Criminal Law and Philosophy

, Volume 3, Issue 3, pp 271–287 | Cite as

History’s Challenge to Criminal Law Theory

Original Paper


After briefly sketching an historical account of criminal law that emphasizes its longstanding reach into social, commercial and personal life outside the core areas of criminal offenses, this paper explores why criminal law theory has never succeeded in limiting the content of criminal codes to offenses that fit the criteria of dominant theories, particularly versions of the harm principle. Early American writers on criminal law endorsed no such limiting principles to criminal law, and early American criminal law consequently was substantively broad. But even with the rise of theories in the mid-nineteenth century that sought to limit criminal law’s reach, codified offenses continued to widely and deeply regulate social life and exceed the limits of those normative arguments. This essay suggests that this practical failure of criminal law theory occurred because it was never adopted by an institutional actor that could limit offense definitions in accord with normative commitments. Legislatures are institutionally unsuited to having their policy actions limited by principled arguments, and courts passed on the opportunity to incorporate a limiting principle for criminal law once they began, in the Lochner era, actively regulating legislative decisions through Constitutional law. The one avenue through which criminal law theory has had some success in affecting criminal codes is through the influence of specialized bodies that influence legislation, especially the American Law Institute advocacy of the Model Penal Code. But the institutional structure of American criminal law policymaking permits an unusually small role for such specialized bodies, and without such an institutional mechanism, criminal law theory is likely to continue to have little effect on actual criminal codes.


Criminalization History Law commission 

The body of criminal law theory concerned with the proper scope of criminalization, or what conduct legitimately may be criminalized, is characterized by frustration. The frustration arises because criminal law theory has been insufficiently influential in shaping actual criminal codes, and in particular it has been unsuccessful in limiting the content and types of offenses. American criminal law scholars widely express concern with both the overreaching scope of American criminal codes and the conceptual incoherence, or absence of organizing and limiting principles, of those codes.

For theory to have influence, it needs a means to affect the institutions of criminal lawmaking. In an important sense what that means is, always and ultimately, simply the persuasive force of argument. But criminal lawmaking turns out to have powerful counter-forces that work against criminal law theory’s commitments for criminal law, making acceptance or implementation of theorists’ arguments difficult; thus criminal law remains unorganized around a coherent set of limiting principles. Criminal law is mostly a legislative product, and so it is legislatures that largely create this unhappy state of affairs. Changing the structure of criminal lawmaking, then, so as to diminish the power of countervailing pressures that conflict with well conceived theory, may be the way—likely the only way—to increase theory’s influence. But such a change in lawmaking practice conflicts with the long history of criminal lawmaking. That history is characterized by legislatures’ use of criminal law broadly and casually as a regulatory tool, meaning that criminal law long has been a primary option to which legislatures turn when seeking to regulate almost any sort of conduct. Criminal law’s breadth and structure is in a sense path dependent; it is a product of a long established practice of making criminal law in ways that provide relatively little opening for the influence of criminal law theory, while leaving relatively greater openings for countervailing influences that argue for expansive criminalization.

Interestingly, there was a long historical period in which, in retrospect, one might have expected the rise, if not triumph, of criminal law theories in judicial doctrines that could have contained the reach of criminal law and limited criminal sanction to blameworthy, harmful conduct. That period followed the first three-quarters of a century of American criminal law, during which criminal law was a widely used regulatory tool that reached invasively across much of social, commercial and private life. Courts and treatise writers of that period largely endorsed criminal law’s substantive breadth (though not always the magnitude or mode of punishment). But in the latter third of the nineteenth century, two shifts occurred that might have joined forces to reshape and constrain criminal law, yet they failed to do so. First, scholars gradually shifted toward the view that criminal law should be reserved for harmful conduct and liability should be limited to offenders who manifested some moral blameworthiness in their choice of action. Second, American courts began to devise doctrines that significantly constrained legislative lawmaking, including criminal lawmaking. Under new constitutional doctrines, they barred criminalization (as well as civil regulation) of certain kinds of conduct. Yet the rise of this constitutionally based judicial power was little influenced by criminal law theory, and this historical moment represents a path not taken for criminal law regulation. This continued to be the story throughout the twentieth century: courts barred legislatures from a range of regulatory activity, including much criminalization, but always without a theory of the limits and premises of substantive criminal law. Criminal law theory failed to influence courts significantly, and it continues to have little persuasive force in legislatures.

In this essay, I want to offer a tentative account to explain how scholarly development of criminal law theory shifted to its current, critical stance and yet neither courts nor legislatures (nor democratic pressure on those bodies) joined that critical view in ways that successfully restrain and reshape criminal lawmaking. I suggest that it is not so much the nature and scope of criminal law that changed; criminal law has always been expansive. Rather, academic or elite assessments of criminal law’s appropriate scope changed. Scholarly opinion is largely persuaded by limiting commitments such as the harm principle, but courts, legislatures and to some degree popular opinion never have been. The story of why criminal law continues to exceed the parameters of a coherent organizing idea is, I suggest, a story about limited ability of such ideas, principles or doctrines to shape law, especially within the structure of American legislative—and broader political—process.

Nevertheless, that story suggests at least a possibility for the greater influence (if not triumph) of criminal law theory’s limiting principles. The path toward those ideas gaining more influence over substantive criminal law lies in changing the practices and institutions in which criminal law is made. That change, in fact, can be seen in familiar models: criminal law theory would more effectively shape criminal lawmaking if criminal law were enacted and revised the way that many other bodies of substantive law increasingly are—through the use of expert committees or groups that shape proposals for legislatures and thereby affect legislative outcomes.

History and Criminal Law’s Scope

The American nation was founded on limited government principles, reacting to experiences of government overreaching, including overreaching in the scope of criminal law.1 Some of those central limited-government principles are, as incorporated in the Bill of Rights, specific corrections to oppressive criminal procedure practices. Yet the breadth of American substantive criminal law has always been broad. As Markus Dubber has noted, there was no rethinking, in the Founding era or early thereafter, of what a republican, limited-government, liberty-maximizing criminal law would look like.2 Americans were content with an expansive criminal law as a tool of wide-ranging regulation of both private and commercial life. One reason, perhaps, is that after Independence, criminal (and civil) regulation was exercised by local authorities rather than distant ones; while much criminal law had been promulgated by state and local governments in the colonial period, all law was, post-independence, much closer to a product of democratic preferences generated by locally accountable authorities. On that view, it was not so much the content or scope of criminal law generally that was objectionable but its democratic malleability—the ability of popular sentiment to effect changes in law’s terms or scope. But with a few notable exceptions—some prominent procedural practices and the reach of capital punishment, for example—much of the scope of English criminal law fit comfortably with American sentiments.

