Updating the study of punishment.

• Author: Meares, Tracey L.

INTRODUCTION I. PUNISHMENT AND SOCIAL SCIENCE: WHAT WORKS, WHAT DOESN'T, AND WHY A. Substitution Effects B. Decision Framing C. Educational Impact of Criminal Law D. Inverse Sentencing Effect E. Impact of Social Control E. Punishment Practice and Legitimacy II. PUNISHMENT AND MODERNITY: COMPUTER CRIMES III. WHAT DOES PUNISHMENT MEAN? IV. CRIMINAL LAW AND THE LEGISLATIVE PROCESS CONCLUSION INTRODUCTION

Criminal law, for much of the nineteenth century and part of the twentieth, was at the forefront of interdisciplinary studies in law. Criminologists borrowed heavily from psychology, sociology, and philosophy in an attempt to understand why people act the way they do and how government should punish them. Yet recently, a movement inward has dominated criminal law scholarship. Suffused by doctrine after doctrine, many criminal law scholars now are content to accept technical legal rules instead of asking whether those rules accord with modern knowledge about human behavior.

Recent years have witnessed a tremendous outpouring of research in economics, psychology, sociology, and other disciplines concerning how institutions, incentives, and rules actually affect behavior. This research has had a significant impact on criminal law scholarship. But it has had almost none on popular criminal law textbooks and thus (we suspect) next to none on the education of criminal lawyers. (1)

The narrowness of conventional criminal law is unfortunate. The implosion may lead to incomplete answers to age-old questions in criminal law, and it has deterred criminal lawyers from asking questions that are commonplace in other areas of law. Perhaps most importantly the failure to fully engage in the classroom the kinds of questions that are being pursued in contemporary scholarship puts our students at risk of being ill-equipped to deal with the pressing questions of criminal justice policy. (2)

This state of affairs is in desperate need of correction. To illustrate, we review some basic themes that a useful casebook on criminal law should cover. We will sketch four areas in which interdisciplinary approaches to thinking about the purposes punishment can and should be incorporated into teaching criminal law. Notably, each case emphasizes nonretributivist approaches to punishment, which we believe have gotten short shrift in criminal law textbooks published most recently. The four areas are: (1) the impact of social science research on our contemporary understandings of punishment; (2) modern doctrinal analogues to theft--computer crimes; (3) expressive values of punishment; and (4) criminal law and the legislative process. Through our examination of these four areas, we hope to demonstrate the problems with the typical, stunted view of punishment and the value of our approach.

1. PUNISHMENT AND SOCIAL SCIENCE: WHAT WORKS, WHAT DOESN'T, AND WHY

   To the detriment of students, legal casebooks largely limit their focus to classic deterrence insights. The standard trope in criminal law, both in scholarship as well as contemporary public understanding, is that enacting high penalties on a particular crime will deter offenders from committing it. (3) The traditional analysis simply turns on whether the penalty is set at an appropriate level to optimize deterrence--balancing the cost of the activity against the cost of enforcement. We do not dispute the central insights of economic theory that people generally act to maximize their preferences and that crime is an area ripe for application of these concepts. However, modern deterrence analysis must incorporate several refinements to the deterrence function, particularly: substitution effects, decision framing, educational impact of laws, social control, and perceived legitimacy. We will discuss each in turn.

   1. Substitution Effects

      The most pervasive economic conception of criminal law, made famous in modern times by economist Gary Becker, views the legal sanction for a given act as its "price" and asks whether that price will exceed the benefit of the criminal act to the criminal. (4) Within the economic tradition itself, some (most preeminently George Stigler) have explained that this calculation misses a crucial variable for optimality: marginal deterrence. The idea is essentially the problem of cliffs--exacting equal penalties for crimes of lesser and greater magnitude leads to crimes of greater magnitude: "If the thief has his hand cut off for taking five dollars, he had just as well take $5,000." (5) The marginal deterrence argument, therefore, is one about creating incentives for individuals to refrain from committing the same crime on a greater scale. While the traditional question asks whether a penalty for X deters X, the marginal deterrence one asks whether a penalty for X may prompt commission of the marginally more severe crime X + 1 because that crime receives the same magnitude of punishment as X. (6) Unfortunately, contemporary casebooks, when they mention deterrence, generally omit this key point. Instead, the analysis is framed as whether a higher penalty on X will produce greater "deterrence," without asking what activity precisely is being deterred and what behavior is being encouraged through the law. (7)

