|
Article Excerpt
I. INTRODUCTION
The Supreme Court's decision in Lawrence v. Texas (1) has intensified the debate over the constitutionality of statutes that impose strict criminal liability for sexual abuse of children. (2) Critics of these statutes have long been troubled by the possibility that an actor might be convicted despite reasonably having been convinced that his partner had reached adulthood. (3) This criticism has taken on new force with Lawrence's apparent "constitutionalization of an individual's right to sexual intimacy." (4) If sexual relations among unmarried adults not only are lawful, but are constitutionally protected, then--according to the critics of strict liability--due process ought to forbid the imposition of criminal liability on an actor who has no reason to suppose that his or her partner in consensual sex has not reached adulthood. (5) What Lawrence appears to require, on this view, is that the Government prove some culpable mental state regarding the age element in the crime of sexual abuse. (6)
So far, the courts have proven resistant to this argument, (7) perhaps with good reason. The obvious alternatives to strict liability are the recklessness and criminal negligence standards, at least in the many states that have adopted the Model Penal Code's approach to culpability. (8) But recklessness and criminal negligence are problematic in this setting. The trouble is that both of these standards would require jurors to calculate the acceptable level of risk on a case-by-case basis. (9) Both standards, in other words, would require jurors to decide--based primarily on the jurors' own estimates of the gravity of the harm that accompanies sexual abuse--what probability a reasonable person would be willing to accept that his or her partner is underage. (10) Most jurors are ill-equipped to make this calculation. Worse, their efforts to make this calculation almost inevitably will be skewed by evidence of the victim's unchastity. (11)
The solution to this problem lies in recognizing a new criterion of culpability--a new species of mens rea--that provides an alternative to strict liability, on the one hand, and to negligence and recklessness, on the other. Under this new criterion of culpability, the legislature, not the jury, would be responsible for deciding what level of risk is acceptable. The legislature would calculate the acceptable level of risk based on legislative assessments of both (1) the gravity of the harm that results from sexual abuse and (2) the general social utility of sexual intercourse. The jury then would be responsible only for determining whether the defendant either culpably failed to perceive or consciously disregarded a risk that exceeded this acceptable level. Thus the jury, instead of being asked to decide whether the defendant was or should have been aware of a "substantial and unjustifiable" risk that his or her partner was underage, (12) would be asked to decide simply whether the defendant was or should have been aware of, say, a realistic possibility that his or her partner was underage.
There are precedents for defining legislatively the acceptable level of risk. For example, the Model Penal Code's self-defense provision--like the self-defense provisions of most state criminal codes--makes the defense available only to actors who "believe" that force is necessary. (13) One effect of this provision is to make the defense unavailable to an actor who is aware of a probability greater than 50% that her use of force is not necessary. (14) The Model Penal Code's influential definition of "knowingly" also defines a fixed level of acceptable risk, albeit a very high one. (15) A person acts "knowingly" only if he "is aware that it is practically certain that his conduct will cause [the proscribed] result." (16)
This Article makes the case for a new variety of mens rea that is the obverse of "knowingly," that is, a new variety of mens rea that establishes a very low, fixed level of acceptable risk. In making the case for this new mens rea, the Article will not return to first principles. It will assume that strict liability, as traditionally defined, (17) is unacceptable for offenses that carry severe punishment. (18) It will also assume, as do legislatures and courts, the basic acceptability of criminal negligence, recklessness, and even so-called general intent as criteria of culpability. (19) Indeed, the better part of this Article will be devoted to showing how the new criterion of culpability is situated in relation to these existing criteria.
Part II.A begins with a brief analysis of recklessness and criminal negligence. This Part develops a kind of standard model for evaluating the justifiability of risk-creating conduct. When the legislature adopts recklessness or criminal negligence as the mens rea for the "social harm" element of an offense, it effectively delegates to the jury the entire responsibility for deciding whether the defendant's conduct was unjustifiable. Accordingly, in defining recklessness and criminal negligence, the legislature provides the jury with formulae for deciding whether the risk was unjustifiable. It is these formulae that will serve as a standard model and a starting point.
Part II.B explains how the legislature implements this standard model when it creates general intent offenses, whose only culpable mental states attach to conduct and attendant-circumstance elements. (20) Focusing on two particular general intent offenses, drunk-driving homicide and drug-induced homicide, this Part will show how these and other general intent offenses can be interpreted as representing antecedent legislative determinations that certain forms of conduct are reckless per se or negligent per se. Put another way: when the legislature creates a general intent offense, it effectively takes upon itself the responsibility for making the same justifiability calculations that are delegated to the jury in cases of negligence and recklessness. This Part demonstrates that there is nothing exceptionable about substituting an antecedent legislative determination of justifiability for the jury's case-specific determination.
