Emory Law Journal

Crime, Punishment, and the Psychology of Self-Control
Rebecca Hollander-Blumoff Associate Professor, Washington University School of Law. Thanks to Susan Appleton, Kimberly Bailey, Kenworthey Bilz, Matt Bodie, Sam Buell, Emily Hughes, Peter Joy, Dan Markel, Alexandra Natapoff, Mae Quinn, Laura Rosenbury, David Zaring, and participants in the Washington University Regional Junior Faculty Workshop and the Junior Criminal Law Scholars Workshop for helpful feedback and suggestions. Thanks also to Elizabeth Chen for terrific research assistance.

Abstract

Criminal law, by design, assigns culpability for intentional, volitional action. Criminal law theory and criminal law doctrine thus both place an important emphasis on an individual’s ability to control his or her behavior. Mostly, the law presumes that individuals can exercise self-control; in special cases, culpability and punishment are lessened because of impaired self-control. Yet despite this central role for self-control, previous scholarship has not examined the implications for criminal law of decades of research in social psychology. This Article presents and explores the implications of the most important findings of recent social psychology research on self-control: first, that individuals think about their actions differently depending on the temporal frame of the potential behavior in ways that often impede self-control efforts, and second, that self-control is a finite resource that can be depleted by other actions and events. The robust findings about when and how individuals are able to exercise control over their behavior help to make sense of criminal law doctrine’s refusal to consider abstract questions of free will in favor of a focus on whether individuals’ actions were under their control on a particular occasion. Additionally, the differences in how psychology and law conceptualize self-control failure, with psychology offering a far broader definition of these instances, demonstrate just how much the law’s theoretical conceptions of self-control failure are normative and socially constructed, rather than positive and observational. The psychology of self-control has important implications for our understanding of the behavior that leads to particular criminal conduct and our conceptions of appropriate punishment.

Introduction

Criminal law rests on the assumption that individuals—most of the time—have free will. 1 See Morissette v. United States, 342 U.S. 246, 250 (1952) (noting the “universal and persistent . . . belief in freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil”); Smith v. Armontrout, 865 F.2d 1502, 1506 (8th Cir. 1988) (en banc) (“The whole presupposition of the criminal law is that most people, most of the time, have free will within broad limits. They are capable of conforming their actions to the requirements of the law, and of appreciating the consequences of failing to do so.”); Nita A. Farahany & James E. Coleman, Jr., Genetics, Neuroscience, and Criminal Responsibility, in The Impact of Behavioral Sciences on Criminal Law 183, 20708 (Nita A. Farahany ed., 2009) (explaining that there is a fundamental difference between theoretical free will and criminal law free will: “the criminal law presumes that individuals actively and consciously choose to engage in criminal conduct”). Indeed, modern philosophers and social scientists have suggested that an assumption of free will is critical to the successful functioning of society, regardless of whether or not it is true. See, e.g., John Tierney, Do You Have Free Will? Yes, It’s the Only Choice, N.Y. Times, Mar. 22, 2011, at D1. They act in ways that they choose to act, exercising control over their own behavior. Criminal law is not generally a system of strict liability: it largely assigns culpability and punishment to actors who have chosen to engage in criminal acts, rather than to actors who have accidentally or involuntarily caused harm. 2 See Wayne R. LaFave, Criminal Law § 5.1, at 253 (5th ed. 2010) (noting that crime typically requires an act or omission as well as requisite state of mind). But see id. § 5.5, at 288–89 (detailing several types of strict liability criminal statutes). Despite this central role of free will and self-control in the conceptualization of criminal responsibility, criminal law scholars have not, to date, considered the implications of decades of research in social psychology on the mechanisms of self-control. 3Only one article has begun to consider the social psychological research on self-control, in the narrow context of the provocation defense in British law. See Richard Holton & Stephen Shute, Self-Control in the Modern Provocation Defence, 27 O.J.L.S. 49 (2007). This Article suggests that examining current social psychology research on self-control offers a novel way to amplify our thinking about crime and punishment, 4In a project similar in type but different in focus, Steven Sherman and Joseph Hoffman offered a careful exploration of the implications of psychological research on anger for voluntary manslaughter and the “heat of passion” defense. Steven J. Sherman & Joseph L. Hoffmann, The Psychology and Law of Voluntary Manslaughter: What Can Psychology Research Teach Us About the “Heat of Passion” Defense?, 20 J. Behav. Decision Making 499 (2007). helping to make sense of the way that the law has developed, casting doubt on the descriptive validity of legal perspectives on self-control and crime, and offering potential guidance as we think about appropriate levels of culpability and punishment.

Criminal law scholars conceptualize self-control and its role in the law in a variety of ways, depending on their disciplinary focus. Advances in neuroscience, for example, have reinvigorated the debate about the possibility of identifying the workings of a “criminal brain” through its unique neurological processes, and about whether free will and self-control are impossible (or, at the least, implausible) in the face of such (potential) biological determinism. 5For discussion of this debate, see Farahany & Coleman, supra note 1, at 206–07; and Owen D. Jones, Behavioral Genetics and Crime, in Context, in The Impact of Behavioral Sciences on Criminal Law, supra note 1, at 125, 138. Others have joined in a more philosophical age-old debate between free will and determinism, exploring the degree to which—if at all—individuals have free will and can exert it in various circumstances. 6 See, e.g., Stephen J. Morse, Culpability and Control, 142 U. Pa. L. Rev. 1587, 1587–88 (1994). In addition, sociological theories largely take a more deterministic approach, suggesting that “delinquency is an adaptation to conditions and social influences in lower-class environments.” Deborah W. Denno, Sociological and Human Developmental Explanations of Crime: Conflict or Consensus?, 23 Criminology 711, 712 (1985). In contrast to both of these approaches, still others use economic principles to characterize criminal behavior as the product of a utility calculation in which a potential criminal actor weighs the costs and benefits of his actions in deciding whether to proceed. 7 See Richard A. Posner, Economic Analysis of Law § 7.2, at 242 (5th ed. 1998); Steven Shavell, Foundations of Economic Analysis of Law 552–56 (2004); Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. Pol. Econ. 169 (1968). Such a characterization only makes explanatory sense if a prospective criminal actor can bring his actions into line with the dictates of the equation—that is, he has free will and can exercise self-control. These approaches to criminal law—neurological, philosophical, and economic—differ in their visions of human behavior and their implications for self-determinism. 8These are not the only lenses through which scholars have analyzed criminal law, and I do not mean to suggest either (1) that they represent an exhaustive list of perspectives on criminal law or (2) that all criminal law scholars writing from the standpoint of neuroscience, philosophy, or economics share a common outlook. I mean only to indicate that several prominent trends in criminal law scholarship share a set of assumptions about the centrality (if not the meaning) of self-control. But all three approaches rest on a shared belief that whether or not individuals have control over their own actions when they engage in criminal acts matters.

Yet even as criminal law and criminal law theorists rely profoundly on the concepts of free will and self-control, there is rigorous debate over what these terms really mean—and what they really mean for criminal law. Modern social psychology research on self-control has developed a working model of how an individual controls her behavior and has offered several intriguing findings. First, one of the most recent theories of self-regulation suggests that systematic differences in the way individuals think about near-term and long-term events and actions may help to explain self-control failures and successes. People tend to construe near-future and distant-future events differently, and this has a direct effect on their efforts to bring their behavior into line with their rational choices. Second, research has supported the proposition that an individual’s self-control is a finite resource that can be used up by other cognitive demands and, furthermore, that an individual can get better at self-control over time. This work has compared the seeming paradox of self-control to a muscle—that is, self-control grows weaker with use in the short term but stronger with use in the long term. What this means for the individual is that, when she has used her self-control in one setting, she will find it harder to exert self-control in a situation that immediately follows. But when she practices controlling her behavior over time, her self-control will eventually grow stronger and will be less subject to quick depletion.

Two important broad insights come from examining this psychological research. First, by considering self-control failure at the micro level—in a particular moment of action or inaction—psychological research on self-control helps uncouple self-control questions from broader questions about the existence of free will. Psychological research supports the idea that individuals differ in the degree to which they can control their own actions. However, psychology does not take a clear position on the roots of self-control failure, leaving open the question of whether societal factors or some innate biological or other characteristics are most influential in dictating the constraints on an individual’s ability to control herself. Whether or not free will “truly” exists is irrelevant, psychological research suggests, to the question of whether or not an individual is able to control his or her behavior in a particular moment. A person either controls herself by acting in a way that is consistent with a particular set of norms 9For further discussion of what psychologists mean by acting in a controlled or uncontrolled manner, see infra Part II. or does not control herself by acting in opposition to it. The roots of failure to control one’s behavior, important though they may be, are separate from the question of an individual’s ability to do so at a specific time and place. Psychology’s robust findings on the fine-grained aspects of self-control suggest that self-control is a concept with meaning and usefulness for the law, regardless of one’s viewpoint about the existence of free will.

Second, taking psychological research on self-control seriously indicates that criminal law may vastly underdescribe the scope of situations in which an individual lacks the ability to control her actions. That is, acts that the law calls “uncontrolled” are a mere subset of the behavior that psychology would call “uncontrolled.” Looking carefully at the psychology research, however, I suggest that psychologists may sometimes be overbroad in their definition of self-control failure: in particular, psychologists often assume a set of shared norms that may not exist. The mismatch between the scope of self-control as described by psychology and criminal law helps to highlight that notions of self-control in the law are inherently constructed by the law itself, rather than reflecting some empirical reality, and that any efforts to define and understand the concept and role of self-control in law as purely positive, rather than normative, are misguided.

In Part I, I briefly demonstrate the ways in which the concept of self-control permeates criminal law, beginning with criminal law theory and then considering criminal law doctrine. Part II provides a concise overview of psychological research on self-control. Part III examines in greater depth psychological research on how information processing about events that occur at different times may intersect with self-control and considers the implications of this research on criminal law. Part IV describes the research on self-control as a finite resource and explores the implications of this “strength” model on criminal law. Part V offers some preliminary thoughts about potential implications in particular areas of criminal law doctrine and in criminal law theory.

I. Self-Control and Criminal Responsibility

Efforts to neatly classify and bring order to criminal law theory and doctrine are stymied by the scope and complexity of the field. Criminal law scholars consider crime from diverse angles, including moral, theoretical, philosophical, economic, sociological, neurological, and biological perspectives. They consider not only the scope of what ought to constitute criminal behavior but also what causes criminal behavior and what appropriate punishment ought to be. And criminal law doctrine encompasses crimes as diverse as assault and embezzlement, homicide and tax fraud. Yet in a field as frustratingly diverse as criminal law, one constant theme is the critical role of human self-control. In the following sections, I show how self-control cuts across diverse theoretical and doctrinal perspectives, forming an insistent motif at the core of criminal law.

I analyze self-control first in the context of criminal law theory and then in criminal law doctrine, and in each setting I further subdivide my analysis into broad conceptual groupings. For criminal law theory, I consider three popular areas of scholarly focus: neuroscience, retributivism/moral theory, and utilitarianism/law and economics. For criminal law doctrine, I consider involuntary acts, duress, and mens rea; insanity; justification and excuse; and mitigation of punishment. However, there may be significant overlap among these categories, and some scholarly perspectives cut across, unite, or simply defy them.

Although these divisions are somewhat blunt, my point here is not to articulate a clear and comprehensive topography of criminal law scholarship but rather to demonstrate the cross-cutting nature of the concept of self-control across the field as a whole. My brief examination of these areas leaves many nuanced issues, debated by those who write in these domains, to the side—not because they are unimportant but rather because they are beyond the scope of this project. My purpose here, again, is simple: to demonstrate the cross-cutting and central nature in criminal law of the very concept of an individual’s ability to control her behavior.

A. Criminal Law Theory

In this section, I explore three prominent strands of thought in current criminal law theory scholarship: the neuroscience perspective, the retributive/moral theory perspective, and the utilitarian/law and economics perspective. 10Again, although the divisions I have drawn are somewhat rough, and criminal scholars’ work may often overlap or even defy these categories, the distinctions are nonetheless useful as a basic framework for discussion. In each of these, scholars grapple with the question of when and how individuals control their behavior and what such control, or lack thereof, means for criminal law. I explain the central role that notions of self-control play in each perspective in turn below.

1. Neuroscience Perspective

Self-control is perhaps the most acute topic of debate in criminal law and neuroscience scholarship, a field that encompasses brain science, brain imaging, and genetics. Growing developments in neuroscience have prompted challenges to classic notions of responsibility. Neurology suggests both a certain level of determinism, because it links biological and chemical reactions in the brain to manifested behavior, and optimism about potential control by the self and/or society, perhaps through chemical or biological intervention.

For example, brain imaging has helped to reduce criminal punishment in certain cases by demonstrating that an individual’s brain tumor or other deformity or injury was one of the causes of criminal behavior. When neuroscience can show that an individual did not make a “free choice” to act in certain way, but instead acted that way because of a biological and identifiable abnormality, juries and courts have sometimes been willing to assign lesser responsibility and punishment. 11 See, e.g., Crook v. State, 813 So. 2d 68, 75 (Fla. 2002) (per curiam) (“[T]he existence of brain damage is a significant mitigating factor that trial courts should consider in deciding whether a death sentence is appropriate in a particular case.”); Tyler v. State, 903 N.E.2d 463, 469 (Ind. 2009) (rejecting the trial court’s imposition of an enhanced sentence in light of the presence of mitigating factors, including “a brain tumor . . . that affected [the defendant’s] ability to control his behavior”). So, too, scholars have suggested that potential solutions to the problem of criminal behavior may be pharmacologically or surgically possible in some of these cases. 12 See, e.g., Henry T. Greely, Neuroscience and Criminal Justice: Not Responsibility but Treatment, 56 U. Kan. L. Rev. 1103, 1104 (2008) (“[N]euroscience is likely to produce a number of different types of interventions—surgical, pharmacological, and other—that will be said to prevent, or to lessen the risk of, continuing criminal conduct.”). But see John F. Stinneford, Incapacitation Through Maiming: Chemical Castration, the Eighth Amendment, and the Denial of Human Dignity, 3 U. St. Thomas L.J. 559, 568 (2006) (arguing against pharmacological punishment for sex offenders).

Others seek to use neuroscience less at the abnormal margins and more in the mainstream of criminal activity, suggesting that when we understand the process by which individuals make decisions about criminal conduct, we might better understand and prevent it. 13Jones, supra note 5, at 142. Some scholars also argue that new technology’s ability to map brain changes and processes provides a foothold for arguments that criminal behavior is not really the responsibility of the actor but is biologically determined. 14 See, e.g., Michele Cotton, A Foolish Consistency: Keeping Determinism Out of the Criminal Law, 15 B.U. Pub. Int. L.J. 1, 44 (2005); Richard E. Redding, The Brain-Disordered Defendant: Neuroscience and Legal Insanity in the Twenty-First Century, 56 Am. U. L. Rev. 51, 118 (2006); Christopher Slobogin, The Civilization of the Criminal Law, 58 Vand. L. Rev. 121, 165 (2005). In particular, some practitioners have made the argument that research on genetic predispositions to criminal behavior, and on the heritability of criminal propensity, demonstrates that certain individuals’ actions are not truly voluntary. 15Deborah Denno notes that, although “[b]ehavioral genetics evidence has not gained widespread acceptance in current case law,” “[a]t least 48 criminal cases have referred to behavioral genetics evidence” in a thirteen-year period between 1994 and 2007. Deborah W. Denno, Behavioral Genetics Evidence in Criminal Cases: 1997–2007, in The Impact of Behavioral Sciences on Criminal Law, supra note 1, at 317, 330–31. Efforts to convince courts to adopt this perspective have been almost completely unsuccessful. 16 See Farahany & Coleman, supra note 1, at 187–205; see also Mobley v. State, 455 S.E.2d 61, 65–66 (Ga. 1995) (rejecting consideration of the defendant’s genetic propensity for criminal behavior). Nonetheless, these challenges based on brain imaging and genetics have generated a robust response from criminal law scholars, even those who are enthusiastic about the possible implications of neuroscience on law. Nita Farahany and James Coleman, for example, forcefully argue that “as a matter of criminal law theory, [behavioral genetics and neuroscience] evidence should not inform the assessment of criminal responsibility in . . . any meaningful way.” 17Farahany & Coleman, supra note 1, at 185. The intersection of neuroscience and criminal law is, indeed, a textbook example of the tensions at issue in defining the scope of free will, self-control, and voluntariness.

2. Retributive/Moral Theory Perspective

Moral theorists suggest that the law punishes a set of behaviors that can be categorized as morally wrong; when people choose to engage in morally wrong acts, they ought to be punished. The retributive justification for criminal law is simple: wrongdoers deserve punishment commensurate with the wrong they intend. 18Kent Greenawalt, Punishment, in 4 Encyclopedia of Crime and Justice 1336, 1338 (1983). Recent research has added social science data into the discussion of how we determine commensurate punishment—that is, how we understand desert. See, e.g., Paul H. Robinson & John M. Darley, Intuitions of Justice: Implications for Criminal Law and Justice Policy, 81 S. Cal. L. Rev. 1 (2007) (discussing and analyzing research relating to desert). As John Rawls explains, “It is morally fitting that a person who does wrong should suffer in proportion to his wrongdoing.” 19John Rawls, Two Concepts of Rules, 64 Phil. Rev. 3, 4–5 (1955). A core belief of a retributivist is that an offender ought to be punished “in proportion to her desert or culpable wrongdoing.” 20Russell L. Christopher, Deterring Retributivism: The Injustice of “Just” Punishment, 96 Nw. U. L. Rev. 843, 860 (2002) (footnotes omitted). Linking the degree of punishment directly to the gravity of the wrong helps to highlight moral theorists’ suggestion that punishment is a restoration of a “moral order” that has been violated. 21 See Greenawalt, supra note 18, at 1338. To preserve such moral order, Kant insists that society has a duty to punish those who have done wrong. 22Kant famously claimed that a society about to disband should still execute its murderers on principle. Immanuel Kant, The Philosophy of Law: An Exposition of the Fundamental Principles of Jurisprudence as the Science of Right 198 (Augustus M. Kelley Publishers 1974) (1887).

The retributive model “requires some notion of free will that attributes to humans responsibility for doing wrong in a way that is not attributed to other animals.” 23 See Greenawalt, supra note 18, at 1339. Notably, Hegel suggests that punishment for criminal acts honors the criminal as a rational being. 24David E. Cooper, Hegel’s Theory of Punishment, in Hegel’s Political Philosophy 151, 153 (Z.A. Pelczynski 1971). Retributive justice is premised on the idea that someone ought to be punished for his wrongful act when that person knew that what he was doing was wrong and did it anyway. A moral punishment for wrongdoing makes no sense when the actor did not understand that the act was wrongful or did not mean to engage in a wrongful act.

The law’s vision of moral responsibility “generally requires a demonstration that the actor made a knowing, rational, and voluntary choice to act in a manner that breached community norms.” 25Peter Arenella, Convicting the Morally Blameless: Reassessing the Relationship Between Legal and Moral Accountability, 39 UCLA L. Rev. 1511, 1518 (1992). As Stephen Morse explains in an article on criminal law culpability and control, “If it is true that an agent really could not help or control herself and was not responsible for the loss of control, blame and punishment are not justified on any theory of morality and criminal punishment.” 26Morse, supra note 6, at 1587–88. When individuals act without volition, they are not liable for their actions. As Morse writes:

Conviction and punishment are justified only if the defendant deserves them. The basic precondition for desert in all contexts, legal and otherwise, is the actor’s responsibility as a moral agent. Any condition or circumstance that sufficiently compromises responsibility must therefore negate desert; a just criminal law will incorporate such conditions and circumstances in its doctrines of excuse. 27Stephen J. Morse, Excusing the Crazy: The Insanity Defense Reconsidered, 58 S. Cal. L. Rev. 777, 781 (1985).

