created an uproar in the British and Australian presses, where the ruling was referred as a "rapists, charter." Curley, Excusing Rape, 5 PHIL. PUB. AFF. 325 (1976). (4) People v. Young, 11 N.Y.2d 274, 183 N.E.2d 319, 229 N.Y.S.2d 1 (1962), rev,d, 12 A.D.2d 262, 210 N.Y.S.2d 358 (1961). (5) MODEL PENAL CODE 211.1 (1) (Proposed Official Draft 1962), reprinted in S.H. KADISH & S.J. SCHULHOFER, CRIMINAL LAW AND ITS PROCESSES at 1179-1248 (1989). (6) Young was convicted of two crimes of third degree assault, but his conviction was reversed by the New York Supreme Court, Appellate Division. However, the New York Court of Appeals reversed the order of the Appellate Division, stating: While the doctrine espoused by the majority below [that one who intervenes in a struggle between strangers under the mistaken but reasonable belief that he is protecting another who he assumes is being unlawfully beaten is thereby exonerated from criminal liability] may have support in some States, we feel that such a policy would not be conducive to an orderly society. We agree with settled policy of law in most jurisdictions that the right of a person to defend another ordinarily should not be greater than such a person's right to defend himself Young, 183 N.E.2d at 319-320. (7) N.Y. PENAL LAW 35.15, quoted in S.H. Kadish & S.J. SCHULHOFER, supra note 5, at 875. In point of fact, this particular articulation of the affirmative defense of reasonable mistake of fact was a result of the difficulties surrounding Young. (8) Greenawalt, The Perplexing Borders of Justification and Excuse, 84 COLUM. L. REV., at 1908 (1984). (9) I should note at the outset that the primary thesis of my paper stands in direct opposition to the Model Penal Code approach. Section 3.09 (2) reads as follows: When the actor believes that the use of force upon or toward the person of another is necessary for any purposes for which such belief would establish a justification under Sections 3.03 to 3.08 but the actor is reckless or negligent in having such belief or in acquiring or failing to acquire any knowledge or belief which is material to the justifiability of his use of force, the justification afforded by those Sections is unavailable in a prosecution for an offense for which recklessness or negligence, as the case may be, suffices to establish culpability. MODEL PENAL CODE 3.09 (2), supra note 5. Thus, on the Model Penal Code approach, actions based upon a faultless mistake of fact are appropriately deemed justified, not excused. Furthermore, on this approach, one can be at fault epistemically (due to negligent or reckless belief acquisition or belief maintenance) and, yet, be deemed justified if the culpability conditions with respect to a material element of the offense require that the person in question acted purposely or knowingly. (10) Greenawalt, supra note 8, at 1897. (11) J.L. Austin, A Plea for Excuses in Philosophical Papers 176 (J.O. Urmson & G.J. Warnock ed. 1979.) (12) D. Husak suggests that we more commonly mean the latter: "The conduct of a defendant who acts under a justification may be exemplary and socially encouraged, but more typically is permissible and socially tolerated." D. Husak, PHILOSOPHY OF CRIMINAL LAW 189 (1987). (13) J. FEINBERG, HARM TO OTHERS 109 (1984). I am grateful to Professor Feinberg for convincing me that excused actions cannot be cases of wrongdoing. (14) I owe to C. Griffin the general idea that the distinction between justification and excuse might be supported on these grounds. (15) J. FEINBERG, The Expressive Function of Punishment, in DOING & DESERVING: ESSAYS IN THE THEORY OF RESPONSIBILITY 100 (1970). Feinberg's analysis is actually more complicated than the quotation provided might suggest. In fact, according to Feinberg, punishment expresses the community's condemnation of the criminal and his actions. Feinberg characterizes condemnation as "a kind of fusing of resentment [the various vengeful attitudes] and reprobation [the stern judgment of disapproval]." He proposes this characterization at 101. (16) Greenawalt, Distinguishing Justifications from Excuses, 49 L. & CONTEMP. PROB. 90 (1986) (emphasis added). (17) K. GREENAWALT, CONFLICTS OF LAW AND MORALITY 287 (1987). (18) See, e.g., K. LEHRER, THEORY OF KNOWLEDGE (1990). (19) See, e.g., A. Goldman, What is Justified Belief, in JUSTIFICATION AND KNOWLEDGE 1-23 (George Pappas ed. 1979). (20) This is M. Engel's account of process reliabilism. M. Engel, Jr., Personal and Doxastic Justification, 67 Phil. Stud. 141 (1992). Indefinite probabilities differ from definite probabilities in the following way. The former specify the probability that something of a given class (for example, the class of perceptual beliefs) will also be a member of another class (for example, the class of true beliefs). In contrast, the latter specify "the probabilities that particular propositions are true or that a certain state of affairs obtains." J. POLLOCK, CONTEMPORARY THEORIES OF KNOWLEDGE 97 (1986). (21) The internalist theory of J. Pollock proves to be an exception to my generalization. Pollock holds that other internal states besides belief states are relevant to epistemic justification. However, Pollock does respect the intuition that I contend is central to the internalist account: the person who is doing the best she can in belief acquisition and belief maintenance is justified. He writes, for example, "If a person has