Implementing Our Theory in “the Real World”
No doubt, our idealized code will seem bizarre to many people. Gone are references to rape, murder, and intention. In their place are risk impo- sitions and legally protected interests. One might object that although our “Golden Rule culpability formulation” works in theory, it can never work practically.
Th is section looks at how one might implement our idealized code in the current legal system. Before turning to the practicalities of imple- menting our code – a process that will not be without its diffi culties – we discuss the current state of the criminal law. We argue that the status quo is far from acceptable, and it should not be entitled to deference simply because it is the status quo. We also argue that our current crim- inal codes, although they appear to be rule based, ultimately rely on standards. Th us, it is no argument in favor of the status quo and against our code that ours is standards based.
Aft er surveying the current state of criminal law, we turn to the ques- tion of whether a rule- or standards-based system is preferable. Here, we discuss the value of having rules, and how values that law is meant to serve are better served by rules than by standards. However, we then discuss a signifi cant problem with rules – the existence of an ineliminable gap between the reasons for promulgating a rule and the reasons that a citizen has to obey it. For a retributivist, this problem is particularly worrisome because any actor who falls within this gap is an innocent (nonculpable) actor who does not deserve punishment – even if he has violated the rule.
Because this gap exists and cannot be eliminated, we argue that in almost all cases, the criminal law should opt for standards. However, we note that in some cases, the pressure for rules may be overwhelming.
We thus discuss the form in which these rules should be enacted and how violations should be punished.
We then turn to a range of other considerations. We argue that our standards-based system is consistent with the principle of legality. We address enforcement concerns, particularly as related to plea bargaining and to our simultaneously wide and narrow conception of a criminal act. And we conclude with a brief discussion of procedural, evidentiary, plea-bargaining, and sentencing considerations.
A. WHATWEARESEEKINGTOREPLACE
Our position radically recasts the criminal law landscape. It challenges the status quo. What burden of persuasion do we have when we seek to replace the current system?
Th e weight to be accorded to the current system depends on how well the current system works. Do we currently punish the guilty and acquit the innocent? Are our statutes narrowly tailored to prohibit only conduct that is culpable? Do our statutes speak with one voice regard- ing the justifi cation for criminal punishment? Do they give suffi cient notice regarding what conduct is prohibited?
Anyone even somewhat familiar with our criminal justice system will quickly realize that the answer to these questions is no. Our crimi- nal law system is defective, in several diff erent respects. We discuss three signifi cant problems here: overcriminalization, confl icting codes, and vagueness.
1. Three Signifi cant Problems with the Current State of Criminal Law a. Overcriminalization: One of the greatest problems with the current criminal law is overcriminalization. We currently punish conduct that does not risk harm to any interest the criminal law might wish to pro- tect. Th ere are two principal types of overcriminalization problems.
Th e fi rst is where the type of conduct prohibited bears only a quite attenuated connection to legally protected interests. As an example, the criminalization of possession of marijuana is arguably unjustifi able because an individual who possesses or uses marijuana is unlikely to harm other people merely by virtue of that possession or use. Indeed, even when the harm principle is invoked,49 there may be little empirical support for how the conduct risks the harm.50
49 See generally Bernard E. Harcourt, “Th e Collapse of the Harm Principle,” 90 J. Crim. L. &
Criminology 109 (1999) (discussing how the regulation of morality has evolved from legal moralism to the co-option of the harm principle).
50 Where a fundamental right is threatened, the Supreme Court does require a more strin- gent relationship between the act and the potential harm. Th us, in Ashcroft v. Free Speech Coalition, 535 U.S. 234 (2002), the Court invalidated a statute banning virtual child pornography on First Amendment grounds. Th e court noted that “[t]he Government has shown no more than a remote connection between speech that might encourage thoughts or impulses and any resulting child abuse. Without a signifi cantly stronger, more direct connection, the Government may not prohibit speech on the ground that
Th e second type of problem is overinclusiveness. Even if we can identify some group of people who will harm others, the criminal stat- utes routinely sweep within their prohibitions conduct that is harm- less.51 It is possible that some people who drive while talking on cell phones cannot do so safely but that others can. To bar everyone from driving while talking on cell phones is to restrict the liberty of even those people whose use of cell phones does not signifi cantly increase the risk of harm to other people.