He further notes that “minimal rationality (a cognitive capacity) and minimal self-control or lack of compulsion (a volitional capacity) are the essential preconditions for responsibility.” 28 Id. at 782. However, Morse has argued that “[o]ut-of-control” action does not mean, necessarily, “unintentional” action 29Morse, supra note 6, at 1595. and has subjected defenses of “out of control” behavior to intense scrutiny. Indeed, he has suggested that “control” issues are largely understood as defects in rational thinking. 30 Id.

So, too, Peter Arenella has noted, “[T]he law’s account of moral agency is simply a description of those persons who have sufficient control over their behavior that they can make a rational choice about whether to comply with the minimal moral demands of the criminal law.” 31Arenella, supra note 25, at 1584–85. A critical assumption behind retribution is thus that an individual could have chosen not to engage in the act—could have controlled his own behavior—but instead chose to act despite an understanding that the behavior was wrong. Retributivists find no use for punishing someone who did not have the capacity to control himself or to choose his actions. However, when someone has the capacity to control himself, but simply does not do so, he is culpable.

3. Utilitarian/Law and Economics Perspective

The law and economics vision of crime suggests that individuals choose to engage in certain behavior because the benefits outweigh the costs and that criminal law provides a set of deterrents against engaging in specific behavior. Thus deterrence provides society with a way to prevent crime by increasing the costs of criminal behavior. 32 Posner, supra note 7, § 7.2, at 243. Under this account, criminal law’s job is to create sanctions such that “the criminal is made worse off by committing the act.” 33 Id. An important assumption of deterrence theory is that individuals will weigh the costs of their behavior (punishment) against the benefits. 34 See Cesare Beccaria, On Crimes and Punishments and Other Writings (Richard Bellamy ed., Richard Davies trans., Cambridge Univ. Press 1995) (1764); Jeremy Bentham, The Principles of Morals and Legislation (Prometheus Books 1988) (1789); H.L.A. Hart, Punishment and Responsibility (2d ed. 2008). This utilitarian theory is embraced by law and economics scholars who presume a rational actor who seeks to maximize utility. This rational actor will engage in an activity only when its benefits outweigh its costs. Thus the key challenge for criminal law is calibrating the punishment appropriately, as weighed against the benefits, to deter crime. The goal is to “set the expected punishment cost at a level that will not deter the occasional crime that is value maximizing.” 35 Posner, supra note 7, § 7.2, at 243. And volumes have been written on creating an optimal level of sanctions for deterrence purposes. 36 See, e.g., Richard A. Bierschbach & Alex Stein, Mediating Rules in Criminal Law, 93 Va. L. Rev. 1197 (2007); Samuel W. Buell, The Upside of Overbreadth, 83 N.Y.U. L. Rev. 1491 (2008); Steven Shavell, Criminal Law and the Optimal Use of Nonmonetary Sanctions as a Deterrent, 85 Colum. L. Rev. 1232 (1985); Avraham D. Tabbach, Does a Rise in Maximal Fines Increase or Decrease the Optimal Level of Deterrence?, 5 Rev. L. & Econ. 53 (2009).

But deterrence only makes sense when, as a general matter, individuals can assess the consequences of their outcomes and choose their course of action accordingly: that is, when they are able to control their own behavior. 37Although I have described the retributivist and utilitarian attitudes toward human control over one’s own actions as similar here, Joshua Dressler contrasts them, stating that “the premise of utilitarianism is that people are generally hedonistic and rational calculators,” whereas “retributivists focus on their view that humans generally possess free will or free choice and, therefore, may justly be blamed when they choose to violate society’s mores.” Joshua Dressler, Understanding Criminal Law 16 (5th ed. 2009) (footnote omitted). Although Dressler’s description may be accurate, both perspectives do rely on an understanding of an individual who is able to choose. The utilitarian view presumes that most people will engage in this rational calculation and decide not to engage in criminal behavior. Those who do engage in the prohibited behavior thus represent a problem for the utilitarian that must be explained in one of several ways: the crime was appropriately committed because its benefits did outweigh its harms; 38 See Posner, supra note 7, § 7.2, at 242–43 (arguing that a starving man ought to steal a trivial amount of food from an unoccupied cabin in the woods). society failed to provide the appropriate level of deterrence; 39 See id. § 7.3, at 250–51 (arguing that the theory of criminal sanction is one of deterrence). the individual miscalculated the cost–benefit equation; the individual had incomplete, missing, or wrong information about the costs and benefits; or the individual correctly calculated the equation but was unable to bring his behavior in line with the appropriate choice.

This last instance, when an individual cannot bring his behavior into line with his rationally calculated choice, is difficult to reconcile with law and economics’ baseline assumptions. For example, in considering the case of an impulsive, heat of passion crime, Posner suggests not that the individual knew that the costs outweighed the benefits and yet could not control himself, but rather that the impulsive individual merely requires a higher level of sanctions in order to balance the equation properly. 40 Id. § 7.4, at 257. On the other hand, Posner suggests that, because the impulsive criminal is less deterrable, perhaps society ought to “buy” less punishment for these individuals. Id. Heat of passion, that is, does not eliminate the possibility of self-control. It merely makes exercising control a less utility-maximizing choice, which thus requires a concomitant increase in punishment to reach deterrence levels. On the other hand, utilitarian theory acknowledges that the imposition of punishment for those who truly are not deterrable serves no function for deterrence purposes: hence a deterrence proponent has no use for punishing an individual who either cannot make an accurate cost–benefit analysis or cannot abide by this calculation’s mandate. 41 See, e.g., Shavell, supra note 7, at 561–62. Of course, generally speaking, a deterrence system has no particular interest in punishing any one criminal. Instead, “people are actually fined or imprisoned only to maintain the credibility of the deterrent.” Posner, supra note 7, § 7.3, at 250. The utilitarian view suggests that there are few criminals who are not capable of controlling their actions to comport with their cost–benefit analyses but also suggests that those who are truly not able to exercise self-control are not appropriately punished under deterrence theory. 42However, the system may need to impose punishment to remain credible for future offenders, and there still may be an independent need to imprison undeterrable criminals for the purpose of incapacitation. Shavell, supra note 7, at 562. As Shavell says:[I]n many cases where a person is not legally responsible for his act, he will be no less dangerous to society than if he were responsible. A person who has an uncontrollable urge to set fires, or who is subject to insane, violent rages, is dangerous to society even though he cannot help himself.Id. In sum, the utilitarian model presumes self-control by the rational criminal actor.

B. Criminal Law Doctrine

Criminal law doctrine generally embraces the idea that individuals are culpable only when they act with intent and volition. A variety of criminal doctrines rely on the concept of “control” when determining culpability and punishment. While scholars debate the premise of true free will and explore the implications for criminal law of its existence or lack thereof, criminal law doctrine largely pushes these metaphysical questions off to the side, instead focusing on the individual’s relationship to the behavior in the moment. The basic underpinnings of criminal law, actus reus and mens rea, require intentional action. Prominently, the concept of self-control is also implicated in inquiries into whether the defendant is insane or, instead, must stand trial. Self-control also plays a significant role in inquiries into whether conduct is justifiable or excusable. Finally, questions about self-control may arise in the context of mitigation or enhancement of punishment. I briefly demonstrate the role of self-control in these areas below.

1. Self-Control and Involuntary Acts, Duress, and Mens Rea

The language of involuntary acts, duress, and mens rea reflects a preoccupation with self-control as a factor in culpability. As numerous scholars have noted, an action that is truly outside of one’s ability to control cannot yield criminal responsibility. The actus reus requirement includes an insistence that an action be voluntary—a deliberate movement by one’s body is distinct, legally, from an involuntary physical movement. As Dressler explains, “With a voluntary act, a human being—a person—and not simply an organ of a human being, causes the bodily action.” 43 Dressler, supra note 37, § 9.02, at 90. Actions that are not the product of the brain’s conscious control lie outside the scope of the actus reus.

However, in cases where there is duress applied by an outside force—for instance, a situation where someone has a gun to one’s head and insists that one perform a criminal act—it is clear that the party is acting rationally and is able to bring her actions in line with her rational preferences to continue, herself, to live. So, the criminal act is not the immediate product of a self-control problem, per se. But because the circumstances that led to the criminal actor’s predicament are outside of her control, the law offers lesser culpability for these acts (except for those involving murder). 44 Id. § 23.01, at 303–04.

Intent and mens rea similarly work to ensure that the law limits criminal responsibility to those who mean to do wrong. The law requires a “culpable state of mind.” 45 Id. § 10.07, at 139, § 11.01, at 145–46. As Hart said, if we did not have responsibility as a principle,

we should lose the ability which the present system in some degree guarantees to us, to predict and plan the future course of our lives within the coercive framework of the law. For the system which makes liability to the law’s sanctions dependent upon a voluntary act not only maximizes the power of the individual to determine by his choice his future fate; it also maximizes his power to identify in advance the space which will be left open to him free from the law’s interference. 46 H.L.A. Hart, Punishment and the Elimination of Responsibility, in Punishment and Responsibility, supra note 34, at 181–82.

The law does not take action unless there is the appropriate mens rea. This is a risk, and “[i]t is the price we pay for general recognition that a man’s fate should depend upon his choice and this is to foster the prime social virtue of self-restraint.” 47 Id. at 182. When someone lacks the capacity to make this choice—whether from an inability to understand the facts of the conduct, comprehend that the conduct violates the law, or conform behavior to the requirements of the law—he does not deserve punishment. 48 See Dressler, supra note 37, § 25.03, at 346 (noting that just punishment depends on the rationality and self-control of the criminal who performed the crime).

2. Self-Control in Insanity

The insanity defense is predicated on lack of self-control. If we accept, as Roscoe Pound says, “the starting point of the criminal law . . . that a criminal was a person possessed of free will who, having before him a choice between right and wrong, had freely and deliberately chosen to go wrong,” then we are inexorably led to his conclusion that “it was evident that if mental disease inhibited or destroyed the will element or precluded the choice of right and wrong, there was no crime.” 49 Roscoe Pound, Criminal Justice in America 126–27 (1930). In essence, a criminal is a person who deliberately chooses to engage in behavior that he knows is wrong; we call this crime because the person could have chosen not to engage in the behavior. That is, that person could have exercised self-control over his actions and opted for a different choice. When that choice-making process is hampered, by whatever means, the system seeks ways to recategorize the resulting conduct.

A variety of perspectives on criminal law support the existence of the insanity defense. For example, punishing those who are insane provides no deterrent effects. As Dressler says, “A person who does not know what she is doing or who cannot control her conduct cannot be deterred by the threat of criminal sanction.” 50 Dressler, supra note 37, § 25.03, at 345. And, similarly, moral retributivists may support the insanity defense because one cannot ascribe blame to severely mentally ill people for their behavior when they simply “lack the capacity to do what other humans are able to do: to act rationally or to control their behavior.” 51 Id. at 346. The exception of the insane person highlights our most basic presumption of free will: unless one is mentally diseased, we assume the capacity for free choice. 52 Id. As Dressler explains, “[M]oral responsibility for one’s actions is dependent on the essential attributes of personhood, namely rationality and self-control.” 53 Id. (footnote omitted). Insane people lack these “essential attributes of personhood” and are therefore not the appropriate subject of a moral evaluation. 54 Id.; accord Morse, supra note 27, at 783.

There are several basic tests of insanity, all of which rely to some degree on a conception of the sane individual’s ability to control his actions. The M‘Naghten test states that a person is insane if, at the time of the act in question, he did not know the nature and quality of the act or did not know that the act was wrong. 55M‘Naghten’s Case, (1843) 8 Eng. Rep. 718 (H.L.) 722; 10 Clark & Finnelly 200. This means that individuals who know what they are doing, and that their act is wrong, are not insane. This test tracks a retributivist account of self-control: we only punish those who could have chosen not to do wrong. When an individual is unaware that particular acts are wrong, he does not have the opportunity to try to bring his actions in line with the law, and so he is not legally responsible for those actions. Criticism of the M‘Naghten test has taken a self-control tack: the test does not include mental illnesses that affect volition, but only focuses on cognitive disability. 56 See Dressler, supra note 37, § 25.04, at 352. The rule, the critics argue, does not adequately protect the individual who is “unable to control her behavior.” 57 Id.

In response to M‘Naghten, some jurisdictions added a third prong to correct this omission: the “irresistible impulse” test. 58 See id. at 353 (citing Clark v. Arizona, 548 U.S. 735, 751 (2006)). This test tries to reach conduct that is apparently beyond the power of the actor to suppress, excusing the actor from culpability if the actor “acted from an irresistible and uncontrollable impulse,” 59 Id. (quoting Commonwealth v. Rogers, 48 Mass. (7 Met.) 500, 502 (1844)) (internal quotation marks omitted). the actor “lost the power to choose between the right and wrong,” 60 Id. (emphasis omitted) (quoting Parsons v. State, 2 So. 854, 866 (Ala. 1887)) (internal quotation mark omitted). or the actions were beyond the control of the actor. 61 Id. (citing Davis v. United States, 165 U.S. 373, 378 (1897)). This test can even cover planned behavior as long as the defendant “lacked the ability to control her conduct.” 62 Id. Self-control has been deeply criticized as the appropriate test for insanity 63 See Morse, supra note 27. because, as Dressler has noted, psychiatrists do not have an accurate measurement device for the capacity for self-control, 64 Dressler, supra note 37, § 25.04, at 354 (citing United States v. Lyons, 731 F.2d 243, 248 (5th Cir. 1984) (en banc)). but it nonetheless remains a common thread in judicial discussion of insanity.

Another test for insanity comes from the American Law Institute’s Model Penal Code. There, a person is not responsible for his criminal conduct if, as a result of a mental disease or defect, he lacked substantial capacity “to appreciate the criminality [wrongfulness] of his conduct or to conform his conduct to the requirements of law.” 65 Model Penal Code § 4.01(1) (1985) (alteration in original). From the self-control perspective, this test suggests that if an individual has a mental defect such that she lacks the ability to control her own conduct, she will not be held responsible for her actions by reason of insanity. The federal insanity rule hews to the M‘Naghten test 66 See Dressler, supra note 37, at 356. but adds that it must be that the individual cannot appreciate the nature or quality of the act or its wrongfulness by virtue of a severe mental disease or defect. 67The product test, also called the Durham test, stated that whenever a person’s behavior is the product of a mental disease or defect, the person should not be held liable by reason of insanity. Durham v. United States, 214 F.2d 862 (D.C. Cir. 1954), overruled by United States v. Brawner, 471 F.2d 969 (D.C. Cir. 1972) (en banc), superseded by statute, Insanity Defense Reform Act of 1984, Pub. L. No. 98-473, 98 Stat. 2057 (codified as amended at 18 U.S.C. §§ 17, 4241–4247 (2006)), as recognized in Shannon v. United States, 512 U.S. 573 (1994). This test has been abandoned. See Brawner, 471 F.2d at 973 (adopting the Model Penal Code test). Its purpose was to streamline the test for insanity in order to give mental health professionals freer rein to testify about a defendant’s insanity without parsing the defendant’s volitional or cognitive abilities. See Dressler, supra 37, at 355. Insanity defenses are not, of course, simply a binary test of whether an individual lacked self-control. 68For example, Pennsylvania courts expressly include lack of self-control in their definition of what insanity is not: “Certainly neither social maladjustment, nor lack of self-control, nor impulsiveness, nor psychoneurosis, nor emotional instability, nor chronic malaria, nor all of such conditions combined, constitute insanity within the criminal-law conception of that term.” Commonwealth v. Neill, 67 A.2d 276, 280 (Pa. 1949). Many individuals who lack control over their behavior in some way are not criminally insane; indeed, very few defendants are excused from liability through the insanity defense. However, defendants found insane do, by and large, share the characteristic that they are not able to exercise self-control.

3. Self-Control in Justification and Excuse

There are many defendants who fail to meet the definition for insanity but are nonetheless mentally compromised in some way that the criminal law believes should make a difference with respect to culpability. In particular, as I describe below, several doctrines suggest lesser culpability for those who are not fully in control of their actions. I do not make an effort here to classify justifications, excuses, defenses, or mistakes, or to offer a taxonomy of these items. There are others who have undertaken that project, 69 See, e.g., Paul H. Robinson, Criminal Law Defenses: A Systematic Analysis, 82 Colum. L. Rev. 199, 203 (1982). and the distinction between justification and excuse is one that has received much scholarly attention. 70 See, e.g., Mitchell N. Berman, Justification and Excuse, Law and Morality, 53 Duke L.J. 1 (2003); Kent Greenawalt, Distinguishing Justifications from Excuses, 49 Law & Contemp. Probs. 89 (1986); Kent Greenawalt, The Perplexing Borders of Justification and Excuse, 84 Colum. L. Rev. 1897 (1984). The only aspect of these concepts critical to my analysis here is how these doctrines understand and rely on self-control as a guiding principle. My purpose is simply to address a variety of the ways in which criminal law incorporates its concern about an individual’s ability to control her actions. I consider below the role of self-control in the context of provocation and intoxication.

a. Provocation

The very definition of provocation relies on a notion of the possibility of losing control of one’s actions. One typical definition suggests that provocation is “adequate” if it is “sufficient to cause an ordinary man to lose control of his actions and his reason,” 71State v. Guebara, 696 P.2d 381, 385 (Kan. 1985) (citing State v. Coop, 573 P.2d 1017, 1020 (Kan. 1978)). indicating the law’s reliance on a reasonable/ordinary person standard in determining when provocation is adequate. Dressler explains that there are two ways this plays out: first, one measures the gravity of the provocation to the reasonable/ordinary person, and second, one measures the level of self-control to be expected of a reasonable/ordinary person. 72 Dressler, supra note 37, § 31.07, at 539.

The law and theory surrounding provocation, or heat of passion, 73There are four basic elements to a provocation/heat of passion claim. First, the actor must act in the heat of passion; second, the passion must have resulted from adequate provocation; third, “the actor must not have had a reasonable opportunity to cool off”; and finally, “there must be a causal link between the provocation, the passion, and the homicide.” Id. at 535 (citing Girouard v. State, 583 A.2d 718, 721 (Md. 1991)). is perhaps the most obvious place in criminal law where the rational actor model clashes directly with moral theorists’ vision; this clash occurs precisely because of the question of control inherent in the doctrine. The rational actor model suggests that the greater the provocation, the less deterrable the criminal actor will be. That is, the immediate benefit to the enraged, provoked actor will outweigh the regular sanction, whereas the immediate utility to a nonenraged, nonprovoked actor will not. 74 See Posner, supra note 7, § 7.2, at 242. The utilitarian model does not readily yield to the potential for an actor whose actions seem “out of control” because of provocation; instead, the suggestion is just that the utility calculation for these individuals looks different. For a moral theorist, however, the “out of control” question is one that directly implicates the degree of blameworthiness, and it is partially for that reason that so much has been written on the question of whether provocation is a justification (something that means that the act in response is not wrong) or an excuse (in which case the act is wrong, but not quite as blameworthy as it might otherwise be). 75 See Joshua Dressler, Provocation: Explaining and Justifying the Defense in Partial Excuse, Loss of Self-Control Terms, in Criminal Law Conversations 319 (Paul H. Robinson et al. eds., 2009) (arguing that justification serves as a partial defense). The nature of the inquiry into whether provocation is an excuse or justification stems from the kind of control the individual had over his behavior. So, for example, Reid Fontaine explains that from an excuse perspective, the person who is provoked and experiences “substantial emotional upset,” without which “he would not have lost self-control,” is punished less than the person who “maintains his control despite being seriously provoked” because “the provocation-resistant person is controlled and rational when he kills the provoker.” 76Reid Griffith Fontaine, Adequate (Non)Provocation and Heat of Passion as Excuse Not Justification, 43 U. Mich. J.L. Reform 27, 45–46 (2009). Under a justification theory, however, the degree of provocation is important regardless of whether the actor experiences the subsequent emotional upheaval that leads to a loss of control. 77 See Dressler, supra note 75.