Interestingly, the scope of substantive criminal law in early American history was—in as much as one can make this sort of comparison across the span of two centuries—roughly as broad as contemporary criminal codes. That comparison is necessarily a normative and contestable one; the activities of ordinary human affairs and social life have changed so much in two centuries that the subjects for criminal regulation have changed dramatically. Nonetheless, I want to defend a generalization about the relatively consistently broad scope of American criminal law across more than two centuries by highlighting examples of the substantial breadth of early criminal codes.

Consider a prominent nineteenth century decision that illustrates not only a typical criminal law of the time but the prevailing judicial view of the law’s legitimacy. Commonwealth v. Alger3 is one of the most prominent state-court public law cases of the nineteenth century. Written by Massachusetts Chief Justice Lemuel Shaw, it is widely taken to be an influential and representative statement of nineteenth century judicial acceptance of a broad definition for the legislative police power, of substantial limitations on private property interests to serve public interests, and of the consequently limited reach of takings doctrine in that era.4 What is often less emphasized is that Commonwealth v. Alger was a criminal prosecution. Cyrus Alger built a wharf on his own property, but extending into Boston Harbor beyond the wharf-limit lines defined by the legislature designed to keep the harbor navigable. Building this wharf in violation of the state code made Cyrus Alger a criminal.

Alger’s conviction, and the criminal form of Massachusetts’ water navigation regulations, are relatively little remarked upon because, for the time, the criminal form of this regulatory enforcement was unremarkable. Throughout the first half of the nineteenth century, criminal law was widely employed as a tool of local regulation under a broadly defined state police power. Common law offenses, carried into the early nineteenth century in state statutes, prohibited buying merchandise outside of designated marketplaces, buying goods in bulk for resale, or reselling any goods within four miles of the place of purchase.5 New Hampshire enforced its criminal provisions against traveling or working unnecessarily on Sundays.6 Broad vagrancy statutes criminalized unemployment and “strolling … without lawful purpose” among a range of other personal life choices.7 Regulations aimed at fire prevention, including building codes (which required brick rather than wooden structures) and tight governance on gunpowder transport and storage, carried substantial fines and prison terms, especially in some urban areas such as New York.8 Criminal ordinances set minimum distances between buildings and roads and punished obstructions in roads, whether caused by loading vehicles or assembling crowds. Rules on weights and measures as well as food purity in market places were backed by criminal sanction. These forms of regulation were part of a large and well established category of public nuisance that justified a range of significant criminal penalties.9

The widespread criminal regulation of nineteenth century private and public life is notable for its breadth and also, through much of that century, for its relatively non-controversial nature. While the subject of regulation was sometimes contested by defendants, as in Alger and Mayo v. Wilson, the use of criminal law as a means to regulate was not. This array of regulatory crimes gives us a picture of the scope of mala prohibita crimes as more ubiquitous than is often suggested. Nineteenth century regulatory crimes stand as direct predecessors to the contemporary range of regulatory crimes. Yet, strikingly from our contemporary perspective, there was much scholarly, judicial and elite support and justification for this broad use of criminal law, and very little of the now-familiar objection that criminal law, in these forms, was overstepping its proper bounds defined by the harm principle, morality or liberty interests.10

This expansive breadth of criminal law is longstanding because few challenges to this scope of regulated activities succeeded in the nineteenth century or earlier (just as, as a matter of constitutional law, few succeed now). This was so because the judiciary affirmed the longstanding authority of government, under the rubric of police power, to regulate nearly any activity, from private personal and familial conduct to property usage and commercial endeavors, whether by civil or criminal statutes, that the legislature deemed necessary for social order or public health. Alger was a prominently cited case for its statement on this point.

… Rights of property, like all other social and conventional rights, are subject to such reasonable limitations in their enjoyment, as shall prevent them from being injurious, and to such reasonable restraints and regulations established by law, as the legislature, under the governing and controlling power vested in them by the constitution, may think necessary and expedient.

…The power we allude to is rather the police power, the power vested in the legislature by the constitution, to make, ordain and establish all manner of wholesome and reasonable laws, statutes and ordinances, either with penalties or without, not repugnant to the constitution, as they shall judge to be for the good and welfare of the commonwealth, and of the subjects of the same.

It is much easier to perceive and realize the existence and sources of this power, than to mark its boundaries, or prescribe limits to its exercise. There are many cases in which such a power is … is so obvious, that all well regulated minds will regard it as reasonable.11

Appellate decisions such as Alger perpetuated a widely accepted view on the legitimacy and necessity of detailed regulation of personal, social and commercial life by criminal as well as civil enforcement mechanisms.

In the latter third of the nineteenth century, federal constitutional law changed in the wake of the post-Civil War amendments (especially the Fourteenth Amendment) to provide a basis for more rigorous judicial supervision of civil and criminal regulation by legislatures (about which more below). Nonetheless, as circumstances changed toward the turn of the twentieth century with such shifts as rising industrialization, urbanization and increased immigration, criminal regulatory offenses directed at these issues were added to nineteenth century criminal codes. For example, Arizona criminally punished both employers and employees if more than twenty percent of a business’s workforce were non-citizens.12 Federal law criminalized employer discrimination against workers on the basis of union membership.13 Pennsylvania criminalized ownership of pharmacies by anyone other than licensed pharmacists.14 Criminal regulation of saloons and dance halls was common.15 In short, legislatures’ affinity for criminal law as a wide-ranging regulatory tool across social and commercial life continued, even as scholarly arguments for restrictions on criminal law’s scope gained acceptance and constitutional law developed doctrines that could have more rigorously restrained criminal law’s expansive scope.