      Once economics is taken seriously, the problems with contemporary criminal law analysis become even more acute. Marginal deterrence is only the tip of the iceberg, for it functions as an illustration of a broader concept at work: substitution effects. Put simply, two products are substitutes when they compete with each other and are complements when they "go together." Consumers will tend to use more of a good--to substitute in favor of the good--when its relative price falls, and to use less of it--to substitute away from the good--when its relative price increases. If the price of tea increases, for example, substitution theory predicts that the demand for coffee would increase. But the demand for other products that go with tea, such as lemons, may drop because tea and lemons are complementary products.

      Instead of framing the deterrence inquiry as simply whether a penalty for crime X will reduce X (the conventional perspective) or lead to X+1 (the marginal deterrence one), another question that has to be asked is whether the penalty on X will distort behavior and lead people to commit an altogether different crime (Y, Z, or some combination of the two). Y and Z may be other crimes, or they may be lawful endeavors. The possibility of lawful endeavors illustrates just how criminal law has unconsciously relied on the substitution concept: The whole point of deterrence is to make the price of a crime high enough so that a criminal will "substitute" forgoing the crime. Just as a high price on train rides means that some people will not take them and ride bicycles instead, a high price on a crime, it is thought, means people will not commit that criminal act. When it comes to crime, however, most of us don't take the economics seriously enough to examine whether an analogue to the bicycles exists: We assume that deterrence works and--poof!--a would-be lawbreaker is now magically converted into a law-abider.

      Consider, for example, the way our government treats crack cocaine. Congress passed dramatic penalties against crack cocaine only a few days after learning of the drug's existence. (8) The mandatory-minimum scheme Congress enacted provides that a minor crack dealer caught with five grams of crack will be in jail for at least sixty months, even on a first offense. (9) Yet legislators never gave serious consideration to what the impact of high crack penalties would be on consumption of other drugs. Much attention has been given to the racial implications of the disparity between powder cocaine and crack cocaine. (10) But none on substitution effects.

      This lack of attention to substitution issues is troubling given the significant disparity between punishments for possessing crack and punishments for possessing other drugs. Simply by weight, the ratio of crack to heroin penalties can be as high as 20:1. (11) Indeed, a dealer can carry 375 grams of heroin and be punished at the same level--5 to 6 years--as the 5-gram crack dealer. When drug dosage is factored into the equation, the crack to heroin punishment ratio can be 80:1 or even higher. (12) As such, drug dealers, who are motivated at least in part by money and a desire to avoid incarceration, would be much better off carrying heroin instead of crack. And while it is difficult to test the substitution effect here without better data, crack consumption has decreased since 1988 while heroin consumption has increased. (13) Perhaps the shift isn't due to substitution effects. But our policymakers, and our casebooks, do not even ask these questions. (14)

      Thinking about criminal law, it is easy to understand how crimes committed for profit, like drug dealing, are ripe candidates for substitution analysis, but it is more difficult to imagine how other crimes can be analyzed in such terms. Yet even crimes of passion may be examined in terms of substitution. (15) Passion, after all, comes in different forms, and a penalty structure may induce people to act in particular ways by assigning costs to particular passionate activities. As Richard Herrnstein puts it, when husbands and wives start throwing dishes at each other, they do not usually throw the fine china. (16)

      Take what seems like the quintessential example in which substitution would not occur: rape. Insofar as these categories are separable--and the argument does not depend on their separation--is rape a crime of sex, violence, or domination? If rapists seek sex, it might follow that legalizing prostitution will reduce the frequency of rape. If they seek to dominate and humiliate, legalized prostitution may provide a substitute as well. (17) To the extent that rapists seek violence, lowering the penalties for other violence, say assaults, may reduce the...