Some states have adopted new, broader terminology meant to encompass provocation and heat of passion. Specifically, the Model Penal Code provides that a homicide that would be classified as murder without some special circumstances can be considered manslaughter if committed as the result of “extreme mental or emotional disturbance for which there is a reasonable explanation or excuse.” 78Model Penal Code § 210.3(1)(b) (1985). Again, self-control plays a role in providing the defendant with a partial defense for his behavior. So, for example, in New York a court may allow a defendant to exercise an “extreme emotional disturbance defense,” which requires that the defendant produce “evidence that he or she suffered from a mental infirmity not rising to the level of insanity at the time of the homicide, typically manifested by a loss of self-control.” 79People v. Smith, 808 N.E.2d 333, 334 (N.Y. 2004) (quoting People v. Roche, 772 N.E.2d 1133, 1138 (N.Y. 2002)). “The defense requires proof of a subjective element, that defendant acted under an extreme emotional disturbance, and an objective element, that there was a reasonable explanation or excuse for the emotional disturbance.” Id. (citing People v. Moye, 489 N.E.2d 736, 738 (N.Y. 1985)). The law suggests that there are some emotional responses that will so overwhelm someone’s ability to think rationally that we may reasonably say that such a person is unable to control his own actions because of his emotional distress. States are not uniform in their approaches to extreme mental or emotional disturbance, and many do not permit the use of this defense. 80 See Mitchell N. Berman & Ian P. Farrell, Provocation Manslaughter as Partial Justification and Partial Excuse, 52 Wm. & Mary L. Rev. 1027, 1040 (2011). Only eleven states have adopted the extreme emotional or mental disturbance language of the Model Penal Code. See id. at 1044.

b. Intoxication

Although it is clear that being under the influence of drugs or alcohol can impair one’s ability to control one’s behavior, courts have been unwilling to allow defendants to evade culpability on this ground. 81 But see Dressler, supra note 37, at 326 (“Although it is true that self-induced intoxication as such never excuses wrongdoing, the condition that intoxication causes, e.g., a clouded mental state, unconsciousness, or insanity, may serve as an exculpatory basis in very limited circumstances.”). For example, the Wisconsin Supreme Court is quite typical in its ruling that loss of self-control caused by voluntary intoxication provides no basis for the insanity defense. 82 See State v. Kolisnitschenko, 267 N.W.2d 321, 326 (Wis. 1978). Indeed, the United States Supreme Court determined that a state’s statutory prohibition against considering voluntary intoxication in determining the existence of a mental state that is an element of a criminal offense did not violate the Due Process Clause. 83Montana v. Egelhoff, 518 U.S. 37, 56 (1996) (plurality opinion). Despite the fact that the intoxication did reduce the individual’s self-control, the Court held that the state need not consider this information. 84 See id. As Justice Ginsburg wrote, it did not violate the Constitution for the statutory regime to adjudge “that two people are equally culpable where one commits an act stone sober, and the other engages in the same conduct after his voluntary intoxication has reduced his capacity for self-control.” 85 Id. at 57 (Ginsburg, J., concurring). The Court further stated that addiction does not offer a viable defense to criminal responsibility when someone commits criminal acts under the influence of a chemical to which he is addicted. 86 See Dressler, supra note 37, § 24.02, at 348 (“The common law treats intoxication resulting from alcoholism or drug addiction as voluntary.”).

For the rational actor, there are several choices for how to think about an intoxication defense under a deterrence model. First, the rational actor ought to include, in his original cost–benefit analysis regarding the use of alcohol or drugs, the possibility of committing an undesirable criminal act because of that use. For deterrence purposes, this could deter the substance abuse that would have otherwise led to the criminal behavior. Alternatively, the utilitarian might say that if an intoxicated individual truly cannot comprehend the costs and benefits, or control his behavior to comport with his calculation, then deterrence is not possible, and so there is no need for punishment.

A moral theorist, instead, might suggest that because the actor had the initial choice to engage in behavior that led to the wrongful act, he must be held responsible for his actions. 87 See, e.g., Paul H. Robinson, Causing the Conditions of One’s Own Defense: A Study in the Limits of Theory in Criminal Law Doctrine, 71 ). A moral theorist may distinguish between situations in which the actor engaged in behavior unwittingly or deliberately. 88 See id. at 31–34. Our general societal stance toward drug and alcohol use, which suggests individual choice as the basic paradigm, 89The widely popular “Just Say No” anti-drug ad campaign of the 1980s and 1990s is one example of this attitude toward substance abuse, suggesting that personal choice and responsibility are the main determinants of drug use. See, e.g., Jerome Beck, 100 Years of “Just Say No” Versus “Just Say Know”: Reevaluating Drug Education Goals for the Coming Century, 22 Evaluation Rev. 15 (1998). typically helps us to place the unplanned consequences of a state of being that we have deliberately entered into in a special category. That is, although we may lack control over our actions while under the influence of a chemical substance, we believe that we had control over the action of taking the chemical substance, and so we do not lessen liability for those later actions. 90Courts have acknowledged that addiction can be considered a type of disease, which suggests that its consequences are beyond the momentary control of the individual; attempts to punish individuals for the status of addiction, rather than for the act of using drugs or for acts undertaken while under their influence, have not succeeded. See, e.g., Robinson v. California, 370 U.S. 660 (1962). Nonetheless, courts have largely been unwilling to exempt the taking of drugs, or acts undertaken under their influence, from criminal liability, even when addiction leads to these events. See, e.g., Powell v. Texas, 392 U.S. 514 (1968) (refusing to hold that acts committed while under the influence of alcohol, even if the defendant were an alcoholic, were exempted from criminal responsibility); Richard C. Boldt, The Construction of Responsibility in the Criminal Law, 140 U. Pa. L. Rev. 2245, 2309 (1992) (noting that, following the decision in Powell, most courts have refused to hold that acts committed while under the influence of alcohol or drugs were exempted from criminal liability, even if the defendant suffered from addiction).

Newer research on the possible genetic and biological roots of addiction has provided a potential critique of the free will assumption underlying the refusal to lessen culpability for intoxication or drug use. 91 See Stephen J. Morse, Addiction, Genetics, and Criminal Responsibility, 69 Law & Contemp. Probs. 165, 169–72 (2006) (describing a strand of modern addiction research as “treat[ing] the intentional conduct of the addict solely as the product of brain mechanisms”). Nonetheless, although there may be an impulse to take drugs or consume alcohol that is extraordinarily difficult to resist—perhaps just as (or more) difficult for some to resist as an impulse in the face of provocation—the law considers drug and alcohol use differently. As one criminal law scholar has said, “[W]e virtually never know whether the agent is in some sense genuinely unable or is simply unwilling to resist[.] . . . [R]esearch evidence concerning the characteristics that help people maintain control when faced with temptation or experiencing impulses is no more than a general guide.” 92 Id. at 187. Concerns about control permeate the discussion surrounding intoxication defenses but point to a different result than in some other areas of law because of our lay perception that the decision to become intoxicated is within the control of the ordinary person. 93 See Robinson, supra note 87, at 7–8.

4. Self-Control in Mitigation of Punishment

Responsibility for criminal behavior is taken into account not just at the time of determining what crime a particular defendant will be charged with or what defenses are available to him but also in sentencing. 94For example, federal sentencing guidelines allow for a downward departure in sentencing when individuals have “significantly reduced mental capacity” that might render them incapable of being fully responsible for their actions. See U.S. Sentencing Guidelines Manual § 5K2.13 (2011). When a crime has been committed in a manner that somehow implicates an inability to control one’s actions, the criminal may receive a lesser punishment than others. 95 See id. § 5K2.12–.13 (providing for lesser punishment when the defendant is unable to control his actions due to coercion and duress, or diminished mental capacity). As noted above, this supports a retributive vision of the law: we do not want to punish someone who lacked the ability to control his own actions because that person did not freely choose to engage in wrongdoing. 96 See supra Part I.A.2. There is also a utilitarian defense for offering lesser punishment; as noted above, it is perhaps an unwise investment in society’s resources to spend money on sanctions that are not able to offer future deterrence. 97 See supra notes 40–42 and accompanying text. As previously noted, this is perhaps counterintuitive to the insight of the law and economics scholar who suggests that, because the impulsive person is harder to deter, he should face a greater sanction.

For example, in death penalty eligible cases, a defendant’s degree of responsibility as determined by the jury plays a role in determining whether or not the death penalty should be imposed. In the sentencing phase of death penalty eligible cases, the defense and the prosecution both seek to present evidence about the defendant’s character and crime that goes beyond establishing culpability. 98 See Gregg v. Georgia, 428 U.S. 153, 193 (1976). The prosecution presents aggravating factors in an effort to show just how bad the defendant’s actions were, and the defense presents mitigating factors in an effort to reduce blameworthiness for the defendant’s actions. 99 Id. However, mitigating and aggravating factors are not always easy for jurors to distinguish. See Joshua N. Sondheimer, Note, A Continuing Source of Aggravation: The Improper Consideration of Mitigating Factors in Death Penalty Sentencing, 41 Hastings L.J. 409, 410 (1990). Death penalty mitigation typically involves an effort by the defense to provide “humanizing and explanatory evidence” that will justify a reduction in punishment. 100Craig M. Cooley, Mapping the Monster’s Mental Health and Social History: Why Capital Defense Attorneys and Public Defender Death Penalty Units Require the Services of Mitigation Specialists, 30 Okla. City U. L. Rev. 23, 52 (2005).

A defendant’s medical history, for example, may suggest problems from conception onward that could include neurological defects, mental retardation, fetal alcohol syndrome, chemical dependencies, or a wide variety of other mental health disorders that do not rise to the level of insanity but nonetheless suggest to the jury that the defendant was in some way incapable of exercising full agency in deciding to commit the crime in question. 101 Id. at 53–54. Similarly, documenting the “extensive poverty, neglect, and abuse” 102Emily Hughes, Mitigating Death, 18 Cornell J.L. & Pub. Pol’y 337, 345 (2009). that may have characterized the defendant’s childhood, and such conditions’ negative long-term effects, also may suggest to a judge and jury that the defendant’s capacity to exercise control over his actions was diminished. The reduction in punishment that mitigation strives for is rooted in self-control: an individual who was afflicted with these medical conditions, or who had this kind of upbringing, the argument goes, is simply unable to make the same choices and exert the same control over his actions that another person might, and thus should not be punished in the same manner.

II. A Brief Overview of Social Psychology Research on Self-Control

The self is the basic building block of human society and human interaction. Social psychologists consider that the self comprises three distinct strands of experience: reflexive consciousness, which is about one’s own construction of a concept of who one is; interpersonal being, which encompasses experiences, behaviors, and relationships; and finally, the executive function, which describes the self’s job as a decision maker and agent. 103Roy F. Baumeister, The Self, in 1 The Handbook of Social Psychology 680, 680 (Daniel T. Gilbert et al. eds., 4th ed. 1998). The ability to control one’s behavior and emotions is of particular interest to psychologists: without this executive function, “the self would be a mere helpless spectator of events, of minimal use or importance.” 104 Id. This latter function was for many years the least studied of the three aspects of the self, but it has garnered more attention in recent years as research on self-regulation and self-control has gained momentum. 105 See id. at 712. As Baumeister explains, it may be that, because so much of the executive function is not dramatic, instead consisting of many small decisions that are easily understood as guided by the rational pursuit of self-interest, it has been understudied. Accordingly, Baumeister notes, “[T]here has been something of a premium put on finding ways in which people act contrary to their rational self-interest”—hence the surge of interest in research on heuristics, biases, and other nonrational decision-making processes. Id. This research “reveal[s] flaws in the executive function” and for that reason helps us understand how the executive function works. Id. at 713. Throughout this Article, following the lead of scholars in the field, I use the terms self-control and self-regulation interchangeably.

Self-regulation research began in the 1980s, initially growing out of learning theory: just as rats might respond to a set of rewards and punishments set up by a laboratory scientist, so too researchers thought that an individual might devise her own set of rewards and punishments to induce or to discourage certain behavior of her own. 106 Id. at 716. Self-regulation researchers have studied how individuals regulate their thoughts, actions, impulses, and feelings. 107 Id. at 717. Researchers have sought to understand why some individuals are able to self-regulate toward future goals and away from undesirable activity, while others find such self-regulation difficult. 108 Id. And researchers have focused energy on how self-regulation affects individuals and society. 109 Id.

Research has demonstrated that a lack of self-control is associated with a wide range of negative outcomes, while high levels of self-control predict a host of positive outcomes. Walter Mischel’s work on delayed gratification is seminal in this area: he found that the ability to delay gratification in early childhood is implicated in many of our most cherished long-term goals, such as graduating from college or completing large projects at work. 110 See, e.g., Yuichi Shoda, Walter Mischel & Philip K. Peake, Predicting Adolescent Cognitive and Self-Regulatory Competencies from Preschool Delay of Gratification: Identifying Diagnostic Conditions, 26 Developmental Psychol. 978, 982 (1990) (finding that very young children’s capacity to delay their gratification in the face of a potential reward significantly correlated with higher SAT scores over a decade later). The ability to delay gratification is implicated any time a large but delayed reward awaits someone who forgoes momentary rewards: that is, in short, most of the time in our lives. Developmental psychologists have identified this trait, which is sometimes called “effortful control,” or the ability to inhibit a dominant response in order to perform a subdominant response, as an important dimension of temperament in children. 111 See Nancy Eisenberg et al., Effortful Control: Relations with Emotion Regulation, Adjustment, and Socialization in Childhood, in Handbook of Self-Regulation: Research, Theory, and Applications 263, 263 (Kathleen D. Vohs & Roy F. Baumeister eds., 2d ed. 2011) (quoting Mary K. Rothbart & John E. Bates, Temperament, in 3 Handbook of Child Psychology: Social, Emotional, and Personality Development 99, 129 (William Damon et al. eds., 6th ed. 2006)). Psychologists have suggested that effortful control, which includes the ability to control one’s attention, to inhibit certain behaviors, and, sometimes, to activate other behaviors, plays a critical role in regulating individual emotion. 112 Id. at 264. Effortful control helps individuals distract themselves from negative emotions and inhibits the expression of negative emotions or behavioral expressions of those emotions. 113 Id.

Research has shown that low effortful control in children is expressly linked to negative qualities such as higher impulsivity. 114 Id. at 270. High effortful control in children is also linked to compliance—particularly compliance that is internally motivated rather than situationally required and that seeks to suppress a particular behavior rather than initiate or sustain a behavior. 115 Id. at 268. Effortful control is also positively related to other positive dimensions of moral development, such as conscience, 116Grazyna Kochanska & Amy Knaack, Effortful Control as a Personality Characteristic of Young Children: Antecedents, Correlates, and Consequences, 71 J. Personality 1087, 1102 (2003) (finding that higher effortful control in small children predicted a more internalized conscience). empathy, 117 See, e.g., Ivanna K. Gurthrie et al., The Relations of Regulation and Emotionality to Children’s Situational Empathy-Related Responding, 21 Motivation & Emotion 87 (1997) (finding that children high in effortful control demonstrated more facial sadness while watching a sad movie than children low in effortful control). and prosocial behavior. 118 See, e.g., Nancy Eisenberg et al., The Relations of Children’s Dispositional Prosocial Behavior to Emotionality, Regulation, and Social Functioning, 67 Child Dev. 974 (1996); Nancy Eisenberg et al., The Relations of Regulation and Emotionality to Resiliency and Competent Social Functioning in Elementary School Children, 68 Child Dev. 295 (1997).

Control and its absence continue to matter as children grow up: at age eighteen, individuals who are low in self-control are more impulsive, danger seeking, aggressive, and interpersonally alienated. 119Avshalom Caspi & Phil A. Silva, Temperamental Qualities at Age Three Predict Personality Traits in Young Adulthood: Longitudinal Evidence from a Birth Cohort, 66 Child Dev. 486, 495 (1995). By twenty-one, low self-control is related to an individual’s number of criminal convictions. 120Bill Henry et al., Staying in School Protects Boys with Poor Self-Regulation in Childhood from Later Crime: A Longitudinal Study, 23 Int’l J. Behav. Dev. 1049, 1067 (1999). Longitudinal studies suggest that the problem of self-control has an important relationship with crime: low self-control at age fourteen predicted criminal offenses at twenty and thirty-two. 121 See Lea Pulkkinen, Offensive and Defensive Aggression in Humans: A Longitudinal Perspective, 13 Aggressive Behav. 197, 203–04 (1987); Lea Pulkkinen, Self-Control and Continuity from Childhood to Late Adolescence, in 4 Life-Span Development and Behavior 63 (Paul B. Baltes & Orville G. Brim, Jr. eds., 1982). Self-control is directly linked to criminal behavior, and it is also linked to the correlate of addiction and substance abuse. 122 See Jay G. Hull & Laurie B. Slone, Alcohol and Self-Regulation, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 466 (Roy F. Baumeister & Kathleen D. Vohs eds., 2004); Michael A. Sayette & Kasey M. Griffin, Self-Regulatory Failure and Addiction, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 505. Because drug use is itself unlawful and being under the influence of drugs or alcohol is a significant predicate for many criminal activities, 123 See Kathryn D. Scott et al., The Role of Alcohol in Physical Assault Perpetration and Victimization, 60 J. Stud. Alcohol 198 (1999) (finding a strong relationship between alcohol use and physical assault perpetration); Samantha Wells et al., Alcohol-Related Aggression in the General Population, 61 J. Stud. Alcohol 626 (2000) (finding that alcohol intoxication is associated with increased levels of physical aggression); Helene Raskin White et al., Problem Drinking and Intimate Partner Violence, 63 J. Stud. Alcohol 205 (2002) (finding that problem drinking is significantly associated with intimate partner violence perpetration). this is yet another way that self-control failure is implicated in criminal behavior. Indeed, psychologist Travis Hirschi has argued that the most important individual-level trait of relevance to an individual’s propensity to commit a crime is self-control; the commonality among all criminal acts is the provision of “immediate benefit at the risk of long-term pain.” 124Travis Hirschi, Self-Control and Crime, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 537, 537; accord Michael R. Gottfredson & Travis Hirschi, A General Theory of Crime 88 (1990). Hirschi and Michael Gottfredson posit that criminal behavior, in contexts as diverse as juvenile delinquency, violent crime, and white collar crime, is based almost wholly on a lack of self-control. 125 Gottfredson & Hirschi, supra note 124, at 88–91.

While some psychologists have focused on identifying the negative and positive effects of self-control on emotions and behavior, others have been working to formulate an understanding of how the process of controlling oneself actually works. Of particular interest to psychologists is how individuals self-regulate to minimize their engagement in undesirable behavior and maximize their engagement in desirable behavior. 126 See Baumeister, supra note 103, at 717. The executive function of the self means that there are few behaviors that we truly lack the capacity to control, according to psychologists; as psychologist and self-control researcher Roy Baumeister has noted, there are almost no “irresistible impulses,” demonstrated by the fact that few serial killers take action when an armed officer is present. 127Roy F. Baumeister, Yielding to Temptation: Self-Control Failure, Impulsive Purchasing, and Consumer Behavior, 28 J. Consumer Res. 670, 671 (2002). In any event, Baumeister suggests that it is not the impulse itself that must be controlled, because impulses are automatic responses to particular stimuli. 128Roy F. Baumeister & Todd F. Heatherton, Self-Regulation Failure: An Overview, 7 Psychol. Inquiry 1, 2 (1996). Instead, resistance to the natural consequences of the impulse is the key to self-control. 129 Id.; Baumeister, supra note 127, at 671. How, then, do individuals regulate their behavior and resist acting on impulses to engage in certain behavior, and how, conversely, do individuals choose effortfully to engage in other behavior?