Early Scholarly Acceptance of Broad Criminal Law

For some sense of the early acceptance of a far-ranging criminal law, we can turn to two prominent treatises of the late eighteenth and early nineteenth centuries. The earlier, and surely more influential, is Blackstone’s Commentaries on the Laws of England, the last volume (the one most concerned with criminal law) of which was published in 1769. Blackstone’s treatise was immensely influential in early American jurisprudence and legislative policy making and remained so for more than a century.16 In particular, as Markus Dubber has described in insightful and meticulous detail, it is Blackstone’s category of “police” that provides the expansive conceptual basis for comprehensive government regulation of citizens’ conduct. “By the public police and oeconomy,” Blackstone wrote, “I mean the due regulation and domestic order of the kingdom: whereby the individuals of the state … are bound to conform their general behavior to the rules of propriety, good neighborhood and good manner: and to be decent, industrious, and inoffensive ….”17

The notion of police was not new with Blackstone, but he provided an influential statement of the police concept that recurred in American state constitutions, codes and American common law throughout the nineteenth century.18 In Dubber’s account, the police power was a concept of state power, in some sense distinct from “law” and “justice” (even though “police” was enforced through criminal and civil law) that grew out of the analogy of a head of household having authority to maintain order in the household and, by extension, the king having whatever authority was needed to maintain order in the kingdom.19 Understandings of social order (and related goals such as public health) were expansive, and thus police power regulated moral and intra-familial conduct as well as the sorts of activity noted above. Church attendance, Sunday work laws, bigamy, child support, sobriety, and gainful employment (in contrast to idleness and vagrancy) were all appropriate and well established subjects for police power regulation, through criminal law or otherwise. Likewise with more familiar health and safety regulation and with the broad range of commercial regulation.

Nathanial Chipman’s 1833 treatise Principles of Government is less well known, but it is useful for present purposes as an American account of government authority that follows Blackstone’s by several decades and provides a significant American statement of premises and justifications for wide ranging government power. As William Novak has explored in some detail, Chipman was one of a group of American legal thinkers who stood in some opposition to the understanding of government as a limitation on individual liberty that traces a lineage from Hobbes, Locke, and Rousseau through Blackstone (and later to Mill and others). Rather than viewing the free individual as one who must sacrifice some liberty to achieve the order arising from a social compact, to Chipman “[man] was not made for independence, but for mutual connexion, mutual dependence, and to this everything in his nature is more or less relative.” “[M]an is by nature a social being” with a “social nature,” and thereby has a “duty to forbear any injury to the society of which he is a member.”20 “The laws of nature” are “laws of social nature,” and “man is formed for a state of society and civil government. He is furnished with appetites, passions, and faculties, which in no other state have either gratification or use. In society, in civil society only, can man act agreeable to the laws of his nature.”21

From this premise of man’s social nature grows a justification for government to mediate the complexities of social interaction and perceived necessities of social order, a natural limitation on claims of individual rights, which are limited by the harm they may do to others, and an affirmative vision for government regulation captured in the common law maxim salus populi suprema lex est—the welfare of the people is the supreme law. Whatever exercise of police power is deemed by legislatures to be important for the welfare of the community finds justification—and near immunity from constitutional or other judicial scrutiny—in this vision of an interdependent polity requiring management of conflicting interests and activities, as well as suppression of immoral activity that is seen as a threat to social order.22

This history of justification and endorsement of broad-ranging government authority to regulate, and in particular to criminalize, in the name of social order and welfare has been somewhat overlooked, at least in debates about the proper scope of criminal law, for some decades now. I offer at this point only some tentative speculations on why that is. Part of the explanation seems to be that criminal law’s traditional scope, both before and after the Founding era, seems incongruous with the rhetoric of limited government power that was a central justification for American independence and subsequent federal constitutional design. Dubber has explored this idea in part, noting that the founding of the American republic included no effort to develop a distinctive, republican, limited-power conception of criminal law and police power. Instead, Americans continued employing expansive regulatory practices of traditional English police power, but with the authority implementing that regulation now being state and local (and, eventually, federal) governments rather than the King and a distant English Parliament.23

Another observation one can take from these early sources as well as from federal and state constitutional design is that the founding era was much more concerned about restraining government’s criminal procedure practices than about constraining the substantive reach of criminal law or “police.” Treatises such as Chipman’s and James Wilson’s give notable attention to abuses of criminal process—as well as excessive use of capital punishment—and almost none to regulation of activity that they consider a realm of personal liberty inappropriate for criminalization; there is little evidence among these writers that such a realm exists. This is, of course, consistent with the focus of the federal Constitution (as well as the Declaration of Independence), which contains several explicit provisions addressing abuses of government power in criminal investigation and adjudication but none, explicitly, on perceived overreaches of substantive criminal law.24 There was simply no consensus through the late nineteenth century that some significant realms of personal activity and social life were inappropriate topics for criminal or civil sanction. That view was destined to change.

The Rise of Constraining Principles

Perhaps the strongest reason why the longstanding acceptance of broad criminal law is now overlooked is that it reflects a view that lost favor, especially among criminal law scholars, starting in roughly the second half the nineteenth century. Sources for this change were, broadly, twofold—developments in political theory, and a sea change in American constitutional law in the wake of the Civil War amendments.

Beginning most notably with Mill’s influential 1859 work On Liberty (though one can certainly trace the ideas earlier in time),25 there gradually developed in the second half of the nineteenth century a strong tradition among criminal law theorists—concomitant with trends among government theorists more broadly—of normative argument precisely for what had not been heard before: significant constraints on the reach of criminal law. With Mill and the rejoinder of Sir James Fitzjames Stephens we begin the terms of the debate, that continued well into the twentieth century at least with the Hart-Devlin debate, of whether criminal law should be constrained by a harm principle or (less substantially) by the scope of prevailing moral standards. While there remain significant debates about the meaning of the harm principle,26 I think it fair to say to the harm principle as a general matter has won this scholarly debate. But with respect to the actual state of criminal law, it was only a Pyrrhic victory. While various articulations of that limitation and theories of external constraints on criminal law have been championed by influential theorists such as Joel Feinberg,27 Douglas Husak,28 and drafters of the Model Penal Code,29 the harm principle and comparable limiting ideas seem to do little work—either through courts or legislatures—in constraining the practice of criminal law definition, perhaps because the definition of “harm” has been contentious and excessively elastic.30 The contemporary scope of criminal codes confirms that such theories have had little success in constraining legislatures and, we can infer, broader public opinion.