Both legislatures and courts may be faulted for the overcriminal- ization problem. Th e political incentives are such that legislatures have every reason to criminalize and no reason not to criminalize. One might hope that courts would step in, but the Supreme Court has had little to say about substantive criminal law.52 Criminal statutes, unless burdening a fundamental right, are subject only to rational basis review.53 Th us, as one commentator has noted, a state could constitu- tionally criminalize eating sausage to prevent obesity.54 Th e ultimate result of such rampant criminalization is that police and prosecutors – not legislatures, judges, or citizens – have the ultimate power in deter- mining when to prosecute and what punishment individuals deserve.55 Th e problem is obvious – unchecked and unguided discretionary power is incompatible with the rule of law.56
b. Haphazard and Confl icting Codes: Beyond the enactment of crimes that punish harmless conduct, the manner in which the criminal laws are
it may encourage pedophiles to engage in illegal conduct.” Id. at 253–254. Th e Court’s solicitude for an actor’s First Amendment right to view virtual child pornography does not carry over to its solicitude, or lack thereof, for the right to be free from criminal punishment.
51 Th is is, of course, quite similar to the last problem. But with the prior problem, it may be that no instance of the criminalized conduct will actually result in harm, whereas, here, some instances of the criminal conduct will.
52 Marcus Dirk Dubber, “Toward a Constitutional Law of Crime and Punishment,” 55 Hastings L.J. 509, 509 (2004) (“It has become a commonplace that there are no meaningful constitutional constraints on substantive criminal law”).
53 Douglas N. Husak, “Guns and Drugs: Case Studies on the Principles Limits of the Criminal Sanction,” 23 Law & Phil. 437, 465–466 (2004).
54 Id. at 476.
55 William J. Stuntz, “Th e Pathological Politics of Criminal Law,” 100 Mich. L. Rev. 505 (2001); Husak, Overcriminalization, supra note 3, at 21.
56 See Husak, Overcriminalization, supra note 3, at 27.
enacted is also highly problematic.57 Because of special-interest-group lobbying, many criminal statutes that are duplicative of already exist- ing ones are enacted. For example, as Paul Robinson and Michael Cahill report, the Illinois criminal code, which already contained a prohibition on theft , also contains a special off ense for theft of delivery containers.58 Oft en, in response to some sort of public outcry over a dramatic crime, legislators view it as politically expedient immediately to criminalize the precise behavior involved. At these times, no thought is given as to how this ad hoc addition to the criminal code will aff ect the code as a whole.59 Th us, the new enactment may use diff erent terms or provide a sentence that is disproportionate to the otherwise similar crimes within the code.
c. Lack of Guidance: If a citizen wishes to know whether he may permis- sibly engage in conduct, the criminal law should provide him with guid- ance. It should tell him what he may and may not do. For the criminal law to give such guidance, the criminal law’s rules should be accessible to those regulated. But that is hardly the case with current criminal codes.
First, many criminal statutes contain vague terms, the meanings of which are determined by courts. However, the decisions of courts are hardly easily available to the average person who seeks to know for any given activity whether it is prohibited and punishable. To the extent that the criminal law requires a juris doctor (or more) to understand its full contours, it gives woefully little guidance to the citizenry.60 Adding insult to injury is the failure of the current criminal law to provide a mistake of law defense.61 Even when a citizen makes a good faith eff ort to learn the law, if she gets it wrong – even if her mistake is reasonable and in good faith – she will not be entitled to any defense.62
57 See generally Paul H. Robinson and Michael T. Cahill, “Can a Model Penal Code Second Save the States from Th emselves?” 1 Ohio St. J. Crim. L. 169 (2003); Husak, Overcriminalization, supra note 3, at 36–39.