Part III explains why none of the criteria of culpability considered so far--not recklessness, nor criminal negligence, nor general intent--will work in sexual abuse cases. Recklessness and criminal negligence founder on the requirement that the jury assign a value to the gravity of the harm in defining the acceptable level of risk. As this Part will show, even appellate judges have proven incapable of putting aside evidence of the victim's unchastity in defining the acceptable level of risk in sexual abuse cases. No better can be expected of jurors. Nor can this problem be solved by the creation of a general intent offense; the facts that signal the presence or absence of a risk that the victim is underage are too various and too subtle to be captured in a per se rule.
Part IV constructs an argument for defining legislatively the acceptable risk of sexual imposition. After explaining how this hybrid approach to the question of justifiability would operate, this Part illustrates the point with a brief analysis of the "knowingly" and "reasonable belief" criteria of culpability, both of which employ the hybrid approach. Then it explains why sexual abuse lends itself to legislative calculation of the acceptable risk and how exactly a fixed probability threshold would operate in the sexual abuse context. Finally, Part V argues that a rule requiring proof that the defendant "was or should have been aware" of a probability exceeding the threshold defined by the legislature provides sufficient assurance of the defendant's culpability.
II. TWO MODELS FOR CALCULATING JUSTIFIABILITY
A. RECKLESSNESS AND NEGLIGENCE
At the core of the Model Penal Code's recklessness and criminal negligence standards lies the idea of unjustifiable risk. (21) Under the Code, a defendant is reckless if he "consciously disregards a Substantial and unjustifiable risk that the material element exists or will result from his conduct." (22) By comparison, he is criminally negligent if he "should be aware of a substantial and unjustifiable risk that the material element exists or will result from his conduct." (23) Both definitions, then, require the jury "to examine the risk and the factors that are relevant to its ... justifiability" for the sake of determining whether the risk posed by the actor's conduct was, in fact, unjustifiable. (24)
It would be natural to suppose that, despite this shared focus on the justifiability of the risk, recklessness and criminal negligence differ fundamentally even in how the justifiability of the risk is measured. Criminal negligence, after all, is said to be an objective standard of culpability, while recklessness is said to be a subjective one. (25) Thus, it might seem as though the definition of recklessness ought to require the jury to address the question of justifiability "from the point of view of the actor's perceptions." (26) Likewise, it might seem as though the definition of negligence ought to require the jury to address the question of justifiability "in terms of the objective view of the situation as it actually existed." (27) Indeed, that is just what the commentary to the Model Penal Code says. The commentary suggests that though the question of justifiability plays a pivotal role in the definitions of both recklessness and negligence, the perspective from which the jury addresses this question differs depending on which standard is being applied. (28)
This aspect of the commentary, though, vastly overstates the differences between recklessness and negligence. First of all, the negligence standard is less objective than the commentary implies. The very terms of the Code's definition of "negligently" require the jury to consider "the circumstances known to [the actor]" in addressing the justifiability of the risk. (29) Thus, as Professor Peter Low has explained:
In spite of its concentration on objective components, the baseline for negligence is the context as the actor perceived it. Negligence, therefore, involves a subjective inquiry (what the actor actually knew about the context) and an objective inquiry (the inferences that should have been drawn from what the actor knew). (30)
Furthermore, the Code's definition of negligence could not really be otherwise in this respect. Any standard of liability that "emphasizes the justifiability of acting from an ex ante perspective" (31) must take into account what the actor knew when she acted. (32)
The Code commentary also overstates the degree to which the recklessness standard requires the jury to evaluate the risk from the defendant's perspective. It is true, of course, that in applying the Code's definition of recklessness the jury must take into account what the defendant knew--that is, "the circumstances known to him"--as it must when it applies the Code's definition of negligence. (33) Beyond this, however, there is nothing subjective about the justifiability calculus. (34) The principal subjective component of the recklessness standard--the requirement that the actor be "conscious[]" of the risk--bears no relation to the requirement that the risk be unjustifiable. The recklessness standard does not, for example, appear to require proof that the defendant actually was aware that the risk was unjustifiable. (35) Nor does the recklessness standard appear to contemplate that the defendant's mistaken factual beliefs (36) or idiosyncratic moral values will play a role in the calculus. (37) Thus, the recklessness and negligence standards require more or less the same analysis of the justifiability of the risk.