Interestingly, the modest effect on criminal codes of this tremendous ideological shift does not seem, in hindsight, to have been the predictable or likely outcome of the emergence of these limiting ideals in the late nineteenth century. There was an avenue for them to gain much more influence. The rise of harm principle occurred at a time when American courts were transforming constitutional doctrine into a more rigorous means for supervising legislatures. Initially, in the wake the Civil War amendments, this transformation took the form of the so-called Lochner era of constitutional jurisprudence, in which the Supreme Court employed due process analysis as a substantive check on legislative action that it interpreted to infringe property and contract rights. The Lochner case itself, in fact, began as a misdemeanor prosecution for violating a statute that limited the number of hours that bakers could work per week—another example of the regulatory use of criminal law. The Supreme Court’s Lochner decision held the criminal statute to be an unconstitutional violation of due process because it interfered “with the right and liberty of the individual to contract” for longer work hours if he chose to do so. Lochner-era due process doctrine provided one avenue the Court could have taken to regulate criminal law more generally, perhaps by articulating other realms of liberty—defined perhaps in harm-principle terms of non-harmful conduct—that are off-limits to civil and criminal regulation.

The Court did in fact do something very close to this through a range of constitutional doctrines in the twentieth century. Brandenburg v. Ohio31 and Cohen v. California32 are but two examples of decisions that held unconstitutional statutes that criminalized speech. In Griswold v. Connecticut, the Court (in the wake of a decriminalization trend by most states on the issue) held unconstitutional a criminal law barring use and prescription of contraceptives.33 In Loving v. Virginia,34 the Court cited a more modest but notable state trend of repeal of miscegenation statutes in holding unconstitutional criminal laws that barred interracial marriage. More recently, in Lawrence v. Texas,35 the Court struck down a statute barring homosexual sodomy on grounds that probably makes void all criminalization of private, consensual, adult sex. These are substantial sources of constraint on contemporary criminal codes; they bar regulations that were once widely adopted. Yet, like Lochner-era due process doctrine, none of these doctrines is a principle distinct to criminal law; all address particular forms of conduct that are off limits to any government regulation, civil or criminal. As a result, none provides the coherent organizing and limiting idea that the harm principle aspires to for criminal law.

Mill’s treatise and the harm principle it articulates has had some modest, occasional successes in state courts, where judges occasionally cite Mill in support of a decision to overturn a criminal statute that intrudes on private, harmless conduct. Courts emphasized the harm principle, for instance, in voiding an ordinance that criminalized possession of a small amount of alcohol for personal use in a county that banned alcohol sales,36 and in voiding a state sodomy statute37 (a decision that presaged Lawrence). But the principle has failed to prompt wider-spread judicial willingness to limit the reach of criminal law statutes that are regulatory in nature and address no obvious harm to others. Repeated challenges to misdemeanor statutes that require motorcyclists to wear helmets, for instance, consistently fail notwithstanding litigants’ frequent invocation of Mill’s harm principle.38

Yet the harm principle as an internal organizing commitment of criminal law did have one notable practical success with the American Law Institute drafters of the Model Penal Code. Through the influence of the MPC on state code revision projects in the 1960s and 1970s, the harm principle exerted some notable constraining influence on criminal law. To be sure, the principle’s power—and the MPC drafters’ power—was limited when confronted with an offense implicating contentious social issues. Consider the MPC’s recommendation to abolish consensual sodomy offenses. The MPC drafters (narrowly) approved the decision to remove consensual sodomy as an offense from their model code, and that model’s influence as an instigator of state code revision led to successful repeal of sodomy statutes as part of a broad code overhaul in a few states, starting with Illinois in 1961 and followed by Connecticut in 1969. But more often, state legislators felt political pressure (or personal conviction) to retain statutes criminalizing consensual sodomy (and sometimes other consensual sex) despite accepting other components of the MPC in their wholesale revision of state criminal laws.39 The harm principle, then, had some considerable influence on state codes because of its acceptance by most members of the American Law Institute, who proved for a time to have substantial influence on state code revision projects. But as an abstract principle it failed to win sufficient acceptance among legislatures and their constituents (or perhaps was sufficiently malleable in its definition of harm) that it could not overcome countervailing policy preferences on particularly contested topics.

The harm principle’s role in sodomy reform provides a strong representative example of the limited role such a conceptual and doctrinal commitment can play in defining the content of substantive criminal law. An ideological or conceptual commitment like the harm principle can have significant influence on elites, who themselves then can occasionally find means to play a significant role in the political process that generates criminal law. But criminal law is a legislative product, and it is not primarily a product (especially in the United States) of academic and legal elites, nor of philosophical or conceptual commitments. Legal change, whether through courts or legislatures, is usually characterized by multiple influences of the sort that characterized sodomy law reform.

In Eskridge’s definitive account, social movements for gay rights, including political protest, extensive litigation strategies in state courts, legislative lobbying, scientific research that informed many of those efforts, shifts in constitutional doctrine (especially the rise of privacy doctrine) and cultural change (including changes in heterosexual sex practices) all played roles, over decades, in leading to decriminalization of sodomy and acceptance of homosexuality.40 Harm principle arguments about sodomy—that consensual sex among adults in private homes harms no one and is thus not a subject for state regulation—were persuasive to many ALI members in the 1950s, but not to most of the broader public until decades later. In the meantime, arguments for criminalization of non-procreative sexual activity were made persuasively to many within terms of the harm principle. That is, advocates for sodomy laws marshaled reasons why even private, consensual sodomy does in fact cause grave social harm—because it degrades morality of participants and of society generally, and it contradicts centuries of moral consensus; because it leads to predatory sexual behavior, especially against children; because it spreads disease, etc.41

Those arguments persuade many fewer now than they did a half century ago, but that attitudinal shift did not come from broad adoption of the harm principle; it came from multifarious processes of social change—social movements, cultural shifts and political action, and also academic writing and lobbying—that courts and (more slowly) legislatures responded to. The sodomy debate demonstrates the expansive capacity of the harm principle to accommodate a wide breadth of criminalization. Note the potential capaciousness of John Gardner’s account of the harm principle:

It is no objection under the harm principle that a harmless action was criminalized, nor even that an action with no tendency to cause charm was criminalized. It is enough to meet the demands of the harm principle that, if the action were not criminalized, that would be harmful….