58 Robinson and Cahill, supra note 57, at 170.
59 Id. at 170–171; see also Paul H. Robinson, Michael T. Cahill, and Usman Mohammad, “Th e Five Worst (and Five Best) American Criminal Codes,” 95 NW. L. Rev. 1, 2 (2000).
60 See John Calvin Jeff ries Jr., “Legality, Vagueness, and the Construction of Penal Statutes,”
71 Va. L. Rev. 189, 207–208 (1985).
61 Id. at 208–209.
62 But see Peter Westen, “Two Rules of Legality in Criminal Law,” 26 Law & Phil. 229 (2007) (arguing that, in most cases, current criminal law mistake-of-law rules accurately track the actor’s culpability).
Even when criminal law statutes are not so vague as to be unintelli- gible, they can be too specifi c and detailed to be comprehended. As Paul Robinson and John Darley have pointed out, the Model Penal Code’s formulation of self-defense is riddled with exceptions to exceptions.63 Not only is it unreasonable to expect a citizen to know the details of such a code, but also it is ridiculous to assume that any citizen in such a situation would have time to consult it. Th e citizenry needs simple rules that it can understand and obey. Notice has no value when understand- ing cannot be achieved.
In summary, current criminal codes suff er from an overcriminal- ization of conduct, haphazard and confl icting statutes, and both overly vague and overly detailed norms. If it manages to achieve retributive justice, that would be miraculous and accidental.
2. Do Our Current Criminal Codes Contain Rules?
Beyond the question of whether our current law serves rule-of-law val- ues is the question of whether it contains rules at all. In our view, the criminal law embeds standards within its statutes, thus resulting in a standards-based system, not a rule-based system.
We should note at the outset that we are not making a normative claim here as to whether rules or standards are preferable. We address that question shortly when discussing how to implement our theory.
For now, the question is an empirical one – what sort of criminal code do we have?
Although our criminal code may have many specifi c criminal stat- utes, ultimately the criminal law is standards based. Indeed, it is stan- dards based in every criminal statute that requires a mens rea of either recklessness or negligence.
Consider fi rst the number of standards embedded within the Model Penal Code. Attempts require the actor take a “substantial step.”64 Some attempts are entitled to mitigation (or even dismissal) if they are
“so inherently unlikely to result . . . in the commission of a crime.”65
63 Paul H. Robinson and John M. Darley, “Does Criminal Law Deter? A Behavioural Science Investigation,” 24 Oxford J. Legal Stud. 173, 181 (2004).
64 Model Penal Code § 5.01(1)(c) (1985).
65 Id. at § 5.05(2).
An actor is guilty of gross sexual imposition if he compels his victim by “any threat that would prevent resistance by a woman of ordinary resolution.”66 One may commit a crime by “loitering” or “prowling.”67 Mistake of law is a defense if the actor “acts in reasonable reliance upon an offi cial statement of law.”68 An actor has a duress defense if he suc- cumbs to a threat “which a person of reasonable fi rmness would have been unable to resist.”69 An actor is held to consent to “reasonable foreseeable hazards of joint participation in an athletic contest.”70 An actor is justifi ed if, among other things, “the harm or evil sought to be avoided is greater than that sought to be prevented by the law defi n- ing the off ense charged.”71 Law enforcement may not use deadly force if it creates a “substantial risk of injury to innocent persons.”72 Parents may not use force on their children if such force is “known to create a substantial risk” of death, serious bodily harm, or “extreme pain” or
“gross degradation.”73
State statutes likewise embed standards. One can be guilty of mur- der if one commits a crime recklessly under circumstances manifesting extreme indiff erence to human life74 or guilty of manslaughter for act- ing recklessly as to death.75 Recklessness is also a suffi cient mens rea for off enses that range from the mundane to the bizarre: aggravated unper- mitted use of indoor pyrotechnics,76 arson,77 assault,78 bigamy,79 child abuse,80 computer crimes,81 criminal mischief,82 criminal nuisance,83