In the past twenty years, psychologists have been particularly interested in questions about how self-regulation works. Some researchers have suggested that self-control represents a triumph of effortful cognition over automatic behavior, 130 See Baumeister & Heatherton, supra note 128. while others have suggested that self-control describes behavior governed by “cool,” cognitive behavior, rather than “hot,” affective, or impulsive responses. 131 See Janet Metcalfe & Walter Mischel, A Hot/Cool-System Analysis of Delay of Gratification: Dynamics of Willpower, 106 Psychol. Rev. 3 (1999). Still others have characterized self-control as behavior that is in accordance with rational preferences, rather than visceral reactions. 132 See George Loewenstein, Out of Control: Visceral Influences on Behavior, 65 Organizational Behav. & Hum. Decision Processes 272, 288–89 (1996). And others have noted that self-control requires actions that comport with long-term, rather than short-term, goals. 133 See Kentaro Fujita et al., Construal Levels and Self-Control, 90 J. Personality & Soc. Psychol. 351, 353–55 (2006).

Although different, these various theories are not completely conceptually unrelated; they all similarly describe a struggle between two types of behavior: on one side, the automatic, “hot,” or visceral short-term-benefit choice, and on the other, the cognitively effortful, “cool,” or rational long-term-benefit choice. Taken together, the varying theories offer a useful picture of some classic self-control challenges. For example, consider smoking. Smoking may represent a failure of self-control because even when one consciously wants not to smoke, one’s automatic response may be to light up a cigarette. So, too, smoking may appeal to the “hot,” affective system and may satisfy a visceral urge, rather than being a cognitively effortful, “cool,” rational preference. And finally, smoking may satisfy a short-term goal of enjoying the experience, but it will fail to satisfy a long-term goal of good health. Similarly, foregoing a fun activity such as watching a movie in order to work and accomplish long-term career goals, or engaging in an unpleasant action such as a painful medical procedure because of its long-term positive health outcome, may likewise involve overriding and replacing automatic or impulsive behavior with more cognitively taxing behavior. This cognitively taxing behavior permits individuals to act in line with their long-term, rather than short-term, goals.

Why are some individuals able to comport themselves in line with long-term goals or cognitively effortful, “cool,” rational preferences, while others are overwhelmed by “hot,” visceral impulses or the lure of the short-term benefit? The psychological self-control literature has developed a three-stage model for how individuals comport their behavior to their wishes—that is, how they exercise control over their actions. 134Baumeister & Heatherton, supra note 128, at 2. First, one must have standards—that is, some kind of benchmark of ideals or goals. 135 Id. A lack of standards, conflicting standards, or confusion about the relevant standard will impede self-regulatory efforts; 136 See Robert A. Emmons & Laura A. King, Conflict Among Personal Strivings: Immediate and Long-Term Implications for Psychological and Physical Well-Being, 54 J. Personality & Soc. Psychol. 1040, 1045–47 (1988) (finding that conflict between personal striving and ambivalence results in negative psychological and health effects, which supports a connection between personal strivings and personality); Elizabeth Van Hook & E. Tory Higgins, Self-Related Problems Beyond the Self-Concept: Motivational Consequences of Discrepant Self-Guides, 55 J. Personality & Soc. Psychol. 625, 632 (1988) (finding that individuals who possess conflicting standards are more likely to experience psychological distress, which “emphasizes the importance of self-evaluative standards in self-guidance and self-assessment”). so, too, setting an inappropriately high or low standard can interfere with self-regulation. 137Todd F. Heatherton & Nalini Ambady, Self-Esteem, Self-Prediction, and Living Up to Commitments, in Self-Esteem: The Puzzle of Low Self-Regard 131 (Roy F. Baumeister ed., 1993). Second, one must monitor one’s behavior to examine whether or not it comports with one’s benchmark. 138Baumeister & Heatherton, supra note 128, at 2. Monitoring requires one to keep a close tab on one’s behavior; so, for instance, behavior that reduces one’s ability to self-monitor, such as drinking alcohol, can impede self-regulation in other areas. 139 Id. Finally, one must be able to engage in what psychologists call “operating” behavior: that is, actually altering the behavior that does not measure up to the benchmark so that it comports with the relevant standard. 140 Id. This understanding of self-control processes was strongly influenced by earlier work by Charles S. Carver and Michael F. Scheier that merged cybernetic theory with self-awareness theory to suggest that self-regulation operates in a “TOTE” feedback loop: Test how one is doing against a standard, Operate on the self to change behavior to better comport with the standard, Test again to see how performance now compares to the standard, and Exit when one has reached the appropriate level of behavior. See Charles S. Carver & Michael F. Scheier, Attention and Self-Regulation: A Control-Theory Approach to Human Behavior 16 (1981). Such a feedback loop could also work to effectuate a goal of increasing distance between oneself and a negative standard. See id. at 17. There may be important differences in the way that the self regulates toward and against standards, but this has been relatively underexplored empirically. This feedback model is well established as a basic premise of self-control research. 141Baumeister, supra note 103, at 717.

The two most intriguing and dominant lines of recent self-control research in modern social psychology concern how individuals evaluate present versus future consequences, and under what circumstances the individual is able to bring her behavior into line with her desires. This first line of research—on “construal level theory”—attempts to unpack the mechanisms of self-control by looking at the implications of differences in the way people process information about events that will occur at different times. The second line of research—under the umbrella of the “strength” model of self-control—looks at the conditions under which self-control does and does not occur, suggesting that because self-control requires such a level of cognitive effort, it is a limited resource that becomes depleted over time. I consider these two perspectives, respectively, in Parts III and IV below, along with their implications for criminal law and theory.

III. Construal Level Theory

A. Research on Construal Level Theory

Focusing closely on the tension between short-term goals or impulses and long-term goals or rational preferences that inheres in self-control problems, social psychologists have tried to understand how people may think differently about near-future and distant-future events to determine if that difference can account for self-control successes and failures. “Construal level theory” in psychology has tried to integrate previous theories about this tension into a coherent theoretical framework by suggesting that the reason people fail in their efforts to act in accordance with long-term goals is because of systematic differences in how people tend to conceptualize events depending on psychological distance from the event. (Although psychological distance can include factors like social distance, most typically it is described as temporal distance. 142 See, e.g., Kentaro Fujita, Seeing the Forest Beyond the Trees: A Construal-Level Approach to Self-Control, 2 Soc. & Personality Psychol. Compass 1475 (2008); Nira Liberman et al., The Effect of Temporal Distance on Level of Mental Construal, 38 J. Experimental Soc. Psychol. 523, 532 (2002). )

Construal level theory suggests that people typically construe events in the distant future at a “high” level, thinking about those events broadly and in an abstract way. 143Yaacov Trope & Nira Liberman, Temporal Construal, 110 Psychol. Rev. 403, 405 (2003). These “high-level construals” encapsulate the central, defining features that convey the essential meaning of the event. 144 Id. In contrast, “low-level construals” capture the concrete, specific, mundane features of the event. 145 Id. For an example, imagine a high-level construal of the Thanksgiving holiday—say, the importance of friends and family and appreciation of one’s blessings—contrasted with a low-level construal—what time the turkey needs to go in the oven or how to fit twelve people at a table for eight. The high-level construal would be what one might experience in thinking about the holiday during August, and the low-level construal is what one experiences in late November.

In one set of studies, researchers first primed participants to think either in low-level or high-level ways about a particular topic by asking some participants to focus on reasons for maintaining good health or personal relationships (typically invoking abstract concepts) and asking others to focus on how one might maintain good health or good personal relationships (typically bringing to the fore a variety of concrete steps that one needs to take). 146Fujita et al., supra note 133, at 354. In one part of the research, participants then assigned monetary value to a variety of items, both for receiving the items immediately and for receiving the items months into the future. When individuals were primed with low-level construal rather than high-level construal, they demonstrated a significant preference for immediate over delayed rewards, while individuals who were primed with high-level construal were less likely to prefer the immediate reward. 147 Id. at 354–55. The researchers concluded that, because of the demonstrated difference in willingness to delay rewards, high-level construal helps individuals exercise self-control. 148 Id. at 355. In another part of this research, the participants were asked to squeeze a handgrip for as long as possible, with the promise that the longer they persisted, the more accurate self-relevant feedback they would receive on their performance in the experiment. Here, individuals primed with a high level of construal held the handgrip for a significantly longer time period than individuals who were primed with low-level construal. 149 Id. at 356. Again, because participants had to maintain control over their handgrip performance—to persist at a fairly unpleasant task to achieve accurate feedback—the researchers concluded that high-level construal helps to facilitate self-control. 150 Id. at 357.

In another set of studies, individuals who imagined scenarios described in high-level terms were more likely to predict that they would feel bad indulging in temptation than those who imagined scenarios described in low-level terms. 151 Id. at 363. That is, potential self-control failure appeared to be more upsetting to individuals who imagined the failure through a high-level lens. Other studies have suggested that individuals’ preferences as to events in the distant future tend to reflect greater self-control than those in the near future, 152 See George Ainslie & Nick Haslam, Hyperbolic Discounting, in Choice over Time 57, 58 (George Loewenstein & Jon Elster eds., 1992); Antonio L. Freitas et al., Abstract and Concrete Self-Evaluative Goals, 80 J. Personality & Soc. Psychol. 410, 417 (2001). which, when considered in conjunction with research on temporal construal, suggests that high-level construal may be a driver of self-control.

Construal level theory has been offered as one potential explanation for the robust findings on hyperbolic discounting that have emerged from experimental work in economics. 153See David Laibson, Golden Eggs and Hyperbolic Discounting, 112 Q.J. Econ. 443, 445–46 (1997) (defining hyperbolic-discount functions as those characterized by a high discount rate over short horizons and a low discount rate over long horizons). Hyperbolic discounting describes a general tendency to weigh events in the present or near future more heavily than those in the more distant future. 154 See, e.g., James Q. Wilson & Richard J. Herrnstein, Crime and Human Nature (1985) (suggesting that temporal discounting is important in explaining criminality because the benefits of crime occur at a different time than its costs); David S. Lee & Justin McCrary, Crime, Punishment, and Myopia 28 (Nat’l Bureau of Econ. Research, Working Paper No. 11491, 2005), available at http://www.nber.org/papers/w11491.pdf. Construal level theory suggests that hyperbolic discounting may take place because “the rich, detailed, and possibly ambiguous information contained in real events is represented more abstractly, in terms of relatively simple and structured mental models, when the events are expected in the distant future than in the future.” 155Trope & Liberman, supra note 143, at 406. While hyperbolic discounting suggests that individuals place more emphasis on near-term events than far-term events, construal level theory adds an additional layer: the events are not just weighted differentially, they are also conceptualized differently. Construal level theory suggests that the individual contemplating committing a criminal act construes events in the near-term and long-term in fundamentally distinct ways that can produce a systematic bias in evaluation.

The difference between construal level theory and hyperbolic discounting is subtle but important. Take, for example, a person contemplating an armed robbery. Hyperbolic discounting means that when that person weighs the benefits of the crime (money) against the possible sanction (prison), he will discount the future sanction by some order of magnitude because it is going to occur, if ever, in the future. This means that even if the costs of prison were commensurate with or outweighed the benefits of the robbery, the costs of prison as computed might not outweigh the benefits after these costs were discounted because they would occur in the future. But construal level theory suggests that the individual would contemplate the immediate act and the future sanction in fundamentally different ways. The immediate act would be most likely construed in its low-level way—for example, what actions are needed to effect the robbery or what immediate benefit it will produce. The high-level construal of the act—its wrongfulness or the potential for serious harm to any of the participants or others—is likely to be less salient. In contrast, the potential future sanction is likely to be construed in broad terms—abstract punishment for wrongdoing—rather than in its specific, concrete terms of imprisonment, daily isolation from family and friends, difficult prison conditions, etc. Rather than performing a simple cost–benefit analysis (even one that is “wrong” due to hyperbolic discounting), the potential criminal actor is comparing dramatically different factors that may be hard to translate into a cost–benefit equation.

Temporal construal also may affect rational probability estimates, further casting doubt on the accuracy of the law and economics vision of a potential criminal actor’s cost–benefit analysis. Recent research on construal level theory and probability has begun to suggest that high-level construal may actually reduce individuals’ estimates about the likelihood of an event. 156Cheryl Wakslak & Yaacov Trope, The Effect of Construal Level on Subjective Probability Estimates, 20 Psychol. Sci. 52 (2009). This means that the potential criminal actor may actually underestimate the likelihood of punishment if that punishment is perceived as temporally distant and therefore envisioned at a high level of mental construal.

Weaving together the strands of self-control research and construal theory, psychologists have suggested that self-control “can be broadly conceptualized as making decisions and acting in accordance with global, high-level construal . . . rather than local, low-level construal.” 157Fujita et al., supra note 133, at 352. In essence, psychologists posit that the regulation of behavior requires one to act in concert with high-level construals rather than low-level construals. Individuals who are able to keep their mental representations of actions in line with high-level construals, then, are better at self-control than those whose mental representations focus on low-level qualities. So, for example, the individual whose mental representation of smoking captures its health effects rather than its pleasurable physical experience will be better able to refuse a cigarette, while the individual whose mental representation of a painful medical test involves maintaining good health rather than unpleasant physical experiences will be more likely to keep her doctor’s appointment.

B. Implications of Construal Level Theory in Criminal Law

If self-control failure, as it is explained in psychology, means a failure to appreciate the abstract nature of an event and to be swamped by its low-level construal in the moment, how might this relate to criminal law? The first thing to consider is what low-level and high-level construals might look like to the potential criminal actor. Any potential object, event, or action 158 Id. can be construed at a high or low level, and high and low levels will be defined differently depending on the characteristics of the person doing the construing as well as the situation. That is, construal level is in the eye of the beholder. As Fujita and colleagues explain:

[F]or a student committed to performing well on final exams, an invitation to attend a party while studying may create a conflict between doing well on final exams (high-level concern that suggests not going to the party) and seeing one’s friends (low-level concern that suggests going to the party). . . . [W]hat are high- and low-level concerns may change as a function of the situation or individual differences. For example, for a person unconcerned about academics but for whom social relationships are highly valued, seeing one’s friends at a party may represent a high-level concern, whereas studying for final exams may represent a low-level concern. 159 Id. at 352–53.

This means that differences in construal are highly contextual, and both the individual and type of crime may implicate different types of construal. Because any construal level analysis will be person-dependent and context-dependent, it may be impossible to draw any broad inferences, but it is worthwhile to at least begin to chart out some possibilities.

Imagine a potential criminal actor who is thinking about committing an armed robbery tomorrow. A basic economic analysis of costs and benefits would weigh factors including the amount of money likely to be made, the positive use of the money (food, other consumable items), and other potential benefits (enjoyment of criminal activity, perhaps) against the costs of the crime (costs of materials used in furtherance of the crime, opportunity costs of other possible activities during that time, potential punishment for the crime, or mental disturbance from engaging in criminal activity). 160Of course, not all of these factors may be applicable for every person contemplating an armed robbery. But, adding in the insights from psychology, it would be easy to distill criminal activity into a classic conflict between short-term and long-term goals. One’s short-term goals are to get the value from the robbery, and one’s long-term goals are presumably to avoid punishment (or, perhaps, not to engage in criminal activity). And under construal theory, one could imagine that a self-control failure occurs because the short-term act is seen in low-level, concrete terms (immediate money), rather than in high-level, abstract terms (violating the law, subjecting oneself to punishment by the state). If one could control oneself, one would choose to have one’s actions comport with the long-term goal of avoiding punishment and to be a law-abiding citizen. If the robbery were going to occur next year instead of tomorrow, construal level theory predicts that the potential actor would be more likely to think about the abstract elements of the crime, rather than its mundane details, and would be more focused on the potential harm to self and society inherent in the act (high-level) instead of the money and purchasing power to be gained (low-level). This actor would find the prospect of the robbery less appealing because it would be seen in its most abstract form: with the potential for harm to self and others in the form of both violence and punishment by the state.

This simple paradigm is appealing, and taken at its face value, it provides support for the idea that all criminal acts are a product of self-control failure. Unfortunately, it does not adequately capture the host of possibilities for both low-level and high-level construal inherent in a decision to engage in criminal activity. So, for instance, low-level construal of an armed robbery contemplated for tomorrow could include both positive and negative features, such as buying a gun, getting money, being arrested or injured, putting food on the table for one’s children, or buying drugs. High-level construal of these same events could include providing for one’s family, righting some societal injustice, paying one’s debts, endangering one’s own or others’ safety, or violating the law.

What this suggests is that any one criminal act might be a triumph of low-level over high-level construal, meaning that it is a failure of self-control—but it might not. It could be that a criminal actor focuses on the high-level construal (say, providing for children) and overcomes his distaste for the low-level features of the crime to comport with those high-level goals. Imagine an armed robber who, much like our patient who opts to undergo a painful medical procedure because of its long-term health benefits, chooses to exercise self-control by carrying out a criminal act to comport with his long-term goals. In another case, it might be that both the abstract, essential elements of the act and the concrete, specific elements of the act are in alignment for the particular individual: perhaps the person thinks that the low-level details of the criminal act are positive and that the high-level meaning of the criminal act is positive. This means that there is no particular need for any self-control mechanism; the person merely acts in accordance with his preferences. Construal level theory’s vision of self-control suggests that crime is a failure of self-control only when an individual has, somewhere in his mind, a goal of not breaking the law. 161Whether this goal is a product of some belief in adherence to the law or merely a desire to avoid punishment may be an important distinction, but for now I do not address this question. Crimes by individuals who do not have this goal cannot be characterized as self-control failures, just as smoking by someone who is an unrepentant smoker or overeating by someone who enjoys being obese cannot be characterized as self-control failures either.

Although the simple paradigm may be inaccurate in a number of cases, there are presumably some situations where it is a reasonable representation of how an individual may perceive potential criminal behavior. In those cases, crimes that are unplanned or immediate may indeed represent a failure of self-control; the individual acts in concert with his low-level construal of immediate events and fails to act in a way that he might were he able to focus on high-level construal. High-level construal, because it focuses on the “cooler” aspects of the event, is thought to more closely track with what economists would call rational thinking. 162 See Fujita, supra note 142, at 1483 (“Cooler, more cognitive thinking is generally linked to more global, high-level construals, which may allow people to transcend the affective, appetitive features of a temptation.”). This means that self-control failure that is a result of temporal construal is a failure to “rationally” weigh the costs and benefits of the action. 163This dovetails with Stephen Morse’s explanation that “control excuses are best understood as irrationality or internal coercion claims.” Morse, supra note 6, at 1656.

For a utilitarian, these effects of temporal construal pose a serious threat to the model of the rational criminal actor. Hyperbolic discounting can be accounted for in the economic model: if individuals regularly discount future events by some factor, we could simply increase punishment to compensate for that discount. But if our prospective audience is not capable of performing a standard cost–benefit analysis because it is construing short-term and long-term events in fundamentally different ways that cannot be directly compared to one another, then the economic model is not particularly helpful. In essence, temporal construal theory suggests that in order to turn potential criminal actors into rational actors, the law must draw the potential criminal actor’s attention to the high-level construal of the event in question. Without that high-level construal to guide them, individuals are more likely to commit crime. How might one best draw a potential criminal actor’s attention to the high-level features of a situation? Are there sanctions with special features that can shake the individual’s natural tendencies of temporal construal? It seems very difficult to deter an action by an individual that will provide him with a host of short-term benefits without in some measure forcing him to consider the abstract and high-level elements of his action, and it is in turn difficult to imagine a system that will be certain to remind him (as he contemplates criminal activity) of these abstract features of crime and punishment. 164Recent government actions regarding public health concerns, such as obesity and road hazards, comport with psychological findings about construal. For example, New York City’s recent push to post calorie information on all fast food is in part an effort to change how individuals construe the choice to eat particular foods. See James Barron, Restaurants Must Post Calories, Judge Affirms, N.Y. Times, Apr. 17, 2008, at B4 (noting that the district court, in upholding the law, stated that providing calorie information will lead at least some consumers to choose lower calorie options, thereby reducing the incidence of obesity). The posting of calories serves as a reminder of the high-level construal of the act of eating, say, a Big Mac. While the low-level construal would focus on the taste of the burger and the enjoyment of satiating one’s hunger, the high-level construal is about the longer term consequences of eating unhealthy, fatty foods.Similarly, campaigns about texting while driving attempt to shift consumer focus away from the low-level construal—immediate gratification of communicating instantly with one’s friends—to the high-level construal of potential danger and longer term risk to life and limb. See, e.g., A Message from Secretary LaHood, Distraction.gov, http://www.distraction.gov/content/about-us/message-from-secretary-LaHood.html (last visited Mar. 27, 2012) (describing a campaign initiated by the U.S. Department of Transportation to end distracted driving). Interestingly, some of these campaigns attempt to invoke grave consequences by depicting them in graphic and detailed ways—fighting a low-level positive construal with a low-level negative construal. See, e.g., PSA Highlights Aftermath of Distracted Driving Crashes, WBALTV.com (Mar. 3, 2011, 6:07 PM), http://www.wbaltv.com/r/27067600/detail.html (noting that videos discourage texting while driving by “depicting the fateful moments leading up to a crash and the heartbreaking news at the hospital”). This approach may be supported by new research suggesting that, under certain conditions, low-level construals may help self-control efforts. See Brandon J. Schmeichel et al., Self-Control at High and Low Levels of Mental Construal, 2 Soc. Psychol. & Personality Sci. 182 (2011). This suggests that a system of deterrence that looks like our present one fails to take into account some fundamental features of human information processing and that the economic model of criminal decision making may not adequately capture the way that individuals make decisions about criminal behavior.