…[T]he harm principle does not say that only harmful wrongs may be criminalized. It says that the criminalization of wrongs is justified only in order to prevent harm. Non-instrumental wrongs, even when perfectly harmless in themselves, can pass this test if their criminalization diminishes the occurrence of them, and the wider occurrence of them would detract from people’s prospects—for example, by diminishing some public good, such as people’s sense of ease with their environment ….42

A conception that expansive leaves much room for debate about meritorious and cognizable harms, and about other limits on criminalization, notably blameworthiness. In light of the perennial difficulty of putting into operation a moral judgment distinct from prevailing popular conceptions of morality, the sort of process and history that characterized sodomy reform, broadly speaking, is how criminal law has always been made and, in all likelihood, will continue to be made. It is the rough product, from legislatures and courts, of shifts in broad social and political views about how to define and address social problems, coupled with legislatures’ longstanding practice of turning to criminal law routinely for small-scale regulatory needs that, at best, marginally accord with harm (and blameworthiness) principles. As a product of such a process, criminal law is unlikely to ever to be characterized by strong, coherent principles that limit its scope; such principles don’t have the power, alone, without support from social, cultural and political consensus, to define criminal lawmaking.

Changing the Practice of Criminal Lawmaking

From this account, is there a path toward resolving the discord between criminal law theory and actual criminal lawmaking? One might divide the options for achieving closer accord between theory and practice into three. First, we could simply revise the theoretical commitment to what criminal law should look like and revert to the sort of broad account that prevailed in the first half of the nineteenth century. Such an understanding would accord with the long tradition (and current practice) of using criminal law as a regulatory form with no meaningful limit regarding harmless conduct. Criminal law could retain its requirements of fault and culpability, but with little constraint as to the topic of criminalization. But this approach seems infeasible as a practical matter, and I will not attempt to further develop the project here; the commitment to normative limits on criminalization is too widely shared for contemporary criminal law theory to simply abandon its premises and uncritically endorse the broad ranging criminalization decisions of legislatures.

If the theoretical side of the equation will not change, that leaves only the prospect of finding ways to pair down criminal codes to more closely align with theoretical commitments. Two other options address this possibility. One has been well developed by others:43 courts could step in to constrain legislative crime definition. As discussed above, courts could devise interventionist doctrines that encompass harm and blameworthiness principles simply to strike down legislative attempts to expand criminal law to conduct that does not fit these parameters. But prospects for this solution seem dim as well. As we have seen, the Supreme Court has never been willing to regulate criminal law’s content in terms internal to the criminal law, no matter how relatively active it has been at times in striking down regulatory action (including criminal statutes) under a range of doctrines.

Further, even if courts took such an interventionist step, it is not clear that that they would, over time, make choices that that scholars widely embrace. Granted, courts can only strike down statutes and not create them;44 thus their active involvement could work only as a one-ratchet of decriminalization. But in a previous era—the Lochner era of roughly the 1880s to the 1930s—when the Supreme Court took an active role in limiting legislatures’ substantive choices about what conduct could be criminalized or otherwise regulated, its aggressive policing of legislative power did not meet with widespread scholarly (or, eventually, political) endorsement. Some decisions are likely widely regarded as salutary. One Lochner-era decision struck down a regulation, enforced with criminal penalties, that allowed only licensed pharmacists to own pharmacy businesses45—not conduct about which the blameworthiness and social harm are obvious. Another struck down a statute that punished with criminal fines employment discrimination against workers on the basis of union membership.46 On the other hand, a typical state court decision of the time, following Supreme Court precedent, upheld a criminal statute regulating the operation of dance halls (for “their tendency is to weaken morals and breed disorder and indolence”).47 It may be that a doctrine constructed specifically on a commitment to harm and blameworthiness principles would guide courts better than did the Lochner-era due process doctrine, which was grounded in protection of contract and property rights. But that seems doubtful given the substantive, contested, historically contingent choices one has to make to determine “harm” or “blameworthiness” outside the core of criminal offenses—the very reason courts have been reluctant to take this doctrinal path.

That leaves a third option: an institutional mechanism to improve legislative lawmaking without judicial supervision. Perhaps the problems of criminal law’s politics—the long history of poor drafting, excessively expansive crime definition, failure to repeal outdated offenses—is remediable by a change in the structure of criminal lawmaking that mediates democratic influences.

American criminal law is perhaps unique in the degree to which it is a product of democratic political process or, put differently, the degree to which it is unmediated in its creation by specialists or expert agencies outside of legislatures. Criminal law is exceedingly wide ranging in its substantive reach, at least on par with tort and administrative law, both of which regulate much of the same conduct and risk creation. Legislatures treat criminal law more a form of law than a substantive body with its own internal integrity and limits; they employ it as a tool available for a wide range of regulatory tasks. This wide-ranging utility challenges any effort to maintain an internal coherence. Narrower bodies of substantive law, such as contracts and commercial sales law, are products mostly not of legislative action but of judicial lawmaking (through common law) and of specialized codes (such as the Uniform Commercial Code) initially drafted by specialized commissions and then offered to legislatures, which frequently adopt them with modest revision. This is the case also with procedural codes for civil and criminal procedure, evidence, and the like: agencies or committees of specialists draft proposals for rules that set the agenda for legislative adoption. “Uniform acts” promulgated by the National Conference of Commissioners on Uniform State Laws have comparable influence (to varying degrees, depending the topic) for more specific topics of substantive law.

Criminal law, however, addresses such a wide range of activities and conduct that its closest counterpart is probably administrative law, which governs many of the same sorts of activities and, collectively, has similarly wide substantive reach—including public capital markets, antitrust, consumer protection, environmental harms and labor relations. With administrative law, however, legislatures delegate much law drafting to agencies, which craft the details of substantive rules, in addition to exercising civil enforcement power. Agencies are hardly apolitical, but their structure creates significant differences in both their lawmaking procedures and their political incentives compared to legislators (or elected prosecutors). The Administrative Procedure Act creates a process for agency lawmaking distinct from legislative process, including (deferential) judicial review of that lawmaking. Nothing really comparable exists in criminal law. Legislatures control the entire crime definition process, largely unmediated by comparable commissions, experts or agencies, and unreviewed by courts.

American criminal lawmaking’s best era—the moment when it reduced the gap between criminal law theory and criminal codes—was when it shifted toward this specialized-commission approach. The American Law Institute’s Model Penal Code had a significant agenda-setting function for state legislatures that sparked a period of code reform in the 1960s and 1970s. Smaller scale examples occasionally occur in states where a legislature will establish a criminal code reform commission or a standing law reform commission to make suggestions for legislation.48 Those bodies have mixed records of success, but when they have any effect they seem usually to be salutary ones. The United Kingdom has a standing Law Reform Commission at the national level that periodically proposes criminal code reform. While the U.K. criminal law scholars also despair of the sprawling state of their criminal code49 (and sometimes despair of the Law Commission’s proposals),50 their commission seems to have some of the beneficial effects of guiding legislative action that American legislatures, especially Congress, could greatly use; at any rate, the Congress has nothing comparable.