66 Id. at § 213.1(2)(a).
67 Id. at § 250.6.
68 Id. at § 2.04(3)(b).
69 Id. at § 2.09(1).
70 Id. at § 2.11(2)(b).
71 Id. at § 3.02(1)(a).
72 Id. at § 3.07(2)(b)(iii).
73 Id. at § 3.08(1)(b).
74 Ala. Code § 13A-6–2 (West 1975).
75 Id. at § 13A-6–3; Conn. Gen. Stat. Ann. § 53a-56 (2007).
76 N.Y. Penal Law § 405.18 (2003).
77 Ala. Code § 13A-7–43 (West 1975) (arson in the third degree).
78 Conn. Gen. Stat. Ann. § 53a-59 (2007); 18 Pa. Cons. Stat. Ann. § 2701 (West 2003).
79 Mo. Rev. Stat. § 568.010 (2007).
80 N.Y. Education Law § 1125 (2001) (child abuse in an educational setting).
81 Conn. Gen. Stat. Ann. § 53a-251 (2007).
82 Id. at § 53a-117.
83 N.Y. Penal Law § 120.20 (2007).
deceptive business practices,84 defacing traffi c signs and signals,85 false advertising,86 hazing,87 interference with police service animals,88 mix- ing, coloring, staining, or other alterations of drugs or medicines,89 obstructing highways,90 public lewdness,91 riot,92 and simulating legal process.93 Moreover, many states have also followed the Model Penal Code’s lead and enacted a blanket misdemeanor for reckless endangerment.94
Th ese examples are just the tip of the iceberg. Juries determine when an act goes beyond “mere preparation” to “dangerous proximity” such that the actor has committed an attempt.95 A jurisdiction that extends complicity to crimes that “naturally and probably” follow from the encouraged act also requires a jury determination about that link- age.96 Th e insanity determination is not rule bound, even though it may appear to be. Empirical evidence shows that juries reach the same con- clusions about legal insanity irrespective of the legal tests.97 And even when the culpability term has a clear legal meaning to law professors, a jury may infuse the term with its own interpretation.98
In the next section, we turn to the question of whether rules or stan- dards are preferable. At this point, we hope to have established at least a prima facie case that (1) the current criminal system should not be
84 Haw. Rev. Stat. § 708–870 (2006).
85 N.J. Stat. Ann. § 2C:17–3.1 (2007).
86 Haw. Rev. Stat. § 708–871 (2006).
87 Md. Code Ann., Crim. Law § 3–607 (2007).
88 Tex. Penal Code Ann. § 38.151 (2007).
89 Mich. Comp. Laws Ann. § 750.18 (2007).
90 Ga. Code Ann. § 16–11–43 (2007).
91 Tex. Penal Code Ann. § 21.07 (2007).
92 Conn. Gen. Stat. Ann. § 53a-175 (2007).
93 Tex. Penal Code Ann. § 32.48 (2007).
94 See, e.g., Ala. Code § 13A-6–24 (West 1975); Conn. Gen. Stat. Ann. § 53a-63 (2007); Md.
Code Ann., Crim. Law § 3–204 (2007); N.Y. Penal Law § 120.20 (2007); 18 Pa. Cons. Stat.
Ann. § 2705 (West 2007).
95 See Michael T. Cahill, “Punishment Decisions at Conviction: Recognizing the Jury as Fault-Finder,” 2005 U. Chi. Legal F. 91, 101 (2005).
96 E.g., People v. Luparello, 231 Cal. Rptr. 832 (Cal. Ct. App. 1986). Of course, this view is antithetical to ours.
97 See Rita J. Simon and David E. Aaronson, Th e Insanity Defense: A Critical Assessment of the Law and Policy in the Post Hinckley Era 126–135 (1988); Henry J. Steadman et al., Before and Aft er Hinckley: Evaluating Insanity Defense Reform 8 (1993).
98 See Darryl K. Brown, “Plain Meaning, Practical Reason, and Culpability: Toward a Th eory of Jury Interpretation of Criminal Statutes,” 96 Mich. L. Rev. 1199 (1998).
entitled to great deference, and (2) the current criminal system, although appearing to be rule based, relies largely on standards.