For someone concerned with retribution, construal level theory poses a different problem. If people have systematic ways of processing information so that events at a psychological distance are seen in their high-level construal and events temporally or otherwise near are seen in a low-level construal, then various societal factors, such as temporal and social distance, ought to be taken into account when deciding how much responsibility an individual has for his actions. In addition, there are individual differences in construal patterns; perhaps these need to be taken into account when considering criminal behavior.

On the other hand, a retributivist may be able to take comfort from the fact that the criminal actor focused on the wrong aspect of the situation, even if there is a natural tendency to do such a thing. Instead of suggesting that the focus on low-level construal—the short-term focus—excuses the actor, a retributivist might consider that the actor inappropriately focused on certain aspects of criminal acts and not others. When so many other individuals are able to prioritize the high-level over the low-level construal, and when both ways to construe a situation are plausible, the retributivist may not feel the need to worry as much about the problem of self-control as an excusing factor.

In addition, this vision of misguided focus may offer support for a differential assignment of responsibility for crimes that are premeditated and ones that are not. Because research findings suggest that events in the longer term future are viewed at a higher level of abstraction, 165 See supra notes 146–48 and accompanying text. the person who contemplates a criminal act in the future is likely to have a higher level construal of that act. Higher level construal invokes classical rational cognition, and so these actors, if they follow through on the action that they contemplate, are thought to have acted rationally, which means that they have self-control. For these actors, then, we have a situation where, under deterrence theory, society has set the sanction too low to deter them. So we punish them merely to make our system credible for others who may recalibrate their own rational analysis accordingly. But for retributivists, the classical rational cognition that is involved in the decision to take action means that this individual is truly morally culpable and deserves punishment.

IV. The Strength Model of Self-Control

A. Research on the Strength Model of Self-Control

Another line of robust self-control research over the last several decades has demonstrated that self-control consumes valuable cognitive resources from a fixed resource base. 166Although the strength model stemmed originally from the automatic versus effortful cognition model of self-regulation, which can be integrated into the temporal construal model, the strength model itself has not been explored vis-à-vis its relationship with temporal construal. As Fujita and colleagues have noted, “[A] construal-level analysis is largely mute with respect to ego-depletion models of self-control. . . . [A] construal-level analysis . . . has little to say about the ego-depletion model’s core tenet that self-control is a limited resource.” Fujita et al., supra note 133, at 364 (citation omitted). This has led to the “strength” model of self-control, 167The strength model has been experimentally tested against two competing models, the schema model and the skill model. The schema model suggests that self-regulation is a set of behaviors that can be activated by the appropriate setting so that, in repeated settings, self-control would remain at a similar level. Mark Muraven et al., Self-Control as Limited Resource: Regulatory Depletion Patterns, 74 J. Personality & Soc. Psychol. 774, 775 (1998). The skill model suggests that individuals ought to learn and improve as they engage in repeated self-regulation, so that their performance in a series of self-control tasks ought to steadily improve. Id. Research has not provided support for these theories but has provided support for the strength model. See id. at 786. also called in the psychology literature “ego depletion.” 168Researchers have settled on the term “ego depletion,” rather than self-control depletion, because the resource pool from which self-control comes seems to be broader, serving more than just a regulatory function. See Roy F. Baumeister et al., Ego Depletion: Is the Active Self a Limited Resource?, 74 J. Personality & Soc. Psychol. 1252, 1252 (1998). The basic concept of this dominant model of self-control is that individuals have a finite supply of energy, or strength, that can be used for self-control; when that self-control has been used up, it is no longer available for some subsequent period of time. 169 Id. The model suggests that different individuals have differing levels of self-control. 170Baumeister & Heatherton, supra note 128, at 3. There are at least two important insights that stem from the strength model of self-control. First, self-control is harder to achieve after other tasks have used up an individual’s self-regulatory strength; 171Baumeister et al., supra note 168, at 1263. and second, just as strength of a muscle grows over time with practice, use of self-control over time can ultimately increase one’s capacity for self-control. 172Baumeister & Heatherton, supra note 128, at 3.

Because self-regulatory strength is a fixed resource, research has suggested that other tasks can deplete this energy. Early research suggested that an initial self-control task would interfere with effectiveness on a subsequent task. In one notable experiment, psychologist Baumeister and his colleagues brought participants into a room in which chocolate-chip cookies had just been freshly baked. 173Baumeister et al., supra note 168, at 1254. The participants were seated at a table on which the fresh cookies, along with a few candies and a bowl of radishes, were already set. Participants were asked either to eat only the cookies or candy, or to eat only the radishes. Subsequently, participants were asked to complete an unsolvable problem-solving task. Individuals who had been asked to eat radishes (and had therefore exerted self-control by not eating the forbidden cookies) spent significantly less time working on the problem than those who had eaten the cookies. 174 Id. at 1255. The numbers were dramatic: the radish eaters spent a little over eight minutes working on the problem, while the cookie eaters spent almost nineteen minutes plugging away. Id. In a control condition where no food was eaten, participants spent about twenty minutes on the task, a number not significantly different from that of the cookie eaters’. Id. This result suggested that those who had exercised self-control in one task found their self-control depleted when it came time to exercise it in a later task. 175 Id. at 1256.

Subsequent research has suggested that it is not only self-control tasks, per se, that deplete self-regulatory strength. Engaging in conscious choices, 176 Id. at 1256–58. engaging in self-control over one’s emotional responses, 177 Id. at 1258–59. undergoing stressful experiences, 178Megan Oaten & Ken Cheng, Academic Examination Stress Impairs Self-Control, 24 J. Soc. & Clinical Psychol. 254, 254 (2005). and being reminded of one’s mortality 179Matthew T. Gailliot et al., Self-Regulatory Processes Defend Against the Threat of Death: Effects of Self-Control Depletion and Trait Self-Control on Thoughts and Fears of Dying, 91 J. Personality & Soc. Psychol. 49, 55–57 (2006). produced similar diminution in research subjects’ performance at other self-control tasks. Psychologists have concluded that there is a resource for executive functions of the self that is affected by a variety of factors. 180 See Baumeister, supra note 103, at 719.

Research on the relationship between emotions and self-control presents a complicated picture. Empirical data suggests that emotional distress causes a failure of self-control, and some research has suggested that this is because people spend their cognitive resources on affect regulation instead of behavior regulation. 181Dianne M. Tice et al., Emotional Distress Regulation Takes Precedence over Impulse Control: If You Feel Bad, Do It!, 80 J. Personality & Soc. Psychol. 53, 53 (2001). In an effort to change one’s mood or emotional state, one might engage in behavior that one believes will change one’s affect, such as overeating, drinking alcohol, or going to a movie when one should be studying, even if that behavior is contrary to how one might act in a state of full self-control. 182 Id. Thus, it is not that emotional distress itself uses up the same resource needed for control of impulses; it is that individuals prioritized another type of self-regulation—regulation over their affect. 183 Id. One study suggested that “emotional distress interfered with self-regulation only when people believed . . . that abandoning impulse control in a particular sphere would allow them to reap rewards or pleasures that would be able to reduce their distress and produce a more positive emotional state.” 184 Id. at 64. In any event, whatever the mechanism at play, researchers agree that “emotional distress is itself often a drain on regulatory strength, insofar as people try to cope with and escape from the bad feelings. These efforts leave less strength left to use for task persistence, resisting temptation, and the like.” 185Baumeister, supra note 103, at 719. Baumeister continues, “Clarifying the relations between emotional distress and regulatory success and failure will be one of the most theoretically interesting challenges of the coming years.” Id. On the other hand, the effect of positive mood on self-control has been more mixed. Some research has found that positive mood boosts self-control, 186 See Barbara L. Fredrickson et al., The Undoing Effect of Positive Emotions, 24 Motivation & Emotion 237, 238–39, 255–56 (2000); Tice et al., supra note 181. but other research has found an opposite effect, 187 See Melissa A. Cyders & Gregory T. Smith, Mood-Based Rash Action and Its Components: Positive and Negative Urgency, 43 Personality & Individual Differences 839, 840–41 (2007); Ayelet Fishbach & Aparna A. Labroo, Be Better or Be Merry: How Mood Affects Self-Control, 93 J. Personality & Soc. Psychol. 158, 158 (2007); Carrie L. Wyland & Joseph P. Forgas, On Bad Mood and White Bears: The Effects of Mood State on Ability to Suppress Unwanted Thoughts, 21 Cognition & Emotion 1513, 1519 (2007). and yet more research has found no effect at all. 188Muraven et al., supra note 167, at 776–79; Harry M. Wallace & Roy F. Baumeister, The Effects of Success Versus Failure Feedback on Further Self-Control, 1 Self & Identity 35, 38 (2002).

One interesting recent finding is that individuals can “pace themselves” with respect to self-control resources, essentially treating self-control as a limited resource. 189Mark Muraven et al., Conserving Self-Control Strength, 91 J. Personality & Soc. Psychol. 524, 525 (2006). In a series of experiments, Muraven and his colleagues showed that when participants in an experiment thought that there would be an upcoming task that would require self-control, they changed their behavior on an earlier task to conserve their self-control. 190 Id. Additionally, participants who had already used some self-control were more concerned with conserving their self-control strength on an intervening task than others who had not done so. 191 Id. at 527. The researchers concluded that “forethought and desire to exert self-control can lead to a breakdown of self-control in certain circumstances.” 192 Id. at 536.

Individuals are more passive after they have used up their regulatory strength, suggesting that there is a finite amount of volitional energy available. Regardless of whether a subsequent task requires active intervention to continue or to desist, individuals are more likely to choose the passive response. 193Baumeister et al., supra note 168, at 1260–61. Individuals whose self-control was depleted were more passive in their responses to movie watching; they watched longer when continuing was passive than when continuing was active. Individuals without this depletion watched the movie for the same length of time, regardless of whether the stopping mechanism was active or passive. Id. So, for example, one study demonstrated that individuals who had engaged in an initial series of choice-making behaviors were more likely, in a second stage of the experiment, to passively watch a broken videotape that showed no discernible picture than to alert the experimenter to the alleged problem. 194Kathleen D. Vohs et al., Making Choices Impairs Subsequent Self-Control: A Limited-Resource Account of Decision Making, Self-Regulation, and Active Initiative, 94 J. Personality & Soc. Psychol. 883, 894–95 (2008).

Recent research has also sought to understand the underlying source of self-regulatory energy. Grounding the self-regulatory function firmly in human physiology, a novel set of experiments looked to blood glucose levels after self-control tasks to determine whether the self-regulatory resource had its origins there. 195Matthew T. Gailliot et al., Self-Control Relies on Glucose as a Limited Energy Source: Willpower Is More than a Metaphor, 92 J. Personality & Soc. Psychol. 325 (2007). This series of studies demonstrated that self-control exertion depletes blood glucose levels and also suggested that increasing blood glucose levels caused a significant change for the better in individuals’ abilities to regulate their behavior after an initial self-control task. 196 Id. at 333. Similarly, researchers have suggested that sleep and rest, too, have a significant effect on the self-regulatory resource. 197 Id. at 334. In sum, the strength model of self-control suggests that there are many demands placed on the resource that individuals also use for controlling their behavior.

The self-control resource appears remarkably difficult to alter as a baseline matter, but there are a number of interventions that can boost performance on a subsequent self-control task after an initial task has depleted the self-control resource. The strength model of self-control research has suggested that, if self-control is like a muscle, perhaps it could be strengthened over time through practice. 198Mark Muraven et al., Longitudinal Improvement of Self-Regulation Through Practice: Building Self-Control Strength Through Repeated Exercise, 139 J. Soc. Psychol. 446, 448 (1999). Researchers posited that strength could be improved in two distinct ways: power (a stronger baseline capacity) and stamina (longer perseverance before fatigue). 199Id. In a short longitudinal study, Muraven and colleagues tested self-control at one point in time, sent participants home with instructions to perform a set of exercises in self-control, and then measured self-control again two weeks later. 200Id. at 451–52. By measuring baseline capacity at the outset and two weeks later, as well as measuring the drop-off in self-control performance between an initial and a subsequent self-control task, the researchers were able to test whether self-control had improved over time and whether any improvement was in the baseline capacity or in the resistance to fatigue. Muraven’s results suggest that self-regulatory exercises during the two-week period did yield an increase in self-regulatory strength—but not in brute power because initial measures were similar. 201Id. 453–56. Instead, individuals who had practiced self-control during the two-week period were better able to continue to exert self-control in a subsequent task in the later session than they had been initially. 202 Id. at 451–52. As noted above, increasing glucose levels after an initial self-control task also raised the level of self-control in a subsequent task, but it had no effect on baseline self-control performance on “nondepleted” individuals. 203Gailliot et al., supra note 195, at 330–33. Similarly, recent research has suggested that self-affirmation—an event that bolsters the integrity of the self—also boosts self-control after initial depletion but has no effect on baseline self-control. 204Brandon J. Schmeichel & Kathleen Vohs, Self-Affirmation and Self-Control: Affirming Core Values Counteracts Ego Depletion, 96 J. Personality & Soc. Psychol. 770, 770–79 (2009). So, too, efforts to increase self-regulatory strength through monetary incentives 205Mark Muraven & Elisaveta Slessareva, Mechanisms of Self-Control Failure: Motivation and Limited Resources, 29 Personality & Soc. Psychol. Bull. 894, 903 (2003). or through boosting positive mood 206Dianne M. Tice et al., Restoring the Self: Positive Affect Helps Improve Self-Regulation Following Ego Depletion, 43 J. Experimental Soc. Psychol. 379, 379, 383–84 (2007). have failed to have an effect on baseline self-control performance.

Of particular note for criminal law, self-control depletion has been found to have an effect on efforts to control aggressive tendencies. Participants in one study behaved more aggressively after a prior, unrelated self-control task than those who had not been taxed in the same way. 207C. Nathan DeWall et al., Violence Restrained: Effects of Self-Regulation and Its Depletion on Aggression, 43 J. Experimental Soc. Psych. 62 (2007). Specifically, participants who restrained themselves from eating a donut were later more aggressive in an unrelated setting; in addition, participants who exerted control over their attention by watching a video while ignoring words on the bottom of the screen or who completed a different mentally challenging task were later more aggressive in response to an insult. 208 Id. at 64–68; see also David Gal & Wendy Liu, Grapes of Wrath: The Angry Effects of Self-Control, 38 J. Consumer Res. 445 (2011) (finding that the exertion of self-control correlates with subsequent angry behavior in a variety of forms). Researchers have suggested that what they call “angry rumination,” where someone spends time reviewing an event that made them angry and “rehearsing acts of revenge,” also depletes self-control resources and increases the likelihood of self-control failure. 209Thomas F. Denson, Angry Rumination and the Self-Regulation of Aggression, in Psychology of Self-Regulation: Cognitive, Affective, and Motivational Processes 233, 234–35 (Joseph P. Forgas et al. eds., 2009).

B. Implications of the Strength Model in Criminal Law

If we take seriously the idea that self-control is a finite resource, this raises several intriguing questions about the “out of control” actor. If factors including the previous exercise of self-control, cognitive stress, and emotional disturbance can deplete this resource, an actor under such conditions is truly at a disadvantage when trying to self-regulate. This insight is captured by the laws that authorize lesser culpability with respect to crimes committed in the heat of passion or in the face of provocation, but it is rejected by the laws that deny consideration of extreme emotional disturbance or other reasons for diminished capacity. If all of these factors affect the self-control resource similarly, this calls for a much greater exploration of the underlying reasons for excusing self-control failures in some depleted instances and not in others.

For utilitarians, the strength model of self-regulation suggests that rather than requiring an increase in sanctions to get that actor to recalibrate his cost–benefit analysis, we may truly feel that there is no sanction that could make him act in accordance with the law because he simply lacks the resources to do so. For that reason, a strength model of self-regulation may suggest to the utilitarian that no sanctions are necessary when the self-control resource has been depleted. 210 But see Bradley R.E. Wright et al., Does the Perceived Risk of Punishment Deter Criminally Prone Individuals? Rational Choice, Self-Control, and Crime, 41 J. Res. Crime & Delinq. 180, 205 (2004) (analyzing data to suggest that the threat of punishment is most important for those who are low in self-control and high in self-reported criminal propensity). There are two countervailing arguments that may rescue sanctions for the utilitarian. First, if self-regulatory strength is a resource that can be self-consciously managed or conserved, then deterrence still plays a crucial role: if an individual is aware that sanctions exist that will make his exercise of self-control critical in the future, then he ought to conserve his strength for those situations. Second, if we believe that self-regulatory strength may be increased over time with practice, the existence of sanctions for uncontrolled actions may provide an incentive for the actor to practice and strengthen his self-control.

For the retributivist, too, the resource model of self-control tracks with the notion that there are occasions when we feel that an individual actor who is “out of control” already had those self-control resources severely taxed by other conditions in a way that ought to lessen liability. Additionally, the idea that the individual may simply have low self-regulatory strength 211 See DeWall et al., supra note 207. may mean that we should consider this an innate defect, just as insanity or diminished capacity may be. And, indeed, if control is a matter of individual differences, where do those differences come from, and do individuals have any meaningful control over their traits? 212 See Arenella, supra note 25; see also David Dolinko, Three Mistakes of Retributivism, 39 UCLA L. Rev. 1623, 1655 (1992) (“The alacrity with which even thoughtful and sophisticated retributivists such as Morse, Morris, and Dressler dismiss suggestions that economic, social, cultural, or psychological deprivations might excuse or mitigate criminal conduct suggests that the harshly punitive attitudes of legal actors and the public that were noted above are not misuses of retributivism but its logical outgrowth.” (footnote omitted)). This leads us back to the most basic questions about free will and its role in law. As Robert Weisberg says, this issue “seems to torture our best thinkers,” 213Robert Weisberg, The Values of Interdisciplinarity in Homicide Law Reform, 43 U. Mich. J.L. Reform 53, 75 (2009). and it is without a doubt outside the scope of my project here. But the strength model adds an extra step in thinking about this issue: because individuals can improve their self-regulatory function over time, we may consider them blameworthy when they have not done so. While it is still true that some may lack the desire to improve, this potential for improvement offers us the possibility that, with practice, anyone could have learned to do better.

V. Broader Implications

A. Doctrinal Implications

If one were to take seriously the psychological research on self-control in the context of criminal behavior, how might one think about specific crimes differently? Because criminal law spans so many different domains, it is worth thinking about the implications in a handful of contexts. One could imagine, for example, that the implications for self-control research might differ between violent crime and white collar crime. I explore only a few domains in criminal law below; the discussion does not seek to exhaust the many areas of criminal liability but merely to show how the psychology research in self-control plays out differently in several areas.