Perspective aside, though, what does it mean for a risk to be unjustifiable in the required sense? Recent tort scholarship has emphasized the availability of different approaches to defining the acceptable or reasonable level of risk-imposition in cases of negligence. One approach, for example, would require the jury to apply "various justice-based standards that take into account the rights and relationships among the parties." (38) Another approach, often associated with the "Learned Hand formula," would require the jury to conduct "a straightforward balancing of costs and benefits." (39)
Whatever the benefits of other approaches, the Model Penal Code's definitions of recklessness and negligence plainly appear to adopt the straightforward balancing approach. (40) This is evident, for example, in the definitions' very use of the word unjustifiable--rather than, say, unreasonable or unacceptable--to define the acceptable level of risk-imposition. (41) In this setting, the word unjustifiable implies that a risk can be justified, or offset, by countervailing social benefits. The same implication can be found in the definitions' command that jurors consider the "purpose of the actor's conduct" along with the "nature and degree" of the risk. (42) Finally, this reading of the negligence and recklessness standards jibes with the Code's version of the choice-of-evils defense, which measures the justifiability of the risk according to a straightforward balancing. (43) It requires the defendant to prove, as an element of the defense, that "the harm or evil sought to be avoided by [the] conduct is greater than that sought to be prevented by the law defining the offense charged." (44)
When the acceptability of a risk is defined by a straightforward balancing of aggregate costs and benefits, it is possible to express the standard of conduct in rough mathematical terms. In torts, for example, where the question of negligence often hinges on the adequacy of safety measures adopted by the defendant, courts sometimes define the acceptable risk in terms of three variables: (1) the probability that the harm will occur, (2) the gravity of the resulting injury if it does, and (3) the cost of taking precautions that would have prevented the harm. (45) On the cost side of the balance, then, is the harm, discounted by the likelihood that it would occur. And on the benefit side is the money saved by the defendant in foregoing precautions that could have prevented the harm. This is the Hand formula, from United States v. Carroll Towing Co.: (46) "[I]f the probability is called P; the injury, L; and the burden, B; liability depends upon whether B is less than L multiplied by P: i.e., whether B < PL." (47)
In criminal law, the benefits side of the equation is less likely to be framed in terms of the "burden of adequate precautions" than in terms of the "nature and purpose of [the] conduct" generally. (48) This, of course, is how the Model Penal Code's definitions of recklessness and criminal negligence define this variable. This different formulation of the social utility variable reflects the fact that the social utility of the actor's conduct in a criminal case less often will depend on whether the defendant has taken "adequate precautions" and more often will depend (in the words of the Code commentary) on whether and to what extent "the actor is seeking to serve a proper purpose [by his conduct], as when a surgeon performs an operation that he knows is very likely to be fatal but reasonably thinks to be necessary because the patient has no other, safer chance." (49) Thus, in the criminal law: if the probability of the harm occurring is called P, the gravity of the harm is called G, and the social utility of the conduct is U, then conduct is unjustifiably risky if PG > U. (50)
In summary, then, the widely accepted Model Penal Code definitions of recklessness and criminal negligence require the jury to decide whether the risk created by the defendant's conduct was unjustifiable. And this assessment of the risk's justifiability takes the same basic form regardless of whether the jury is applying a recklessness standard or a criminal negligence standard. In either case, the jury is required to assess the risk on the basis of "the circumstances known to [the defendant]" when he or she acted. (51) And, in either case, the assessment of the risk requires the jury to determine whether the risk posed by the defendant's conduct was "justified" by the countervailing social benefits. In other words, the assessment of the risk requires the jury to balance the aggregate costs and benefits in roughly the way described in the Hand formula.
B. GENERAL INTENT AS NEGLIGENCE PER SE
Crimes of negligence and recklessness are only two of the ways that legislatures target unjustified risk imposition. Legislatures also target unjustified risk imposition by creating so-called general intent offenses, whose only culpable mental states attach to conduct and attendant-circumstance elements. (52) These general intent offenses are relevant to our inquiry because they show that the legislature sometimes can make for itself the very justifiability calculation that is delegated to the jury in cases of negligence and recklessness. Offenses of this kind are utterly commonplace, as we will see. But the relationship between these general intent offenses, on the one hand, and offenses of negligence and recklessness, on the other, can best be illustrated by two particular offenses: drunk-driving homicide and drug-induced homicide.