B. IMPLEMENTINGAPRACTICALCODE
1. Rules versus Standards: In General
Our view as articulated thus far is a standards-based view. Indeed, to this point, we have argued for only one standard: take only those risks to legally protected interests that, as you perceive those risks, are justi- fi ed by your reasons for acting. One might wonder, however, whether the criminal law should be structured in this way. Should the criminal law be standards based, or should it contain discrete determinate rules?
Of course, the debate between rules and standards is not a new one.99 But the question for us is quite specifi c: Should the criminal law simply use one (or many) standards? Or should criminal law rules be criminal law rules?
In this section, we argue that, with a few exceptions, criminal law serves its function best by being standards based. Importantly, most of the values that underlie having rules are consequentialist. As we have mentioned, there is an ineliminable gap between when a legislator should create rules and when a citizen should follow them. A citizen can thus violate a justifi ed rule justifi ably. Such a citizen should not be punished because he is not culpable and therefore does not deserve punishment. A criminal law that truly cares about an actor’s culpability cannot punish a nonculpable actor simply to preserve the (consequentialist) value of rules.
2. The Argument for Rules over Standards
Th e value of rules is that they authoritatively settle moral disagree- ments.100 Th at is, even when individuals are ethically well-disposed
99 Th e problem is perhaps most famously embodied in the debate between Oliver Wendell Holmes and Benjamin Cardozo. Holmes sought to establish per se rules of conduct, believing that the “featureless generality” of negligence would ultimately give way to spe- cifi c rules. See Baltimore & O.R.R. v. Goodman, 275 U.S. 66, 70 (1927); Oliver Wendell Holmes, Th e Common Law 111 (1881). In response, Cardozo noted that such rules could not take into account all the circumstances so as to adjudicate negligence correctly in future cases. See also Pokora v. Wabash Ry. Co., 292 U.S. 98, 104 (1934).
100 For further discussion of this argument, see Larry Alexander and Emily Sherwin, Th e Rule of Rules: Morality, Rules, and the Dilemmas of Law ch. 1 (2001).
actors, they need the assistance of posited, determinate rules.
Authoritative settlement by determinate rules resolves problems of coordination, expertise, and effi ciency.
Rules solve coordination problems.101 In some cases, there are sev- eral incompatible ways to act and no reason to prefer one solution to another.102 Which side of the street to drive on is one example. In other instances, rules solve social coordination problems: in a world of imper- fect information, and in which the morally right thing to do turns at least in part on what others are likely to do, rules provide actors with a basis for such a prediction.103
In other instances, the rules refl ect the expertise of their promul- gators. We may believe that a particular authoritative decision maker has greater moral and factual expertise than the typical rule subject.
Although there may be reason to doubt this superior expertise in any given case, in general, legal rules can resolve questions about how to act that most individuals on their own may not be capable of resolving as well because they lack the rule promulgator’s information or exper- tise.104 Rules also avert errors. When, owing to complexity, actors must look at a multitude of factors in order to determine what to do, they may simply get the calculations wrong.105
Finally, rules reduce decision-making costs. It is simply more effi - cient for us to have a traffi c law that tells us how fast to go than for us to calculate a safe speed each moment that we are driving.106 A rule that dispenses with the necessity of complex calculations can also be said to promote predictability because everyone will arrive at the same result – what the rule prescribes – rather than diff erent results through diff erent calculations.107
All of these benefi ts of rules stem from both rule addressees and rule enforcers having the same understanding of the rule.108 In other words,
101 Id. at 56; Frederick Schauer, Playing by the Rules: A Philosophical Examination of Rule- Based Decision-Making in Law and in Life § 7.7 (1991).
102 Alexander and Sherwin, supra note 100, at 56.
103 Id. at 57–58.
104 Id. at 55.
105 Schauer, supra note 101, at 150.
106 Id. at § 7.3.
107 Id. at 137.
108 Id. at 138.