Because certain types of violent crime may often be a product of strong emotion, the interaction between emotion and self-control is important in this context. As noted above, an individual’s capacity for self-control is diminished in the face of emotional distress. Research suggests that this is because an individual’s executive function capacity is too busy working to improve affect. Sometimes, this means that self-control in other areas will suffer as individuals make decisions that they believe will improve their mood. Here, it may be that the very behavior that society would like individuals to control, the natural consequence of their violent impulse, is directly connected to the negative emotion they are experiencing and that the action they believe will ameliorate their negative emotional state is the act of violence itself. If affect regulation does trump self-control, some violent crime may be explained as an effort to improve mood.

In terms of how violent crime relates to the strength model of self-control, it seems that individuals who had other demands on their executive function resources would be more likely to engage in violent behavior—in other words, more likely to give in to the behavior suggested by a violent impulse and less likely to be able to exert control over their behavior and refrain from acting on the impulse. This vision of self-control tracks well with the law’s attempt to carve out defenses for more impulsive and less premeditated criminal behavior. But the law distinguishes between the various demands on our cognitive resources in a way that the psychology research does not. That is, a taxonomy of ways to exhaust the cognitive resource that self-control depends upon is not freighted, in psychology, with any meaning: each way in which the resource can be depleted is merely another in a list of potential depleting factors. But in law, these factors are evaluated based on different criteria, including whether the factors were, themselves, self-induced, and may also be evaluated differently depending on whether one takes a retributivist or a utilitarian tack.

In addition, those individuals who are low in self-regulatory strength as a general matter will also be more likely to engage in violent behavior. The law’s use of the reasonable or ordinary person standard implies that there may be a level of self-regulatory strength that we consider “reasonable.” And yet empirical work has never measured the population in order to identify such a level. Perhaps this is not a problem; if there is only a small percentage of the population who violate the standard, we may assume that we have struck the right point on the spectrum. But the self-regulatory resource model suggests that more work on identifying further parameters of this strength would be useful.

Interestingly, if one critical aspect of self-control is identifying the appropriate norm, 214 See supra notes 134–41 and accompanying text. it may be that certain violent crimes represent a clash of norms. Consider a cultural norm of self-defense: certain cultures suggest that the appropriate response to a threat or a provocation would be to defend oneself or defend one’s honor through attack, whereas other cultures might suggest that the appropriate response would be to seek help or to leave the scene. 215For discussions of use of the “cultural defense,” in which defendants in the United States, typically from another country, attempt to introduce evidence of their culture of origin as the basis for lack of culpability, see James J. Sing, Note, Culture as Sameness: Toward a Synthetic View of Provocation and Culture in the Criminal Law, 108 Yale L.J. 1845 (1999); and Sharon M. Tomao, Note, The Cultural Defense: Traditional or Formal?, 10 Geo. Immigr. L.J. 241 (1996). Using the self-control feedback loop, someone who has identified the self-defense or the substantial provocation norm may then conform his or her behavior to the wrong (for society) norm. Some theorists, however, believe that this type of situation represents an acquiescence to impulse and an abandonment of self-regulation; someone may believe it is “appropriate, reasonable, or even desirable to abandon self-control.” 216Baumeister & Heatherton, supra note 128, at 9. These scholars suggest that culture plays a significant role in determining when individuals choose to lose control over their behavior. Misregulation also suggests that people may simply be wrong about their calculations of costs and benefits of certain behavior. As Baumeister has explained, self-control failure may occur because “[p]eople . . . are trying to control the wrong aspect of the process or because they are trying to control something that is essentially immune to control.” 217 Id. at 11.

Premeditated crime suggests a different role for self-control research. Premeditated crime draws attention to the failure of research in psychology to fully explore behaviors in individuals that comport with those individuals’ goals, even if those goals violate society’s norms. But a foundational aspect of self-control theory in psychology may provide a useful frame for this problem, as well: the feedback loop that researchers have posited as crucial in appropriate self-monitoring requires, initially, the identification of an appropriate standard to test behavior against. When someone identifies a standard that requires him to engage in behavior that involves planned violent crime, he has chosen wrongly—but it does not mean that he is not engaged in effective self-regulation. In this case, the psychology of self-control suggests that the criminal actor has identified a (wrong) standard, but that he is effective in making his behavior match his goals. Alternatively, the psychology of self-control may suggest that the criminal actor has been irrational in his weighing of the pros and cons of his action because of temporal construal. In either case, however, the premeditated criminal actor, if he is aware of an alternative standard for his conduct or has overweighted the concrete gains of the action against the more abstract harms, could be held to a higher standard because his actions are fully self-regulated. In this situation, legal doctrine tracks the insights of psychology by penalizing premeditating criminals with a higher level of punishment.

White collar crime presents similar but perhaps even more complex problems vis-à-vis the psychology of self-control. Corporations are filled with highly educated individuals who have regularly delayed gratification in some domains in order to achieve longer term gains. They are, in Mischel’s basic terms, successful grown-ups. 218See infra text accompanying note 110. The research on self-control resource development would suggest that these are likely to be individuals who are highly practiced in self-control and who thus have a strong self-control resource that ought to be slower to be depleted. This would suggest, then, that these white collar crimes are less likely to represent a failure of self-control and more likely, as with premeditated violent crime, to represent a problem with identification of the appropriate norm by which to regulate one’s conduct. In a self-control feedback loop, individuals in corporations may be most concerned with controlling their behavior to comport with relevant office policies and directives from superiors, rather than with the law. Therefore, self-control may be quite high in corporate settings; it is just that it is used in some instances to conform one’s behavior to a corporate norm that encourages or even mandates criminal behavior. 219There is certainly evidence that corporate norms have a strong effect on individuals within the corporation. See, e.g., Tom R. Tyler & Steven L. Blader, Can Businesses Effectively Regulate Employee Conduct? The Antecedents of Rule Following in Work Settings, 48 Acad. Mgmt. J. 1143 (2005).

However, as with other premeditated crime, issues of temporal construal may change the way a potential wrongdoer’s rational cost–benefit calculus looks. The immediate, low-level benefits of wrongdoing such as altering financial records—pleasing one’s superiors, posting quarterly gains, raising stock prices—may be more appealing than the high-level construal of violating professional and legal rules. However, high-level construal of financial wrongdoing may include long-term career enhancement, in which case one could not suggest that such criminal activity is a self-control failure as psychology defines it. In some financial crimes, positive high-level construal might not only suggest that the crime is not a product of self-control failure but that it is actually a product of self-control.

If white collar crime represents a situation in which self-control is fully present, both utilitarian and retributive arguments would suggest a corresponding increase in punishment. The utilitarian would argue that the costs of the behavior must be raised in order to alter the very deliberate calculus about the criminal behavior, while a retributivist would suggest that the full nature of the self-control provides support for stronger penalties.

Drug-related crimes raise very different self-control issues. Crime that is committed while under the influence of drugs directly implicates self-control failures: drugs and alcohol have an immediate and powerful effect on the executive function’s capacity. Because drugs and alcohol impair cognitive functioning, they necessarily have the effect of diminishing self-control. 220For example, roughly one-quarter to one-third of prisoners report that the crime that they committed occurred while they were under the influence of drugs. See Office of Nat’l Drug Control Policy, Exec. Office of the President, ONDCP Drug Policy Information Clearinghouse Fact Sheet: Drug Data Summary (2003), available at http://www.expomed.com/drugtest/files/drugdata.pdf. As noted above, although psychology research—unlike law—does not make normative distinctions between types of depletion, the self-control strength model still provides some useful insight into this conduct. Even though the law may be willing to acknowledge and consider demands on self-control in determining culpability, the understanding that self-control is a strength that can be developed through practice and exercise, as well as a resource that can be conserved when necessary, helps make sense of the law’s typical lack of openness to defenses based on alcohol and drug use. Individuals who use or are addicted to alcohol or drugs are not developing their self-control strength; indeed, they are doing quite the opposite. And because someone has depleted his self-regulatory strength through his own choice of action, 221I leave to the side the free will issue implicated in addiction. See supra text accompanying notes 91–93. the law does not accommodate an effort to alter the degree of his culpability or punishment. 222For crime related to drug sales, rather than drug use, the picture is somewhat different. This crime, using a self-control lens, looks more like other types of premeditated crimes because the individual’s self-control has not been impaired by an outside influence. The first problem is a lack of the correct standard by which to monitor one’s actions; the second concern is with the problem of temporal construal and its distortion of present benefits to impair rational behavior.

B. Theoretical Implications

In contrast to legal theorists, who assume that individuals—including criminals—mostly can control themselves, even if they sometimes choose not to, psychologists writing about criminal behavior have concluded that “crimes practically define failure of self-control as it is usually conceived.” 223Hirschi, supra note 124, at 540. As noted above, there is a host of evidence that low self-control is related to negative outcomes, 224 See, e.g., Ozlem Ayduk et al., Regulating the Interpersonal Self: Strategic Self-Regulation for Coping with Rejection Sensitivity, 79 J. Personality & Soc. Psychol. 776, 777 (2000). and there is a compelling empirical case that poor self-control is a critical factor in criminal behavior. 225 See Wright et al., supra note 210. So, on the one hand, much of criminal law and criminal law scholarship assumes as a baseline that most criminal behavior is fully controlled behavior. Some smaller subset of criminal behavior is due to “noncontrol,” that is, what amounts to a conscious choice not to control oneself that is treated as identical to control itself. And some still smaller subset of criminal behavior is due to a “true” incapacity to exert self-control. In contrast, the psychological perspective assumes that most criminal behavior is a product of an individual’s inability to control his actions. 226 See, e.g., Hirschi, supra note 124, at 540; Kathleen D. Vohs & Roy F. Baumeister, Understanding Self-Regulation: An Introduction, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 1, 8.

This tremendous difference may be accounted for primarily by different conceptions of self-control failure between the disciplines of law and psychology; these differences help to highlight the essential features and implications of each vision. For legal scholars, failure of self-control is only exculpatory when we believe someone “truly” cannot exercise control; any other behavior is assumed to be a product of self-control. That is, when someone simply does not, rather than cannot, control his behavior, the law suggests that control continues to play a central role because that individual is, at some level, in control of his decision to abandon control. Therefore, lack of self-control only matters when we believe that the lack is compulsory, rather than optional. But literature in psychology (and, relatedly, in criminology 227Criminology focuses on “what actually causes crime and on how best to control it . . . .” Stuart A. Scheingold, Constructing the New Political Criminology: Power, Authority, and the Post-Liberal State, 23 Law & Soc. Inquiry 857, 859 (1998). The scope of criminology goes well beyond concerns about self-control, but it is a well-worn trope in the criminology literature that self-control failure plays a significant role in producing criminal behavior. See, e.g., Gottfredson & Hirschi, supra note 124, at 89–91. ) suggests that all crime occurs because someone either could not or did not exercise control, and furthermore suggests that the distinction between “could not” and “would not” is immaterial. Indeed, psychological research challenges the very existence of this divide, largely by ignoring it. Any time someone violates societal norms, the psychological perspective stamps this a failure of self-control.

This lack of nuance in the psychological literature calls out for attention. When psychologists assert that all crime is a product of self-control failure, this embeds an assumed set of shared norms that is simply not accurate. Research that appreciates and examines differences in behavior between those who share and do not share particular norms is critical to the development of deeper and more sophisticated connections between law and psychology.

While the conceptual differences in “self-control” between the disciplines of law and psychology may play some role in accounting for complete lack of critical examination by legal scholars of empirical research in social psychology on the mechanisms of self-control, these distinctions may themselves serve as a useful catalyst for discussion about the role of self-control in criminal law. While social psychology focuses not on the roots of self-control failure writ large but rather on the causes of self-control failure writ small, so too legal doctrine makes the same leap: questions of free will, writ large, are outside the scope of what the criminal law will consider, 228 See, e.g., State v. Sikora, 210 A.2d 193, 203 (N.J. 1965) (“[W]e cannot accept a thesis that responsibility in law for a criminal act perpetrated by a legally sane defendant, can be considered nonexistent or measured by the punishment established for a crime of lower degree, because his act was motivated by subconscious influences of which he was not aware, and which stemmed inevitably from his individual personality structure.”). but purportedly smaller questions of individual control over particular acts remain of deep concern. 229 See, e.g., Zamora v. Phillips, No. 04-CV-4093 (JFB), 2006 WL 2265079, at *6 (E.D.N.Y. Aug. 8, 2006) (“In determining whether a petitioner has acted out of a loss of self control, the court will look at the petitioner’s conduct before and after the homicide.”). And yet social psychology casts a far wider net over self-control failure. Even narrowing the scope of self-control failure to situations where an individual truly does want to comport with the relevant societal norm, the temporal construal problems that inhere even in efforts to act rationally and the vision of self-control as a resource that is fairly easily depleted both suggest that self-control failures in psychology are far more common than the failures of self-control that the law is willing to recognize and allow.

In particular, the contrast between what legal scholars and psychologists conceptualize as self-control failure points up quite starkly that the law makes normative determinations about when a loss of self-control is meaningful and when it is not. 230Others have explicitly noted the normative nature of criminal law efforts around doctrine that centers on questions of self-control. See Stephen J. Morse, The Irreducibly Normative Nature of Provocation/Passion, 43 U. Mich. J.L. Reform 193, 205 (2009). For example, as noted above, the law does not permit lesser punishments when an individual fails to exercise self-control after drinking or using drugs, even though the individual’s acts might be no less out of his control than the acts might be in the face of provocation or heat of passion. While psychology makes no special distinctions between the types of demands placed on one’s self-control strength, the law draws many fine lines that do not, in fact, delineate control or lack thereof as they purport to do but instead reflect a normative judgment about the type of behavior involved.

Conclusion

Considering crime as a problem bound up with issues of self-control—a premise central to criminal law theory and doctrine—suggests that legal scholars and lawyers should take the psychology of self-control seriously. But caveats are certainly in order. One important problem is highlighted in the above discussion about temporal construal. Typically, self-control literature posits self-control problems as those in which an individual must either resist engaging in an appealing but undesirable activity, like smoking or eating unhealthy food, or must persist in an unappealing but desirable activity, such as exercising or maintaining a budget. Yet the psychology literature does not typically differentiate between subjective and objective norms, so that discussions of self-control behaviors often center around societally undesirable—or desirable—activities without any test of the individual perspective on those activities. Thus an “overeater” is perceived as having a self-control problem whether or not he himself considers overeating to be a problem; a smoker who does not view smoking or its potential consequences as problematic is nonetheless considered to suffer a self-regulation failure. 231Thus, some researchers have described self-control from a motivational perspective: “If the person does not have a strong inner desire to reach [a particular] standard, then merely knowing the standard is not likely to produce effective self-regulation.” Joseph P. Forgas et al., The Psychology of Self-Regulation: An Introductory Review, in Psychology of Self-Regulation: Cognitive, Affective, and Motivational Processes, supra note 209, at 1, 11. In the case of criminal behavior, a typical self-control failure as understood by psychology would be when an individual adopted the objective norm of criminalized behavior and understood that the behavior was undesirable 232Even if the temporal construal issues made this calculation tip in favor of the criminal activity, that calculation represents a different situation from one in which the actor truly believes that the action is completely appropriate, per some internal standard. but was not able to control his behavior in order to comport with the law. The self-control literature in psychology does not offer a perfect paradigm to capture the behavior of the individual who adopts—and adheres to—a subjective norm that violates the law. Another concern for application to law is the relative paucity of psychology research on how, exactly, self-control can be developed. However, as the self-control research continues to grow, insights from the law, and questions that arise from seriously considering this research in light of our legal system, should be taken more seriously in helping to develop appropriate research questions for investigation.

Nonetheless, important principles emerge in considering psychological research on self-control in the criminal law context. Research findings that self-control is better when individuals can focus on high-level abstract concepts about actions can help us think about what kinds of systems could make such focus possible, but we must acknowledge that not everyone’s abstract concepts about actions will, themselves, comport with the requirements of criminal law. Similarly, findings that self-control is a finite resource may help us to make better sense of the places in our criminal law system where we hold an individual less culpable because of other factors that have reasonably depleted that resource, and also may help us to be more aware of the normative nature of distinctions among ways in which the resource was depleted. Self-control research in psychology also offers a suggestion for both prevention and rehabilitation: if self-control is a resource that can be strengthened over time with practice, perhaps this is a fruitful area for further development in our at-risk populations. Considering the major findings of social psychology research on self-control as they relate to criminal law both sheds light on, and poses further questions for, our conceptualizations of self-control as a guiding principle in crime and punishment.

Footnotes

Associate Professor, Washington University School of Law. Thanks to Susan Appleton, Kimberly Bailey, Kenworthey Bilz, Matt Bodie, Sam Buell, Emily Hughes, Peter Joy, Dan Markel, Alexandra Natapoff, Mae Quinn, Laura Rosenbury, David Zaring, and participants in the Washington University Regional Junior Faculty Workshop and the Junior Criminal Law Scholars Workshop for helpful feedback and suggestions. Thanks also to Elizabeth Chen for terrific research assistance.

1 See Morissette v. United States, 342 U.S. 246, 250 (1952) (noting the “universal and persistent . . . belief in freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil”); Smith v. Armontrout, 865 F.2d 1502, 1506 (8th Cir. 1988) (en banc) (“The whole presupposition of the criminal law is that most people, most of the time, have free will within broad limits. They are capable of conforming their actions to the requirements of the law, and of appreciating the consequences of failing to do so.”); Nita A. Farahany & James E. Coleman, Jr., Genetics, Neuroscience, and Criminal Responsibility, in The Impact of Behavioral Sciences on Criminal Law 183, 20708 (Nita A. Farahany ed., 2009) (explaining that there is a fundamental difference between theoretical free will and criminal law free will: “the criminal law presumes that individuals actively and consciously choose to engage in criminal conduct”). Indeed, modern philosophers and social scientists have suggested that an assumption of free will is critical to the successful functioning of society, regardless of whether or not it is true. See, e.g., John Tierney, Do You Have Free Will? Yes, It’s the Only Choice, N.Y. Times, Mar. 22, 2011, at D1.

2 See Wayne R. LaFave, Criminal Law § 5.1, at 253 (5th ed. 2010) (noting that crime typically requires an act or omission as well as requisite state of mind). But see id. § 5.5, at 288–89 (detailing several types of strict liability criminal statutes).

3Only one article has begun to consider the social psychological research on self-control, in the narrow context of the provocation defense in British law. See Richard Holton & Stephen Shute, Self-Control in the Modern Provocation Defence, 27 O.J.L.S. 49 (2007).

4In a project similar in type but different in focus, Steven Sherman and Joseph Hoffman offered a careful exploration of the implications of psychological research on anger for voluntary manslaughter and the “heat of passion” defense. Steven J. Sherman & Joseph L. Hoffmann, The Psychology and Law of Voluntary Manslaughter: What Can Psychology Research Teach Us About the “Heat of Passion” Defense?, 20 J. Behav. Decision Making 499 (2007).

5For discussion of this debate, see Farahany & Coleman, supra note 1, at 206–07; and Owen D. Jones, Behavioral Genetics and Crime, in Context, in The Impact of Behavioral Sciences on Criminal Law, supra note 1, at 125, 138.

6 See, e.g., Stephen J. Morse, Culpability and Control, 142 U. Pa. L. Rev. 1587, 1587–88 (1994). In addition, sociological theories largely take a more deterministic approach, suggesting that “delinquency is an adaptation to conditions and social influences in lower-class environments.” Deborah W. Denno, Sociological and Human Developmental Explanations of Crime: Conflict or Consensus?, 23 Criminology 711, 712 (1985).

7 See Richard A. Posner, Economic Analysis of Law § 7.2, at 242 (5th ed. 1998); Steven Shavell, Foundations of Economic Analysis of Law 552–56 (2004); Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. Pol. Econ. 169 (1968).

8These are not the only lenses through which scholars have analyzed criminal law, and I do not mean to suggest either (1) that they represent an exhaustive list of perspectives on criminal law or (2) that all criminal law scholars writing from the standpoint of neuroscience, philosophy, or economics share a common outlook. I mean only to indicate that several prominent trends in criminal law scholarship share a set of assumptions about the centrality (if not the meaning) of self-control.