Most state criminal codes, in addition to the usual general prohibitions on reckless and criminally negligent homicide, now have vehicular-homicide statutes in which the driver's liability hinges exclusively on his or her intoxication at the time of the fatal accident. (53) These statutes do not require the jury to make a determination that the defendant was reckless or negligent with respect to the result element--that is, the death of a person. (54)
Instead, the statutes' only required mental states pertain to the conduct and attendant-circumstance elements. (55) They usually require, first, that the defendant act purposely with respect to the conduct element--namely driving a motor vehicle--and, second, that the defendant act knowingly with respect to an attendant circumstance element--namely, the fact that the defendant had consumed an intoxicant. (56)
A second, related example of a general intent offense is the crime of drug-induced homicide. A substantial minority of states now have specific drug-induced homicide statutes that supplement the states' broader general proscriptions on reckless and negligent homicide. (57) These statutes generally require, first, that the defendant deliver one of several specified controlled substances--for example, heroin, methamphetamine, or cocaine--and, second, that another person die as the result of ingesting the controlled substance. (58) The statutes do not require the Government to prove that the defendant was reckless or negligent with respect to the social harm that is the target of the statute. (59) Instead, by way of mens rea, they typically require the Government only to prove that the defendant knew that he or she was delivering the controlled substance. (60)
To a lawyer schooled in the Model Penal Code's strict, elemental approach to culpability, the crimes of drunk-driving homicide and drug-induced homicide look very different from crimes like reckless homicide and criminally negligent homicide. They look different because they lack a culpable mental state with respect to one of their material elements. The Model Penal Code requires, with a few insignificant exceptions, that every material element of every criminal statute be assigned a culpable mental state. (61) Under the Code, a criminal statute that lacks a mental state with respect to some material element is said to impose "absolute liability," and absolute liability is absolutely forbidden. (62) The offenses of reckless and criminally negligent homicide satisfy the Model Penal Code's standard model; they require proof of some culpable mental state with respect to every element, including the result.
In contrast, the crimes of drunk-driving homicide and drug-induced homicide lack a culpable mental state with respect to one of their elements. Indeed, they lack a culpable mental state with respect to the element that is the principal determinant of punishment: the death of a person. Drunk-driving homicide carries a penalty that is dozens or even hundreds of times more severe than the penalty for ordinary drunk driving. (63) Likewise, drug-induced homicide carries a penalty that is substantially more severe than the penalty for ordinary drug dealing. (64) And yet, for both of these offenses, the element that triggers this enhanced liability--the death of a person--has no attached culpable mental state. From the perspective of the Model Penal Code's drafters, the fact that these statutes operate to impose liability for homicide "based on culpability required for the underlying [offense] without separate proof of any culpability with regard to the death" makes them fundamentally problematic. (65)
But there is another way of looking at these offenses. As Professor Mark Kelman has pointed out, general intent offenses like drunk-driving homicide and drug-induced homicide are related to reckless and criminally negligent homicide in much the same way that negligence per se is related to ordinary negligence. (66) What differentiates these two relatively novel forms of homicide from the more traditional offenses of reckless manslaughter and criminally negligent homicide is just the identity of the decision-maker. (67) Instead of requiring a jury to make an ad hoc, case-by-case determination of unreasonableness, these statutes "require an antecedent [legislative] judgment of per se unreasonableness." (68) The legislature, not the jury, assumes the responsibility for balancing the three factors in the justifiability calculus: the social utility of the actor's conduct, the gravity of the potential harm, and the probability that the harm will come about as a result of the conduct.
The first of these factors--the social utility of the actor's conduct--clearly lends itself to antecedent legislative determination in both drunk-driving homicide and drug-induced homicide. Both statutes define with specificity the conduct that will trigger liability--drunk driving in the one case, and the delivery of drugs in the other. There is nothing controversial in the legislature's antecedent determination that the delivery of heroin, cocaine, or methamphetamine to other persons lacks any redeeming social value. To be sure, drunk driving has greater social value than drug dealing; it serves the same basic need for mobility that ordinary driving does. But, again, there is nothing very controversial about the legislature's assumption of the responsibility for assigning a value to the utility of driving. The social value of driving, though not constant, varies only within a narrow range. And for those rare cases where driving serves a compelling social need--say, the delivery of a sick child to the hospital--the necessity or choice-of-evils defense is available to the driver. (69)
The second factor in the justifiability calculus the gravity of the potential harm--is even more clearly within the competence of the legislature. The gravity of the potential harm is a factor separate, and different in kind, from the probability of the harm occurring. The gravity of the harm presents a question of value, while the probability of the harm occurring is a question of fact. (70) Questions of value, of course, are uniquely well-suited to legislative determination. (71) And the value to be assigned to the death of a person is no exception. (72) There is nothing controversial, then, in the legislature's decision to assign a value to the gravity of the potential harm from either drug dealing or drunk driving.
It is the third factor in the justifiability calculus--the probability of the harm occurring--that appears to make these statutes problematic. Recall that the unjustifiability of conduct has to be judged on the basis of what the defendant knew when he or...
|