Those comparisons, I believe, point us toward a more feasible and thorough solution for the dysfunctional process of American criminal lawmaking. The expansiveness of criminal law, due both to its history and to the weak practical constraints of the harm principle, means that political debate and policymaking apparatus will always play a large role in defining offense content and defining boundaries that distinguish crime from tort and other subjects of civil law regulation. Choices in the institutional design of policymaking bodies strengthen some modes of politics over others, and make some substantive outcomes more likely than others. Criminal law needs legislative action to be more thoroughly delegated to, and mediated by, commissions or agencies that play a substantial agenda-setting role for lawmaking. The structure and operational practices of democratic institutions matter, and a strong, institutionalized, agenda-setting role for a criminal lawmaking agency seems the most likely mechanism for bringing the products of legislative crime definition into alignment with what scholars hope and insist are the defining principles of criminal law.

Even without this sort of change in the structure of legislative criminal lawmaking, it is worth noting that not all democratic lawmaking structures are created equally and some seem more likely to produce moderate criminal law policies than others. We have some evidence that existing variations in democratic practices, without strong agency mediation, improves substantive criminal law—or, more precisely, criminal punishment policy. Incarceration polices in fact vary tremendously among American states. A national average incarceration rate of 443 per 100,000 belies wide state-level differences, from rates of 117 to 736 per 100,000, with neighboring states and states comparable on some common demographic factors showing wide divergence from each other.51 Criminologist Vanessa Barker has offered an innovative explanation for some of these variations. She suggests that some forms of democratic practices and modes of sate governance are much more conducive to producing moderate penal policies than others.52 Specifically, states with strong traditions of civic involvement among citizens and participatory democracy by government actors—frequent use of public hearings, town hall meetings, citizen commissions and the like when developing legislative policies on criminal sentencing—are more likely to yield moderate policies than states with populist traditions and mechanisms (including California’s citizen initiative and referendum options). Barker’s study focuses on only a few states as case studies, and her focus is sentencing policy rather than substantive crime definition. It nonetheless suggests that some models of democratic lawmaking, in the context of some existing political cultures, hold the promise of more moderate, thoughtful criminal lawmaking even without a shift toward greater influence for administrative bodies in the process. But changing a jurisdiction’s public culture and governmental structure, institutionalized political practices is surely more difficult even than the substantial innovation of devising a prominent agenda-setting role for expert commissions.


The long history of legislatures employing criminal law across a wide scope of regulatory projects, and courts’ longstanding reluctance to police the defining parameters of criminal law, leaves little hope that either of those branches will be persuaded to enforce conceptual coherence in criminal codes by consistently respecting criminal law’s core principles. Legislatures respond more strongly to other forces, particularly interest groups, broad cultural shifts and social movements, than to principled argument and normative theory—though theory can contribute to cultural shifts and social movements. Yet different forms of legislative organization can increase the influence of theory and temper the influence of countervailing arguments. For criminal law theory to have an effect on the real world production of criminal offense definition, appropriate and familiar institutional mechanisms have to be in place for those theoretical commitments to influence lawmaking. Until that sort of structural change occurs in the lawmaking process, the feasible possibilities for second-best remedies by courts are likely at best to mitigate the harms of poorly drafted, excessively sprawling criminal codes that have long burdened American jurisdictions.


  1. 1.

    See, e.g., Hay et al. (1975).

  2. 2.

    Dubber (2005, Chap. 4).

  3. 3.

    61 Mass. (7 Cush.) 53, (Mass. 1851).

  4. 4.

    Novak (1996, pp. 19–21) (“Commonwealth v. Alger occupied a central place in nineteenth century jurisprudence.”).

  5. 5.

    See, e.g., Revised Code of Va., Chap. 138 (1819).

  6. 6.

    Mayo v. Wilson, 1 N.H. 53 (N.H. 1817) (upholding a civil challenge to an arrest under criminal statute); Laws of N.H. at 346–49 (1815) (defining “Crimes … Profanation of Sabbath).

  7. 7.
    For a relatively modern example of this sort of longstanding statute, see Papachristou v. City of Jacksonville, 405 U.S. 156 (1972). Jacksonville Ordinance Code § 26—57 held to be unconstitutionally vague in that case defined criminal vagrancy to include:

    ‘Rogues and vagabonds, or dissolute persons who go about begging, common gamblers, persons who use juggling or unlawful games or plays, common drunkards, common night walkers, … persons wandering or strolling around from place to place without any lawful purpose or object, habitual loafers, disorderly persons, persons neglecting all lawful business and habitually spending their time by frequenting houses of ill fame, gaming houses, or places where alcoholic beverages are sold or served, persons able to work but habitually living upon the earnings of their wives or minor children ….

  8. 8.

    Laws of the State of New York, 2 vols. (Albany, N.Y. 1813), 2 (1813), c. 86 (R.L.), 363–70. A Digest of the Laws of New Jersey, 1709–1861, 300 (John T. Nixon ed.) (3d ed. 1861); W. Novak (1996, pp. 57–58).

  9. 9.

    See W. Novak (1996, p. 61); see, e.g., Fisher v. McGirr, 67 Gray 1 (Mass. 1854) (describing and approving criminal sanctions for gunpowder regulation).

  10. 10.

    This account of criminal law’s substantive scope is consistent with Gerald Leonard’s history of American criminal law theory, in which he recounts a dominant view in favor of criminal law as a tool for prevention of social harm with relatively little regard for culpability, a position that was gradually supplanted by the familiar one limiting guilt to those with individual responsibility and moral culpability. See Leonard (2003).

  11. 11.
    Commonwealth v. Alger, 61 Mass. (7 Cush.) 53, 85 (1851). Around the same time as Alger, Thorpe v. Rutland & Burlington Railroad, 27 Vt. 140, 156 (1854), provided another oft-cited statement of this expansive view of legislative authority for civil and criminal regulation:

    One with any degree of familiarity with this subject would never question the right depending upon invincible necessity … to subject persons and property to such regulations as the public security and health may require, regardless of private convenience …. [D]oubts in regard to the extent of governmental authority come from those who have had small experience.