9For further discussion of what psychologists mean by acting in a controlled or uncontrolled manner, see infra Part II.

10Again, although the divisions I have drawn are somewhat rough, and criminal scholars’ work may often overlap or even defy these categories, the distinctions are nonetheless useful as a basic framework for discussion.

11 See, e.g., Crook v. State, 813 So. 2d 68, 75 (Fla. 2002) (per curiam) (“[T]he existence of brain damage is a significant mitigating factor that trial courts should consider in deciding whether a death sentence is appropriate in a particular case.”); Tyler v. State, 903 N.E.2d 463, 469 (Ind. 2009) (rejecting the trial court’s imposition of an enhanced sentence in light of the presence of mitigating factors, including “a brain tumor . . . that affected [the defendant’s] ability to control his behavior”).

12 See, e.g., Henry T. Greely, Neuroscience and Criminal Justice: Not Responsibility but Treatment, 56 U. Kan. L. Rev. 1103, 1104 (2008) (“[N]euroscience is likely to produce a number of different types of interventions—surgical, pharmacological, and other—that will be said to prevent, or to lessen the risk of, continuing criminal conduct.”). But see John F. Stinneford, Incapacitation Through Maiming: Chemical Castration, the Eighth Amendment, and the Denial of Human Dignity, 3 U. St. Thomas L.J. 559, 568 (2006) (arguing against pharmacological punishment for sex offenders).

13Jones, supra note 5, at 142.

14 See, e.g., Michele Cotton, A Foolish Consistency: Keeping Determinism Out of the Criminal Law, 15 B.U. Pub. Int. L.J. 1, 44 (2005); Richard E. Redding, The Brain-Disordered Defendant: Neuroscience and Legal Insanity in the Twenty-First Century, 56 Am. U. L. Rev. 51, 118 (2006); Christopher Slobogin, The Civilization of the Criminal Law, 58 Vand. L. Rev. 121, 165 (2005).

15Deborah Denno notes that, although “[b]ehavioral genetics evidence has not gained widespread acceptance in current case law,” “[a]t least 48 criminal cases have referred to behavioral genetics evidence” in a thirteen-year period between 1994 and 2007. Deborah W. Denno, Behavioral Genetics Evidence in Criminal Cases: 1997–2007, in The Impact of Behavioral Sciences on Criminal Law, supra note 1, at 317, 330–31.

16 See Farahany & Coleman, supra note 1, at 187–205; see also Mobley v. State, 455 S.E.2d 61, 65–66 (Ga. 1995) (rejecting consideration of the defendant’s genetic propensity for criminal behavior).

17Farahany & Coleman, supra note 1, at 185.

18Kent Greenawalt, Punishment, in 4 Encyclopedia of Crime and Justice 1336, 1338 (1983). Recent research has added social science data into the discussion of how we determine commensurate punishment—that is, how we understand desert. See, e.g., Paul H. Robinson & John M. Darley, Intuitions of Justice: Implications for Criminal Law and Justice Policy, 81 S. Cal. L. Rev. 1 (2007) (discussing and analyzing research relating to desert).

19John Rawls, Two Concepts of Rules, 64 Phil. Rev. 3, 4–5 (1955).

20Russell L. Christopher, Deterring Retributivism: The Injustice of “Just” Punishment, 96 Nw. U. L. Rev. 843, 860 (2002) (footnotes omitted).

21 See Greenawalt, supra note 18, at 1338.

22Kant famously claimed that a society about to disband should still execute its murderers on principle. Immanuel Kant, The Philosophy of Law: An Exposition of the Fundamental Principles of Jurisprudence as the Science of Right 198 (Augustus M. Kelley Publishers 1974) (1887).

23 See Greenawalt, supra note 18, at 1339.

24David E. Cooper, Hegel’s Theory of Punishment, in Hegel’s Political Philosophy 151, 153 (Z.A. Pelczynski 1971).

25Peter Arenella, Convicting the Morally Blameless: Reassessing the Relationship Between Legal and Moral Accountability, 39 UCLA L. Rev. 1511, 1518 (1992).

26Morse, supra note 6, at 1587–88.

27Stephen J. Morse, Excusing the Crazy: The Insanity Defense Reconsidered, 58 S. Cal. L. Rev. 777, 781 (1985).

28 Id. at 782.

29Morse, supra note 6, at 1595.

30 Id.

31Arenella, supra note 25, at 1584–85.

32 Posner, supra note 7, § 7.2, at 243.

33 Id.

34 See Cesare Beccaria, On Crimes and Punishments and Other Writings (Richard Bellamy ed., Richard Davies trans., Cambridge Univ. Press 1995) (1764); Jeremy Bentham, The Principles of Morals and Legislation (Prometheus Books 1988) (1789); H.L.A. Hart, Punishment and Responsibility (2d ed. 2008).

35 Posner, supra note 7, § 7.2, at 243.

36 See, e.g., Richard A. Bierschbach & Alex Stein, Mediating Rules in Criminal Law, 93 Va. L. Rev. 1197 (2007); Samuel W. Buell, The Upside of Overbreadth, 83 N.Y.U. L. Rev. 1491 (2008); Steven Shavell, Criminal Law and the Optimal Use of Nonmonetary Sanctions as a Deterrent, 85 Colum. L. Rev. 1232 (1985); Avraham D. Tabbach, Does a Rise in Maximal Fines Increase or Decrease the Optimal Level of Deterrence?, 5 Rev. L. & Econ. 53 (2009).

37Although I have described the retributivist and utilitarian attitudes toward human control over one’s own actions as similar here, Joshua Dressler contrasts them, stating that “the premise of utilitarianism is that people are generally hedonistic and rational calculators,” whereas “retributivists focus on their view that humans generally possess free will or free choice and, therefore, may justly be blamed when they choose to violate society’s mores.” Joshua Dressler, Understanding Criminal Law 16 (5th ed. 2009) (footnote omitted). Although Dressler’s description may be accurate, both perspectives do rely on an understanding of an individual who is able to choose.

38 See Posner, supra note 7, § 7.2, at 242–43 (arguing that a starving man ought to steal a trivial amount of food from an unoccupied cabin in the woods).

39 See id. § 7.3, at 250–51 (arguing that the theory of criminal sanction is one of deterrence).

40 Id. § 7.4, at 257. On the other hand, Posner suggests that, because the impulsive criminal is less deterrable, perhaps society ought to “buy” less punishment for these individuals. Id.

41 See, e.g., Shavell, supra note 7, at 561–62. Of course, generally speaking, a deterrence system has no particular interest in punishing any one criminal. Instead, “people are actually fined or imprisoned only to maintain the credibility of the deterrent.” Posner, supra note 7, § 7.3, at 250.

42However, the system may need to impose punishment to remain credible for future offenders, and there still may be an independent need to imprison undeterrable criminals for the purpose of incapacitation. Shavell, supra note 7, at 562. As Shavell says:[I]n many cases where a person is not legally responsible for his act, he will be no less dangerous to society than if he were responsible. A person who has an uncontrollable urge to set fires, or who is subject to insane, violent rages, is dangerous to society even though he cannot help himself.Id.

43 Dressler, supra note 37, § 9.02, at 90.

44 Id. § 23.01, at 303–04.

45 Id. § 10.07, at 139, § 11.01, at 145–46.

46 H.L.A. Hart, Punishment and the Elimination of Responsibility, in Punishment and Responsibility, supra note 34, at 181–82.

47 Id. at 182.

48 See Dressler, supra note 37, § 25.03, at 346 (noting that just punishment depends on the rationality and self-control of the criminal who performed the crime).

49 Roscoe Pound, Criminal Justice in America 126–27 (1930).

50 Dressler, supra note 37, § 25.03, at 345.

51 Id. at 346.

52 Id.

53 Id. (footnote omitted).

54 Id.; accord Morse, supra note 27, at 783.

55M‘Naghten’s Case, (1843) 8 Eng. Rep. 718 (H.L.) 722; 10 Clark & Finnelly 200.

56 See Dressler, supra note 37, § 25.04, at 352.

57 Id.

58 See id. at 353 (citing Clark v. Arizona, 548 U.S. 735, 751 (2006)).

59 Id. (quoting Commonwealth v. Rogers, 48 Mass. (7 Met.) 500, 502 (1844)) (internal quotation marks omitted).

60 Id. (emphasis omitted) (quoting Parsons v. State, 2 So. 854, 866 (Ala. 1887)) (internal quotation mark omitted).

61 Id. (citing Davis v. United States, 165 U.S. 373, 378 (1897)).

62 Id.

63 See Morse, supra note 27.

64 Dressler, supra note 37, § 25.04, at 354 (citing United States v. Lyons, 731 F.2d 243, 248 (5th Cir. 1984) (en banc)).

65 Model Penal Code § 4.01(1) (1985) (alteration in original).

66 See Dressler, supra note 37, at 356.

67The product test, also called the Durham test, stated that whenever a person’s behavior is the product of a mental disease or defect, the person should not be held liable by reason of insanity. Durham v. United States, 214 F.2d 862 (D.C. Cir. 1954), overruled by United States v. Brawner, 471 F.2d 969 (D.C. Cir. 1972) (en banc), superseded by statute, Insanity Defense Reform Act of 1984, Pub. L. No. 98-473, 98 Stat. 2057 (codified as amended at 18 U.S.C. §§ 17, 4241–4247 (2006)), as recognized in Shannon v. United States, 512 U.S. 573 (1994). This test has been abandoned. See Brawner, 471 F.2d at 973 (adopting the Model Penal Code test). Its purpose was to streamline the test for insanity in order to give mental health professionals freer rein to testify about a defendant’s insanity without parsing the defendant’s volitional or cognitive abilities. See Dressler, supra 37, at 355.

68For example, Pennsylvania courts expressly include lack of self-control in their definition of what insanity is not: “Certainly neither social maladjustment, nor lack of self-control, nor impulsiveness, nor psychoneurosis, nor emotional instability, nor chronic malaria, nor all of such conditions combined, constitute insanity within the criminal-law conception of that term.” Commonwealth v. Neill, 67 A.2d 276, 280 (Pa. 1949).

69 See, e.g., Paul H. Robinson, Criminal Law Defenses: A Systematic Analysis, 82 Colum. L. Rev. 199, 203 (1982).

70 See, e.g., Mitchell N. Berman, Justification and Excuse, Law and Morality, 53 Duke L.J. 1 (2003); Kent Greenawalt, Distinguishing Justifications from Excuses, 49 Law & Contemp. Probs. 89 (1986); Kent Greenawalt, The Perplexing Borders of Justification and Excuse, 84 Colum. L. Rev. 1897 (1984).

71State v. Guebara, 696 P.2d 381, 385 (Kan. 1985) (citing State v. Coop, 573 P.2d 1017, 1020 (Kan. 1978)).

72 Dressler, supra note 37, § 31.07, at 539.

73There are four basic elements to a provocation/heat of passion claim. First, the actor must act in the heat of passion; second, the passion must have resulted from adequate provocation; third, “the actor must not have had a reasonable opportunity to cool off”; and finally, “there must be a causal link between the provocation, the passion, and the homicide.” Id. at 535 (citing Girouard v. State, 583 A.2d 718, 721 (Md. 1991)).

74 See Posner, supra note 7, § 7.2, at 242.

75 See Joshua Dressler, Provocation: Explaining and Justifying the Defense in Partial Excuse, Loss of Self-Control Terms, in Criminal Law Conversations 319 (Paul H. Robinson et al. eds., 2009) (arguing that justification serves as a partial defense).

76Reid Griffith Fontaine, Adequate (Non)Provocation and Heat of Passion as Excuse Not Justification, 43 U. Mich. J.L. Reform 27, 45–46 (2009).

77 See Dressler, supra note 75.

78Model Penal Code § 210.3(1)(b) (1985).

79People v. Smith, 808 N.E.2d 333, 334 (N.Y. 2004) (quoting People v. Roche, 772 N.E.2d 1133, 1138 (N.Y. 2002)). “The defense requires proof of a subjective element, that defendant acted under an extreme emotional disturbance, and an objective element, that there was a reasonable explanation or excuse for the emotional disturbance.” Id. (citing People v. Moye, 489 N.E.2d 736, 738 (N.Y. 1985)).

80 See Mitchell N. Berman & Ian P. Farrell, Provocation Manslaughter as Partial Justification and Partial Excuse, 52 Wm. & Mary L. Rev. 1027, 1040 (2011). Only eleven states have adopted the extreme emotional or mental disturbance language of the Model Penal Code. See id. at 1044.

81 But see Dressler, supra note 37, at 326 (“Although it is true that self-induced intoxication as such never excuses wrongdoing, the condition that intoxication causes, e.g., a clouded mental state, unconsciousness, or insanity, may serve as an exculpatory basis in very limited circumstances.”).

82 See State v. Kolisnitschenko, 267 N.W.2d 321, 326 (Wis. 1978).

83Montana v. Egelhoff, 518 U.S. 37, 56 (1996) (plurality opinion).

84 See id.

85 Id. at 57 (Ginsburg, J., concurring).

86 See Dressler, supra note 37, § 24.02, at 348 (“The common law treats intoxication resulting from alcoholism or drug addiction as voluntary.”).

87 See, e.g., Paul H. Robinson, Causing the Conditions of One’s Own Defense: A Study in the Limits of Theory in Criminal Law Doctrine, 71 ).

88 See id. at 31–34.

89The widely popular “Just Say No” anti-drug ad campaign of the 1980s and 1990s is one example of this attitude toward substance abuse, suggesting that personal choice and responsibility are the main determinants of drug use. See, e.g., Jerome Beck, 100 Years of “Just Say No” Versus “Just Say Know”: Reevaluating Drug Education Goals for the Coming Century, 22 Evaluation Rev. 15 (1998).

90Courts have acknowledged that addiction can be considered a type of disease, which suggests that its consequences are beyond the momentary control of the individual; attempts to punish individuals for the status of addiction, rather than for the act of using drugs or for acts undertaken while under their influence, have not succeeded. See, e.g., Robinson v. California, 370 U.S. 660 (1962). Nonetheless, courts have largely been unwilling to exempt the taking of drugs, or acts undertaken under their influence, from criminal liability, even when addiction leads to these events. See, e.g., Powell v. Texas, 392 U.S. 514 (1968) (refusing to hold that acts committed while under the influence of alcohol, even if the defendant were an alcoholic, were exempted from criminal responsibility); Richard C. Boldt, The Construction of Responsibility in the Criminal Law, 140 U. Pa. L. Rev. 2245, 2309 (1992) (noting that, following the decision in Powell, most courts have refused to hold that acts committed while under the influence of alcohol or drugs were exempted from criminal liability, even if the defendant suffered from addiction).

91 See Stephen J. Morse, Addiction, Genetics, and Criminal Responsibility, 69 Law & Contemp. Probs. 165, 169–72 (2006) (describing a strand of modern addiction research as “treat[ing] the intentional conduct of the addict solely as the product of brain mechanisms”).

92 Id. at 187.

93 See Robinson, supra note 87, at 7–8.

94For example, federal sentencing guidelines allow for a downward departure in sentencing when individuals have “significantly reduced mental capacity” that might render them incapable of being fully responsible for their actions. See U.S. Sentencing Guidelines Manual § 5K2.13 (2011).

95 See id. § 5K2.12–.13 (providing for lesser punishment when the defendant is unable to control his actions due to coercion and duress, or diminished mental capacity).

96 See supra Part I.A.2.

97 See supra notes 40–42 and accompanying text. As previously noted, this is perhaps counterintuitive to the insight of the law and economics scholar who suggests that, because the impulsive person is harder to deter, he should face a greater sanction.

98 See Gregg v. Georgia, 428 U.S. 153, 193 (1976).

99 Id. However, mitigating and aggravating factors are not always easy for jurors to distinguish. See Joshua N. Sondheimer, Note, A Continuing Source of Aggravation: The Improper Consideration of Mitigating Factors in Death Penalty Sentencing, 41 Hastings L.J. 409, 410 (1990).

100Craig M. Cooley, Mapping the Monster’s Mental Health and Social History: Why Capital Defense Attorneys and Public Defender Death Penalty Units Require the Services of Mitigation Specialists, 30 Okla. City U. L. Rev. 23, 52 (2005).

101 Id. at 53–54.

102Emily Hughes, Mitigating Death, 18 Cornell J.L. & Pub. Pol’y 337, 345 (2009).

103Roy F. Baumeister, The Self, in 1 The Handbook of Social Psychology 680, 680 (Daniel T. Gilbert et al. eds., 4th ed. 1998).

104 Id.

105 See id. at 712. As Baumeister explains, it may be that, because so much of the executive function is not dramatic, instead consisting of many small decisions that are easily understood as guided by the rational pursuit of self-interest, it has been understudied. Accordingly, Baumeister notes, “[T]here has been something of a premium put on finding ways in which people act contrary to their rational self-interest”—hence the surge of interest in research on heuristics, biases, and other nonrational decision-making processes. Id. This research “reveal[s] flaws in the executive function” and for that reason helps us understand how the executive function works. Id. at 713. Throughout this Article, following the lead of scholars in the field, I use the terms self-control and self-regulation interchangeably.

106 Id. at 716.

107 Id. at 717.

108 Id.

109 Id.

110 See, e.g., Yuichi Shoda, Walter Mischel & Philip K. Peake, Predicting Adolescent Cognitive and Self-Regulatory Competencies from Preschool Delay of Gratification: Identifying Diagnostic Conditions, 26 Developmental Psychol. 978, 982 (1990) (finding that very young children’s capacity to delay their gratification in the face of a potential reward significantly correlated with higher SAT scores over a decade later).

111 See Nancy Eisenberg et al., Effortful Control: Relations with Emotion Regulation, Adjustment, and Socialization in Childhood, in Handbook of Self-Regulation: Research, Theory, and Applications 263, 263 (Kathleen D. Vohs & Roy F. Baumeister eds., 2d ed. 2011) (quoting Mary K. Rothbart & John E. Bates, Temperament, in 3 Handbook of Child Psychology: Social, Emotional, and Personality Development 99, 129 (William Damon et al. eds., 6th ed. 2006)).

112 Id. at 264.

113 Id.

114 Id. at 270.

115 Id. at 268.

116Grazyna Kochanska & Amy Knaack, Effortful Control as a Personality Characteristic of Young Children: Antecedents, Correlates, and Consequences, 71 J. Personality 1087, 1102 (2003) (finding that higher effortful control in small children predicted a more internalized conscience).

117 See, e.g., Ivanna K. Gurthrie et al., The Relations of Regulation and Emotionality to Children’s Situational Empathy-Related Responding, 21 Motivation & Emotion 87 (1997) (finding that children high in effortful control demonstrated more facial sadness while watching a sad movie than children low in effortful control).

118 See, e.g., Nancy Eisenberg et al., The Relations of Children’s Dispositional Prosocial Behavior to Emotionality, Regulation, and Social Functioning, 67 Child Dev. 974 (1996); Nancy Eisenberg et al., The Relations of Regulation and Emotionality to Resiliency and Competent Social Functioning in Elementary School Children, 68 Child Dev. 295 (1997).

119Avshalom Caspi & Phil A. Silva, Temperamental Qualities at Age Three Predict Personality Traits in Young Adulthood: Longitudinal Evidence from a Birth Cohort, 66 Child Dev. 486, 495 (1995).

120Bill Henry et al., Staying in School Protects Boys with Poor Self-Regulation in Childhood from Later Crime: A Longitudinal Study, 23 Int’l J. Behav. Dev. 1049, 1067 (1999).

121 See Lea Pulkkinen, Offensive and Defensive Aggression in Humans: A Longitudinal Perspective, 13 Aggressive Behav. 197, 203–04 (1987); Lea Pulkkinen, Self-Control and Continuity from Childhood to Late Adolescence, in 4 Life-Span Development and Behavior 63 (Paul B. Baltes & Orville G. Brim, Jr. eds., 1982).