no reason to be suspicious of his apparent memories, then he is doing the best he can if he simply accepts them. Consequently, he is justified." J. POLLOCK, supra note 20, at 50. I should note, however, that there is considerable debate regarding which pair of epistemological concepts appropriately supports the distinction between doing the best one can epistemically given one's resources and doing the best one can epistemically all things considered. Thus, there are several proposals which differ from my account. For example, Goldman suggests the distinction between weak and strong justification. A. Goldman, Strong and Weak Justification, 2 PHILOSOPHICAL PERSPECTIVES 51-69 (1988). J. Pollock proposes a distinction between subjective and objective justification. Pollock, A Plethora of Epistemological Theories, in JUSTIFICATION AND KNOWLEDGE 93-113 (G. Pappas ed. 1979). H. Komblith distinguishes between epistemically responsible action and belief formation founded upon "inferences licensed by rules of ideal reasoning." Kornblith, Justified Belief and Epistemically Responsible Action, 92 Phil. Rev. 34 (1983). (22) See, e.g., Lehrer & Cohen, Justification, Truth and Coherence, 55 Synthese 191-207 (1983) and Cohen, Justification and Truth, 46 PHIL. STUD. 279-96 (1984). (23) Alston, Concepts of Epistemic Justification, 68 Monist 67 (1985). (24) Id. at 67-68. (25) L. BONJOUR, THE STRUCTURE OF EMPIRICAL KNOWLEDGE 38 (1985). (26) Id. at 38-45. (27) As the editors of Criminal Justice Ethics have pointed out to me, it is possible that the individual suffering from an insane delusion will turn out to be epistemically faultless on my account since his beliefs are internally epistemically justified. This, I think, is the right result. In some cases at least, the criminally insane individual is doing the best he can with what he has and is thus faultless, though mistaken. However, just because internal epistemic justification is necessary and sufficient for epistemic faultlessness, it does not follow that my account cannot accommodate the distinction in the law between faultless mistake of fact and insanity. The editors are correct that it appears to be externalist considerations which distinguish the two. For example, we must ask if the agent's beliefs rested merely upon a localized error or if he was instead systemically mistaken about the world. Thus, it is important to note that while my account of faultlessness is an internalist one, the analysis of faultless mistake of fact appeals to both internalist and externalist considerations. (28) M. Moore also appeals to the distinction between the justification of beliefs and the justification of actions in his analysis of Greenawalt's account. Of the reasonable, but mistaken, agent, he writes: "[He] may idiomatically be said to have been |justified, in doing what he did because he was epistemically justified in believing what he believed." Moore concludes, however, that "[His] action is not justified; it was wrongful." M. MOORE, ACT AND CRIME: THE PHILOSOPHY OF ACTION AND ITS IMPLICATIONS FOR CRIMINAL LAW 183 (1993). (29) Greenawalt, supra notes 8 & 16. Greenawalt does not use the terminology that I have used here, but I see no reason why he should find it unacceptable. (30) Greenawalt, supra note 8, at 1908 & supra note 16, at 102. In the latter, Greenawalt refers to actions based on a faultless mistake of fact as actions based upon "the actor's blameless perception of the facts." (31) Greenawalt, supra note 16, at 101-103. (32) Fletcher, The Right and the Reasonable, 98 HARV. L. REV. 949-82 (1985). For example, Fletcher writes, "Justification-harmony with the Right--is an objective phenomenon. Mere belief cannot generate a justification, however reasonable the belief might be." Fletcher makes this point at 972. (33) Greenawalt, supra note 8, at 1918-27. G. Fletcher, for one, holds that justification and excuse can be distinguished in this fashion. G. FLETCHER, RETHINKING THE CRIMINAL LAW 76062, 830, 859 (1978). (34) Greenawalt, supra note 16, at 102. (35) Greenawalt, supra note 8, at 1908. (36) Id. at 1919. I might point out that it is not clear that we would "expect and hope that a similar actor with a similar set of available facts would make the same choice." Actually, it would have been preferable had Young not chosen to interfere in the situation. As will later become evident, the conflicting intuitions about the case arise because "what Young did" is ambiguous. We do want people to try to help those in need, but we do not want individuals to interfere with lawful arrests. However, I leave these issues aside until a later stage in the paper. (37) Id. at 1916 n.55. (38) Thus, we have a case in which (internal) epistemic justification and sincerity of belief come apart. Accordingly, we should not accept the following remarks of Chief Judge Wachtler in his opinion, People v. Goetz, 68 N.Y.2d 111, 497 N.E.2d 50 (1986): "Interpreting the statute to require only that the defendant's belief was ,reasonable to him,' as done by the plurality below, would hardly be different from requiring only a genuine belief; in either case, the defendant's own perceptions could completely exonerate him from any criminal liability." As my example shows, a person can have a sincere belief which is not even reasonable to him. Roughly, the fact that the belief is not reasonable to him entails that he is unjustified in holding the