  12. 12.

    See Traux v. Raich, 239 U.S. 33 (1915) (describing and holding unconstitutional an Arizona statute).

  13. 13.

    See Adair v. United States, 208 U.S. 161 (1908) (holding statute unconstitutional).

  14. 14.

    Louis K. Liggett Co. v. Baldridge, 278 U.S. 105 (1928) (holding statute unconstitutional).

  15. 15.

    See, e.g., Dwyer v. Colorado, 261 p. 858, 859 (Colo. 1927) (approving criminal regulation of dance halls).

  16. 16.

    Dubber (2005, at p. xii); Freund (1904, p. 2 & n. 2).

  17. 17.

    Blackstone (1979, p. 162), quoted in Dubber (2005, p. xii).

  18. 18.

    See, e.g., Dubber (2005, p. 59) (citing Freund).

  19. 19.

    Dubber (2005, pp. 47–93).

  20. 20.

    Chipman (1833, pp. 46 & 209); see also Novak (1996, pp. 30–48).

  21. 21.

    Quoted in Novak (1996, p. 31).

  22. 22.

    See generally Novak (1996, Introduction and Chap. 1).

  23. 23.

    See Dubber (2005, Chap. 4).

  24. 24.

    Modern constitutional law has of course developed some limitations on what activity government can criminalize or otherwise regulate. See, e.g., Griswold v. Connecticut, 381 U.S. 479 (1965); Brandenburg v. Ohio, 395 U.S. 444 (1969); Roe v. Wade, 410 U.S. 113 (1973); Lawrence v. Texas, 539 U.S. 558 (1973).

  25. 25.

    See, e.g., Jefferson (1984, p. 285) (“the legitimate powers of government extend to such acts only as are injurious to others.”). See also Beccaria (1764, pp. 17–21).

  26. 26.

    Compare, e.g., Duff (1990, pp. 111–112), and Gardner (2007, pp. 30–32).

  27. 27.

    See generally, Feinberg (1984).

  28. 28.

    Husak (2008, Chaps. 2 & 3).

  29. 29.

    See also Hall (1960, pp. 221–222) (argued that the principle of harm is “an essential organizational construct” of the criminal law).

  30. 30.

    Harcourt (1999); Lacey et al. (2003).

  31. 31.

    395 U.S. 444 (1969) (criminal syndicalism statute violates First Amendment).

  32. 32.

    403 U.S. 15 (1971).

  33. 33.

    Griswold v. Connecticut, 381 U.S. 479 (1965) (overturning statute criminalizing contraceptive use by married couples); see also Eisenstadt v. Baird, 405 U.S. 438 (1972) (extending Griswold to overturn ban on contraception for unmarried people).

  34. 34.

    388 U.S. 1, 6 & n. 5 (1967) (listing states that abolished miscegenation crimes); Wadlington (1966, p. 1189, n. 2) (noting that only seventeen states still had antimiscegenation laws in 1966).

  35. 35.

    539 U.S. 558 (2003).

  36. 36.

    Commonwealth v. Campbell, 117 S.W. 383 (Ky. 1909).

  37. 37.

    Commonwealth v. Bonadio, 415 A.2d 47 (Pa. 1980). See also State v. Douglas, 21 Misc.2d 551, 202 N.Y.S.2d 160 (N.Y. 1959) (citing Mill in overturning an obscenity statute on First Amendment grounds).

  38. 38.

    See, e.g., Picou v. Gillum, 874 F.2d 1519 (11th Cir. 1989); Commonwealth v. Kautz, 491 A.2d 864 (Pa. Super. 1985).

  39. 39.

    The now-definitive history of the Model Code’s effect on reform of sodomy statutes is. Eskridge (2008, pp. 118–126, 144–147, 161–165).

  40. 40.

    See Eskridge (2008).

  41. 41.

    See Eskridge (2008). For some of these arguments, see Bowers v. Hardwick, 478 U.S. 186 (1986).

  42. 42.

    Gardner (2007, pp. 29–30).

  43. 43.

    Stuntz (2001); Husak (2008).

  44. 44.

    I hold aside here courts’ significant power to expand statutes’ reach through broad interpretation, which one might argue is a form of creating new crimes.

  45. 45.

    Louis K. Liggett Co. v. Baldridge, 278 U.S. 105 (1928) (in action brought to enjoin enforcement of criminal statute prohibiting non-pharmacists from owning pharmacies, statute held to violate Fourteenth Amendment due process rights).

  46. 46.

    Adair v. United States, 208 U.S. 161 (1908) (overturning misdemeanor conviction of railroad agent who discharged an employee based on his union membership, holding that the statute violated defendant’s Fifth Amendment Due Process right to liberty and property).

  47. 47.

    Dwyer v. Colorado, 82 Colo. 574, 576, 261 p. 858, 859 (1927) (affirming conviction for operating unlicensed dance hall and holding criminal statute does not violate due process).

  48. 48.

    For an example of state crime commission successfully recommending repeal of criminal offenses, see Va. State Crime Comm’n, The Reorganization and Restructuring of Title 18.2 (2004) (recommending the revision of Virginia’s criminal penalties). See also Act of Apr. 12, 2004, ch. 459, 2004 Va. Acts 657 (repealing crimes suggested by commission).

  49. 49.

    Ashworth (2000). Moreover, some scholars are skeptical of the Law Commission’s proposals for criminal law reform. See Gardner (2007, pp. 33–55).

  50. 50.

    See, e.g., Gardner (2007, p. 33).

  51. 51.

    As a representative year, I use figures from Bureau of Justice Statistics (1998, tbl. 5.4). For more recent data showing comparable contrasts, see Sabol et al. (2007).

  52. 52.

    Barker (2004, 2006). Barker’s four types of state political structures or types of governance are populist, participatory democracy, pragmatism, and patronage-oriented. See Barker (2006, p. 10 & tbl. 1).



I would like to thank Kim Ferzan and Doug Husak, co-directors of the Rutgers University Institute for Law and Philosophy, for inviting me to participate in the conference on criminal law theory at which this paper was presented. I thank the participants there for their comments, and especially Doug Husak and Paul Robinson for close readings and extensive comments on an earlier version of the manuscript.