122 See Jay G. Hull & Laurie B. Slone, Alcohol and Self-Regulation, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 466 (Roy F. Baumeister & Kathleen D. Vohs eds., 2004); Michael A. Sayette & Kasey M. Griffin, Self-Regulatory Failure and Addiction, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 505.

123 See Kathryn D. Scott et al., The Role of Alcohol in Physical Assault Perpetration and Victimization, 60 J. Stud. Alcohol 198 (1999) (finding a strong relationship between alcohol use and physical assault perpetration); Samantha Wells et al., Alcohol-Related Aggression in the General Population, 61 J. Stud. Alcohol 626 (2000) (finding that alcohol intoxication is associated with increased levels of physical aggression); Helene Raskin White et al., Problem Drinking and Intimate Partner Violence, 63 J. Stud. Alcohol 205 (2002) (finding that problem drinking is significantly associated with intimate partner violence perpetration).

124Travis Hirschi, Self-Control and Crime, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 537, 537; accord Michael R. Gottfredson & Travis Hirschi, A General Theory of Crime 88 (1990).

125 Gottfredson & Hirschi, supra note 124, at 88–91.

126 See Baumeister, supra note 103, at 717.

127Roy F. Baumeister, Yielding to Temptation: Self-Control Failure, Impulsive Purchasing, and Consumer Behavior, 28 J. Consumer Res. 670, 671 (2002).

128Roy F. Baumeister & Todd F. Heatherton, Self-Regulation Failure: An Overview, 7 Psychol. Inquiry 1, 2 (1996).

129 Id.; Baumeister, supra note 127, at 671.

130 See Baumeister & Heatherton, supra note 128.

131 See Janet Metcalfe & Walter Mischel, A Hot/Cool-System Analysis of Delay of Gratification: Dynamics of Willpower, 106 Psychol. Rev. 3 (1999).

132 See George Loewenstein, Out of Control: Visceral Influences on Behavior, 65 Organizational Behav. & Hum. Decision Processes 272, 288–89 (1996).

133 See Kentaro Fujita et al., Construal Levels and Self-Control, 90 J. Personality & Soc. Psychol. 351, 353–55 (2006).

134Baumeister & Heatherton, supra note 128, at 2.

135 Id.

136 See Robert A. Emmons & Laura A. King, Conflict Among Personal Strivings: Immediate and Long-Term Implications for Psychological and Physical Well-Being, 54 J. Personality & Soc. Psychol. 1040, 1045–47 (1988) (finding that conflict between personal striving and ambivalence results in negative psychological and health effects, which supports a connection between personal strivings and personality); Elizabeth Van Hook & E. Tory Higgins, Self-Related Problems Beyond the Self-Concept: Motivational Consequences of Discrepant Self-Guides, 55 J. Personality & Soc. Psychol. 625, 632 (1988) (finding that individuals who possess conflicting standards are more likely to experience psychological distress, which “emphasizes the importance of self-evaluative standards in self-guidance and self-assessment”).

137Todd F. Heatherton & Nalini Ambady, Self-Esteem, Self-Prediction, and Living Up to Commitments, in Self-Esteem: The Puzzle of Low Self-Regard 131 (Roy F. Baumeister ed., 1993).

138Baumeister & Heatherton, supra note 128, at 2.

139 Id.

140 Id. This understanding of self-control processes was strongly influenced by earlier work by Charles S. Carver and Michael F. Scheier that merged cybernetic theory with self-awareness theory to suggest that self-regulation operates in a “TOTE” feedback loop: Test how one is doing against a standard, Operate on the self to change behavior to better comport with the standard, Test again to see how performance now compares to the standard, and Exit when one has reached the appropriate level of behavior. See Charles S. Carver & Michael F. Scheier, Attention and Self-Regulation: A Control-Theory Approach to Human Behavior 16 (1981). Such a feedback loop could also work to effectuate a goal of increasing distance between oneself and a negative standard. See id. at 17. There may be important differences in the way that the self regulates toward and against standards, but this has been relatively underexplored empirically.

141Baumeister, supra note 103, at 717.

142 See, e.g., Kentaro Fujita, Seeing the Forest Beyond the Trees: A Construal-Level Approach to Self-Control, 2 Soc. & Personality Psychol. Compass 1475 (2008); Nira Liberman et al., The Effect of Temporal Distance on Level of Mental Construal, 38 J. Experimental Soc. Psychol. 523, 532 (2002).

143Yaacov Trope & Nira Liberman, Temporal Construal, 110 Psychol. Rev. 403, 405 (2003).

144 Id.

145 Id.

146Fujita et al., supra note 133, at 354.

147 Id. at 354–55.

148 Id. at 355.

149 Id. at 356.

150 Id. at 357.

151 Id. at 363.

152 See George Ainslie & Nick Haslam, Hyperbolic Discounting, in Choice over Time 57, 58 (George Loewenstein & Jon Elster eds., 1992); Antonio L. Freitas et al., Abstract and Concrete Self-Evaluative Goals, 80 J. Personality & Soc. Psychol. 410, 417 (2001).

153See David Laibson, Golden Eggs and Hyperbolic Discounting, 112 Q.J. Econ. 443, 445–46 (1997) (defining hyperbolic-discount functions as those characterized by a high discount rate over short horizons and a low discount rate over long horizons).

154 See, e.g., James Q. Wilson & Richard J. Herrnstein, Crime and Human Nature (1985) (suggesting that temporal discounting is important in explaining criminality because the benefits of crime occur at a different time than its costs); David S. Lee & Justin McCrary, Crime, Punishment, and Myopia 28 (Nat’l Bureau of Econ. Research, Working Paper No. 11491, 2005), available at http://www.nber.org/papers/w11491.pdf.

155Trope & Liberman, supra note 143, at 406.

156Cheryl Wakslak & Yaacov Trope, The Effect of Construal Level on Subjective Probability Estimates, 20 Psychol. Sci. 52 (2009).

157Fujita et al., supra note 133, at 352.

158 Id.

159 Id. at 352–53.

160Of course, not all of these factors may be applicable for every person contemplating an armed robbery.

161Whether this goal is a product of some belief in adherence to the law or merely a desire to avoid punishment may be an important distinction, but for now I do not address this question.

162 See Fujita, supra note 142, at 1483 (“Cooler, more cognitive thinking is generally linked to more global, high-level construals, which may allow people to transcend the affective, appetitive features of a temptation.”).

163This dovetails with Stephen Morse’s explanation that “control excuses are best understood as irrationality or internal coercion claims.” Morse, supra note 6, at 1656.

164Recent government actions regarding public health concerns, such as obesity and road hazards, comport with psychological findings about construal. For example, New York City’s recent push to post calorie information on all fast food is in part an effort to change how individuals construe the choice to eat particular foods. See James Barron, Restaurants Must Post Calories, Judge Affirms, N.Y. Times, Apr. 17, 2008, at B4 (noting that the district court, in upholding the law, stated that providing calorie information will lead at least some consumers to choose lower calorie options, thereby reducing the incidence of obesity). The posting of calories serves as a reminder of the high-level construal of the act of eating, say, a Big Mac. While the low-level construal would focus on the taste of the burger and the enjoyment of satiating one’s hunger, the high-level construal is about the longer term consequences of eating unhealthy, fatty foods.Similarly, campaigns about texting while driving attempt to shift consumer focus away from the low-level construal—immediate gratification of communicating instantly with one’s friends—to the high-level construal of potential danger and longer term risk to life and limb. See, e.g., A Message from Secretary LaHood, Distraction.gov, http://www.distraction.gov/content/about-us/message-from-secretary-LaHood.html (last visited Mar. 27, 2012) (describing a campaign initiated by the U.S. Department of Transportation to end distracted driving). Interestingly, some of these campaigns attempt to invoke grave consequences by depicting them in graphic and detailed ways—fighting a low-level positive construal with a low-level negative construal. See, e.g., PSA Highlights Aftermath of Distracted Driving Crashes, WBALTV.com (Mar. 3, 2011, 6:07 PM), http://www.wbaltv.com/r/27067600/detail.html (noting that videos discourage texting while driving by “depicting the fateful moments leading up to a crash and the heartbreaking news at the hospital”). This approach may be supported by new research suggesting that, under certain conditions, low-level construals may help self-control efforts. See Brandon J. Schmeichel et al., Self-Control at High and Low Levels of Mental Construal, 2 Soc. Psychol. & Personality Sci. 182 (2011).

165 See supra notes 146–48 and accompanying text.

166Although the strength model stemmed originally from the automatic versus effortful cognition model of self-regulation, which can be integrated into the temporal construal model, the strength model itself has not been explored vis-à-vis its relationship with temporal construal. As Fujita and colleagues have noted, “[A] construal-level analysis is largely mute with respect to ego-depletion models of self-control. . . . [A] construal-level analysis . . . has little to say about the ego-depletion model’s core tenet that self-control is a limited resource.” Fujita et al., supra note 133, at 364 (citation omitted).

167The strength model has been experimentally tested against two competing models, the schema model and the skill model. The schema model suggests that self-regulation is a set of behaviors that can be activated by the appropriate setting so that, in repeated settings, self-control would remain at a similar level. Mark Muraven et al., Self-Control as Limited Resource: Regulatory Depletion Patterns, 74 J. Personality & Soc. Psychol. 774, 775 (1998). The skill model suggests that individuals ought to learn and improve as they engage in repeated self-regulation, so that their performance in a series of self-control tasks ought to steadily improve. Id. Research has not provided support for these theories but has provided support for the strength model. See id. at 786.

168Researchers have settled on the term “ego depletion,” rather than self-control depletion, because the resource pool from which self-control comes seems to be broader, serving more than just a regulatory function. See Roy F. Baumeister et al., Ego Depletion: Is the Active Self a Limited Resource?, 74 J. Personality & Soc. Psychol. 1252, 1252 (1998).

169 Id.

170Baumeister & Heatherton, supra note 128, at 3.

171Baumeister et al., supra note 168, at 1263.

172Baumeister & Heatherton, supra note 128, at 3.

173Baumeister et al., supra note 168, at 1254.

174 Id. at 1255. The numbers were dramatic: the radish eaters spent a little over eight minutes working on the problem, while the cookie eaters spent almost nineteen minutes plugging away. Id. In a control condition where no food was eaten, participants spent about twenty minutes on the task, a number not significantly different from that of the cookie eaters’. Id.

175 Id. at 1256.

176 Id. at 1256–58.

177 Id. at 1258–59.

178Megan Oaten & Ken Cheng, Academic Examination Stress Impairs Self-Control, 24 J. Soc. & Clinical Psychol. 254, 254 (2005).

179Matthew T. Gailliot et al., Self-Regulatory Processes Defend Against the Threat of Death: Effects of Self-Control Depletion and Trait Self-Control on Thoughts and Fears of Dying, 91 J. Personality & Soc. Psychol. 49, 55–57 (2006).

180 See Baumeister, supra note 103, at 719.

181Dianne M. Tice et al., Emotional Distress Regulation Takes Precedence over Impulse Control: If You Feel Bad, Do It!, 80 J. Personality & Soc. Psychol. 53, 53 (2001).

182 Id.

183 Id.

184 Id. at 64.

185Baumeister, supra note 103, at 719. Baumeister continues, “Clarifying the relations between emotional distress and regulatory success and failure will be one of the most theoretically interesting challenges of the coming years.” Id.

186 See Barbara L. Fredrickson et al., The Undoing Effect of Positive Emotions, 24 Motivation & Emotion 237, 238–39, 255–56 (2000); Tice et al., supra note 181.

187 See Melissa A. Cyders & Gregory T. Smith, Mood-Based Rash Action and Its Components: Positive and Negative Urgency, 43 Personality & Individual Differences 839, 840–41 (2007); Ayelet Fishbach & Aparna A. Labroo, Be Better or Be Merry: How Mood Affects Self-Control, 93 J. Personality & Soc. Psychol. 158, 158 (2007); Carrie L. Wyland & Joseph P. Forgas, On Bad Mood and White Bears: The Effects of Mood State on Ability to Suppress Unwanted Thoughts, 21 Cognition & Emotion 1513, 1519 (2007).

188Muraven et al., supra note 167, at 776–79; Harry M. Wallace & Roy F. Baumeister, The Effects of Success Versus Failure Feedback on Further Self-Control, 1 Self & Identity 35, 38 (2002).

189Mark Muraven et al., Conserving Self-Control Strength, 91 J. Personality & Soc. Psychol. 524, 525 (2006).

190 Id.

191 Id. at 527.

192 Id. at 536.

193Baumeister et al., supra note 168, at 1260–61. Individuals whose self-control was depleted were more passive in their responses to movie watching; they watched longer when continuing was passive than when continuing was active. Individuals without this depletion watched the movie for the same length of time, regardless of whether the stopping mechanism was active or passive. Id.

194Kathleen D. Vohs et al., Making Choices Impairs Subsequent Self-Control: A Limited-Resource Account of Decision Making, Self-Regulation, and Active Initiative, 94 J. Personality & Soc. Psychol. 883, 894–95 (2008).

195Matthew T. Gailliot et al., Self-Control Relies on Glucose as a Limited Energy Source: Willpower Is More than a Metaphor, 92 J. Personality & Soc. Psychol. 325 (2007).

196 Id. at 333.

197 Id. at 334.

198Mark Muraven et al., Longitudinal Improvement of Self-Regulation Through Practice: Building Self-Control Strength Through Repeated Exercise, 139 J. Soc. Psychol. 446, 448 (1999).

199Id.

200Id. at 451–52.

201Id. 453–56.

202 Id. at 451–52.

203Gailliot et al., supra note 195, at 330–33.

204Brandon J. Schmeichel & Kathleen Vohs, Self-Affirmation and Self-Control: Affirming Core Values Counteracts Ego Depletion, 96 J. Personality & Soc. Psychol. 770, 770–79 (2009).

205Mark Muraven & Elisaveta Slessareva, Mechanisms of Self-Control Failure: Motivation and Limited Resources, 29 Personality & Soc. Psychol. Bull. 894, 903 (2003).

206Dianne M. Tice et al., Restoring the Self: Positive Affect Helps Improve Self-Regulation Following Ego Depletion, 43 J. Experimental Soc. Psychol. 379, 379, 383–84 (2007).

207C. Nathan DeWall et al., Violence Restrained: Effects of Self-Regulation and Its Depletion on Aggression, 43 J. Experimental Soc. Psych. 62 (2007).

208 Id. at 64–68; see also David Gal & Wendy Liu, Grapes of Wrath: The Angry Effects of Self-Control, 38 J. Consumer Res. 445 (2011) (finding that the exertion of self-control correlates with subsequent angry behavior in a variety of forms).

209Thomas F. Denson, Angry Rumination and the Self-Regulation of Aggression, in Psychology of Self-Regulation: Cognitive, Affective, and Motivational Processes 233, 234–35 (Joseph P. Forgas et al. eds., 2009).

210 But see Bradley R.E. Wright et al., Does the Perceived Risk of Punishment Deter Criminally Prone Individuals? Rational Choice, Self-Control, and Crime, 41 J. Res. Crime & Delinq. 180, 205 (2004) (analyzing data to suggest that the threat of punishment is most important for those who are low in self-control and high in self-reported criminal propensity).

211 See DeWall et al., supra note 207.

212 See Arenella, supra note 25; see also David Dolinko, Three Mistakes of Retributivism, 39 UCLA L. Rev. 1623, 1655 (1992) (“The alacrity with which even thoughtful and sophisticated retributivists such as Morse, Morris, and Dressler dismiss suggestions that economic, social, cultural, or psychological deprivations might excuse or mitigate criminal conduct suggests that the harshly punitive attitudes of legal actors and the public that were noted above are not misuses of retributivism but its logical outgrowth.” (footnote omitted)).

213Robert Weisberg, The Values of Interdisciplinarity in Homicide Law Reform, 43 U. Mich. J.L. Reform 53, 75 (2009).

214 See supra notes 134–41 and accompanying text.

215For discussions of use of the “cultural defense,” in which defendants in the United States, typically from another country, attempt to introduce evidence of their culture of origin as the basis for lack of culpability, see James J. Sing, Note, Culture as Sameness: Toward a Synthetic View of Provocation and Culture in the Criminal Law, 108 Yale L.J. 1845 (1999); and Sharon M. Tomao, Note, The Cultural Defense: Traditional or Formal?, 10 Geo. Immigr. L.J. 241 (1996).

216Baumeister & Heatherton, supra note 128, at 9.

217 Id. at 11.

218See infra text accompanying note 110.

219There is certainly evidence that corporate norms have a strong effect on individuals within the corporation. See, e.g., Tom R. Tyler & Steven L. Blader, Can Businesses Effectively Regulate Employee Conduct? The Antecedents of Rule Following in Work Settings, 48 Acad. Mgmt. J. 1143 (2005).

220For example, roughly one-quarter to one-third of prisoners report that the crime that they committed occurred while they were under the influence of drugs. See Office of Nat’l Drug Control Policy, Exec. Office of the President, ONDCP Drug Policy Information Clearinghouse Fact Sheet: Drug Data Summary (2003), available at http://www.expomed.com/drugtest/files/drugdata.pdf.

221I leave to the side the free will issue implicated in addiction. See supra text accompanying notes 91–93.

222For crime related to drug sales, rather than drug use, the picture is somewhat different. This crime, using a self-control lens, looks more like other types of premeditated crimes because the individual’s self-control has not been impaired by an outside influence. The first problem is a lack of the correct standard by which to monitor one’s actions; the second concern is with the problem of temporal construal and its distortion of present benefits to impair rational behavior.

223Hirschi, supra note 124, at 540.

224 See, e.g., Ozlem Ayduk et al., Regulating the Interpersonal Self: Strategic Self-Regulation for Coping with Rejection Sensitivity, 79 J. Personality & Soc. Psychol. 776, 777 (2000).

225 See Wright et al., supra note 210.

226 See, e.g., Hirschi, supra note 124, at 540; Kathleen D. Vohs & Roy F. Baumeister, Understanding Self-Regulation: An Introduction, in Handbook of Self-Regulation: Research, Theory, and Applications, supra note 111, at 1, 8.

227Criminology focuses on “what actually causes crime and on how best to control it . . . .” Stuart A. Scheingold, Constructing the New Political Criminology: Power, Authority, and the Post-Liberal State, 23 Law & Soc. Inquiry 857, 859 (1998). The scope of criminology goes well beyond concerns about self-control, but it is a well-worn trope in the criminology literature that self-control failure plays a significant role in producing criminal behavior. See, e.g., Gottfredson & Hirschi, supra note 124, at 89–91.

228 See, e.g., State v. Sikora, 210 A.2d 193, 203 (N.J. 1965) (“[W]e cannot accept a thesis that responsibility in law for a criminal act perpetrated by a legally sane defendant, can be considered nonexistent or measured by the punishment established for a crime of lower degree, because his act was motivated by subconscious influences of which he was not aware, and which stemmed inevitably from his individual personality structure.”).

229 See, e.g., Zamora v. Phillips, No. 04-CV-4093 (JFB), 2006 WL 2265079, at *6 (E.D.N.Y. Aug. 8, 2006) (“In determining whether a petitioner has acted out of a loss of self control, the court will look at the petitioner’s conduct before and after the homicide.”).

230Others have explicitly noted the normative nature of criminal law efforts around doctrine that centers on questions of self-control. See Stephen J. Morse, The Irreducibly Normative Nature of Provocation/Passion, 43 U. Mich. J.L. Reform 193, 205 (2009).

231Thus, some researchers have described self-control from a motivational perspective: “If the person does not have a strong inner desire to reach [a particular] standard, then merely knowing the standard is not likely to produce effective self-regulation.” Joseph P. Forgas et al., The Psychology of Self-Regulation: An Introductory Review, in Psychology of Self-Regulation: Cognitive, Affective, and Motivational Processes, supra note 209, at 1, 11.

232Even if the temporal construal issues made this calculation tip in favor of the criminal activity, that calculation represents a different situation from one in which the actor truly believes that the action is completely appropriate, per some internal standard.