belief. Admittedly, however, the fact that a belief is (internally) unjustified may come to bear on our conclusions regarding the sincerity of the defendant's belief. (39) Greenawalt, supra note 16, at 95. He makes this point at 103 as well. (40) But see Alston, supra note 23, at 59. (41) Greenawalt, supra note 16, at 101. (42) Fletcher suggests this approach to justification. He writes, "The fallacy in [the] reasoning [which moves from the acceptable claim that self-defense must include a subjective element, viz., justificatory intent, to the assimilation of putative self-defense with self-defense] is assuming that self-defense must be either objective or subjective, not both. The contrary view, which holds that self-defense affects the rights of victims and third parties, calls for an objective dimension of wrongful aggression as well as a subjective dimension of justificatory intent." G. Fletcher, supra note 33, at 768. Greenawalt criticizes this sort of approach: "The position that justification depends on consonancy with both subjective appraisal and actual fact fails to explain adequately why the absence of either support is enough to term an act unjustified, and why the presence of both is needed to make an act justified." Greenawalt, supra note 16, at 102. These concerns, unfortunately, are beyond the scope of this paper. (43) Greenawalt, supra note 8, at 1918. (44) One important feature of the principle is the following. It construes neither responsibility nor voluntariness as absolute notions. Thus, individuals can be more or less responsible for their actions and their actions can be more or less voluntary. On this account, to show that an individual is not responsible enough to be held criminally liable for what he has done, it will not be necessary to show that his actions were involuntary. Rather, it will be sufficient to show that his actions were significantly nonvoluntary. My analysis relies upon J. Feinberg's very insightful discussion of these matters. J. Feinberg, Harm to Self 104-127 (1986). (45) Smith, Culpable Ignorance, 92 PHIL REV. 547-571 (1983). (46) J. Feinberg, supra note 44. (47) Id. at 152. (48) Id. at 129. (49) J.L. Austin, supra note 11, at 176 (emphasis added). (50) As J. Feinberg points out, a significant amount of line drawing is necessary in the legal realm, especially surrounding issues of responsibility. J. FEINBERG,, roblematic Responsibility in Law and Morals, in DOING & DESERVING: ESSAYS IN THE THEORY OF RESPONSIBILITY 25-37 (1970). (51) This is not quite right. There are severe difficulties with trying to conceive of a possible world in which only one fact is altered. D. Lewis writes, We might think it best to confine our attention to worlds where kangaroos have no tails and everything else is as it actually is; but there are no such worlds. Are we to suppose that kangaroos have no tails but that their tracks in the sand are as they actually are? Then we shall have to suppose that these tracks are produced in a way quite different from the actual way. Are we to suppose that kangaroos have no tails but that their genetic makeup is as it actually is? Then we shall have to suppose that genes control growth in a way quite different from the actual way (or else that there is something, unlike anything there actually is, that removes the tails). And so it goes; respects of similarity and difference trade off. If we try too hard for exact similarity to the actual world in one respect, we will get excessive differences in some other respect. D. LEWIS, COUNTERFACTUALS 9 (1973). (52) As J. Feinberg holds, matters of luck pervade this internal arena as well. He suggests that although one would think that we must avoid arbitrariness in moral matters, "it is a mistake to think that by restricting responsibility to an inner jurisdiction we can thereby make precise its vagueness and eliminate its contingencies." J. Feinberg, supra note 50, at 37. Therefore, a simple appeal to intentions will not solve the present problem since intention formation itself is, in many circumstances, highly subject to matters beyond the actor's control. (53) Even if this were true, and I shall argue that it is not, we should be careful here regarding any conclusions we might draw from this claim. As S. Kagan correctly maintains, showing that some factor is morally determinate in one circumstance does not show that it is determinate, or even relevant, in all circumstances. This would apply to the factor at issue in the case at hand, namely, how things happen to turn out. Kagan, The Additive Fallacy, 99 Ethics 5-31 (1988). (54) T. NAGEL, MORAL LUCK, in MORTAL QUESTIONS 21-38 (1979). (55) Id. at 28. (56) Id. (57) Id. at 29. (58) Id. at 30. (59) Id. at 28-31. (60) Id. at 34. (61) This is T. Nagel's example. Id. at 30-31. (62) Thus, I follow E. Beardsley, who writes, "Praise and blame will be treated as correlative concepts such that, for everything that is said about one, a corresponding statement about the other could be made ...." Beardsley, Determinism and Moral Perspectives, in REASON & RESPONSIBILITY 409 (J. Feinberg ed. 1989). Beardsley's article originally appeared in 21 PHIL. PHEN. RES. 1-20 (1960).
|Printer friendly Cite/link Email Feedback|
|Author:||Price, Terry L.|
|Publication:||Criminal Justice Ethics|
|Date:||Jun 22, 1993|
|Previous Article:||Religious grounds in liberal politics.|
|Next Article:||Against Excess: Drug Policy for Results.|