  1. Ashworth, A. (2000). Is criminal law a lost cause. Law Quarterly Review, 116, 225–256.Google Scholar
  2. Barker, V. (2004). The politics of punishing: A comparative historical analysis of American democracy and punishment variation, 1965-present (Doctoral dissertation).Google Scholar
  3. Barker, V. (2006). The politics of punishing: Building a state governance theory of American imprisonment variation. Punishment and Society, 8, 5–32.CrossRefGoogle Scholar
  4. Beccaria, C. (1764; Edward Ingraham trans. 1819). An essay on crimes and punishment. Philadelphia: Philip H. Nicklin.Google Scholar
  5. Blackstone, W. (1769). Commentaries on the laws of England (Vol. 4). Chicago: University of Chicago Press.Google Scholar
  6. Bureau of Justice Statistics. (1998). Correctional populations in the United States, 1998.Google Scholar
  7. Chipman, N. (1833). Principles of government. Burlington: Edward Smith.Google Scholar
  8. Dubber, M. (2005). The police power. New York: Columbia University Press.Google Scholar
  9. Duff, R. A. (1990). Intention, agency and criminal liability. Oxford: Basil Blackwell.Google Scholar
  10. Eskridge, W. N. (2008). Dishonorable passions: Sodomy laws in America, 1861–2003. New York: Viking.Google Scholar
  11. Feinberg, J. (1984). The moral limits of criminal law. New York: Oxford University Press.Google Scholar
  12. Freund, E. (1904). The police power: Public policy and constitutional rights. Chicago: Callaghan & Co; University of Chicago Press.Google Scholar
  13. Gardner, J. (2007). Offences and defences. Oxford: Oxford University Press.CrossRefGoogle Scholar
  14. Hall, J. (Ed.). (1960). General principles of the criminal law (2nd ed.). New York: Bobbs-Merrill Co.Google Scholar
  15. Harcourt, B. (1999). The collapse of the harm principle. Journal of Criminal Law and Criminology, 90, 109–194.CrossRefGoogle Scholar
  16. Hay, D., et al. (1975). Albion’s fatal tree: Crime and society in eighteenth century England. New York: Pantheon Books.Google Scholar
  17. Husak, D. (2008). Overcriminalization. New York: Oxford University Press.Google Scholar
  18. Jefferson, T. (1984). Notes on the State of Virginia in writings. Library Classics of the United States, Inc.: New York.Google Scholar
  19. Lacey, N., Wells, C., & Cook, O. (Eds.). (2003). Reconstructing criminal law (3rd ed.). Cambridge: Cambridge University Press.Google Scholar
  20. Laws of the State of New York (2 vols., 1813). Albany, N.Y.Google Scholar
  21. Leonard, G. (2003). Towards a legal history of American criminal theory: culture and doctrine from Blackstone to the Model Penal Code. Buffalo Criminal Law Review, 6, 691–833.Google Scholar
  22. Nixon, J. T. (Ed.). A Digest of the Laws of New Jersey, 1709–1861 (3rd ed., 1861).Google Scholar
  23. Novak, W. J. (1996). People’s welfare: Law and regulation in nineteenth century America. Chapel Hill: University of North Carolina Press.Google Scholar
  24. Revised Code of Virginia (1819).Google Scholar
  25. Sabol, W., Minton, T., & Harrison, P. (June 2007). Prison and jail inmates at midyear, 2006. Bureau of Justice Statistics Bulletin.Google Scholar
  26. Stuntz, W. J. (2001). The pathological politics of criminal law. Michigan Law Review, 100, 505–600.CrossRefGoogle Scholar
  27. Virginia Act of April 12, 2004, ch. 459, 2004 Va. Acts 657.Google Scholar
  28. Virginia State Crime Commission, The Reorganization and Restructuring of Title 18.2 (2004). Available at$file/HD15_2004.pdf.
  29. Wadlington, W. (1966). The loving case: Virginia’s anti-miscegenation statute in historical perspective. Virginia Law Review, 52, 1189–1223.CrossRefGoogle Scholar


  1. Adair v. United States, 208 U.S. 161 (1908).Google Scholar
  2. Bowers v. Hardwick, 478 U.S. 186 (1986).Google Scholar
  3. Brandenburg v. Ohio, 395 U.S. 444 (1969).Google Scholar
  4. Cohen v. California, 403 U.S. 15 (1971).Google Scholar
  5. Commonwealth v. Alger, 61 Mass. (7 Cush.) 53, (Mass. 1851).Google Scholar
  6. Commonwealth v. Bonadio, 415 A.2d 47 (Pa. 1980).Google Scholar
  7. Commonwealth v. Campbell, 117 S.W. 383 (Ky. 1909).Google Scholar
  8. Commonwealth v. Kautz, 491 A.2d 864 (Pa. Super. 1985).Google Scholar
  9. Dwyer v. Colorado, 261 P. 858, 859 (Colo. 1927).Google Scholar
  10. Eisenstadt v. Baird, 405 U.S. 438 (1972).Google Scholar
  11. Fisher v. McGirr, 67 Gray 1 (Mass. 1854).Google Scholar
  12. Griswold v. Connecticut, 381 U.S. 479 (1965).Google Scholar
  13. Lawrence v. Texas, 539 U.S. 558 (2003).Google Scholar
  14. Louis K. Liggett Co. v. Baldridge, 278 U.S. 105 (1928).Google Scholar
  15. Loving v. Virginia, 388 U.S. 1, 6 & n.5 (1967).Google Scholar
  16. Mayo v. Wilson, 1 N.H. 53 (N.H. 1817).Google Scholar
  17. Papachristou v. City of Jacksonville, 405 U.S. 156 (1972).Google Scholar
  18. Picou v. Gillum, 874 F.2d 1519 (11th Cir. 1989).Google Scholar
  19. Roe v. Wade, 410 U.S. 113 (1973).Google Scholar
  20. State v. Douglas, 21 Misc.2d 551, 202 N.Y.S.2d 160 (N.Y. 1959).Google Scholar
  21. Thorpe v. Rutland & Burlington Railroad, 27 Vt. 140, 156 (1854).Google Scholar
  22. Traux v. Raich, 239 U.S. 33 (1915).Google Scholar

Copyright information

© Springer Science+Business Media B.V. 2009

Authors and Affiliations

  1. 1.University of Virginia School of LawCharlottesvilleUSA

Personalised recommendations