Jump to ContentJump to Main Navigation
Modern Control Theory and the Limits of Criminal Justice$

Michael Gottfredson and Travis Hirschi

Print publication date: 2020

Print ISBN-13: 9780190069797

Published to Oxford Scholarship Online: October 2019

DOI: 10.1093/oso/9780190069797.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 13 June 2021

General Theory, Public Policy, and the Limits of Criminal Justice

General Theory, Public Policy, and the Limits of Criminal Justice

(p.131) Chapter 7 General Theory, Public Policy, and the Limits of Criminal Justice
Modern Control Theory and the Limits of Criminal Justice

Michael Gottfredson

Travis Hirschi

Oxford University Press

Abstract and Keywords

In this chapter, the general theory of crime depicted in self-control theory is taken as valid, and the implications for criminal justice are explored. The historical connections between classical theory and criminal sanctions are described, and the relations between classical deterrence theories and control theory are examined. The classical theory assumption that deterrence places limits of effectiveness on state sanctions is used in conjunction with the modern notion of self control. The result is that modern control theory, supported by contemporary research on the effectiveness of criminal sanctions, explains why criminal sanctions have limited effectiveness for crime and sets limits on the appropriate use of criminal sanctions. Modern control theory, using classical school assumptions of human nature and choice, shows why public policy should focus on early socialization and prevention.

Keywords:   criminal sanctions, self control, crime, public policy, deterrence theory, incapacitation, causal theory, crime prevention

It is better to prevent crimes than to punish them.

—C. Beccaria (1764/1993, 103)

Public policies about crime and violence are in disarray. The police are at once blamed for rising crime rates, lauded for “smart policing” when they concentrate force on high-crime neighborhoods, and criticized for harsh and discriminatory treatment of suspects. Police leaders claim responsibility for major declines in crime at the same time they shift blame to general social conditions when crime rates appear to increase. In the United States, the huge increase in incarceration during the past 40 years, now routinely referred to as “mass incarceration,” has been supported by politicians at every level, but it is now widely condemned as enormously costly in both financial and human terms. Mass incarceration has resulted in large racial and class differences in imprisonment, and the negative effects of parental incarceration on children exacerbate racial and economic disadvantage across generations. Support for rehabilitation as a purpose for the criminal justice system once characterized US sentencing policy, was claimed by both academicians and politicians to be the source of much that went wrong with the justice system in the 1980s and 1990s, and yet returns as the best way to think about reintegration of offenders post-incarceration. Under the guise of restorative justice, it makes a renewed claim for a central place in criminal justice practices. The juvenile justice system, once the (p.132) hallmark of the prevention ideal in the US justice system, comes under attack for leniency and for ignoring dangerous offenders, resulting in widespread adoption of mechanisms to “waive” juveniles to adult courts for more “serious” offenses.

In New York City, police leaders claim major crime reductions in violence have been achieved by responding harshly to minor instances of “disorder,” such as vagrancy, selling cigarettes without a proper license, and public drug use, and yet these same tactics cause citizen–police encounters that sometimes end in violence and death. In California, proposals to reintroduce discretionary release from prison to save money and avoid injustices are made by the governor who once called for the end of the rehabilitative ideal in corrections and advocated mandatory sentencing. Yet his proposals result in immediate condemnation by prosecutors and chiefs of police, who argue that even minor modifications of harsh sentencing laws cause unacceptable increases in the crime rate and who band together to fight a public referendum supportive of reductions in mass incarceration.

The role of guns in causing or deterring violence is a major issue for presidential politics and public commentary. Debates rage about whether further restrictions on the purchase or possession of guns would have any effect on violent crime (or might increase it), while enormous public concern about what may be done to respond to school and other rampage shootings dominates public discourse about violence. Legislation legalizing marijuana for recreational use is said to reduce unwarranted criminal justice intrusion into the lives of citizens and at the same time is blamed for increases in crime. In California, a public referendum was passed that promised financial relief by creating modest reductions of prison terms for nonviolent offenders; law enforcement quickly claimed that the referendum was responsible for increases in the homicide rate—a claim academicians just as quickly disputed.

In the academic literature, major shifts in the conventional wisdom about research on criminal justice policies are taking place. For many years, the dominant view among scholars has been that research fails to document substantial effects for police tactics on the amount of crime; today, this view is being displaced by the claim by police researchers that “smart policing” and saturation policing focused on “hot spots” can effectively and fairly reduce crime in the cities. It has even been suggested that more police will result in less imprisonment because it can lower crime rates. Other scholars are sharply critical of these claims.

With respect to sentencing policy, researchers focused on “criminal careers,” a tradition dominant in criminology for the past 30 years, seek to tie observational studies of criminal careers to incapacitation and deterrence. The criminal career advocates now must confront a wave of evidence (p.133) that such capabilities have eluded them and a growing consensus that imprisonment rates do not much affect the crime rate. Decades of research, underwritten with substantial research budgets from the federal government, searching through cohorts of delinquents and other offenders seeking reliable markers that would enable the selection among them for potential lengthy criminal careers (and thereby provide the criminal justice system the ability to selectively incapacitate high-rate offenders) has yielded no practical results.

A recent (and, it turns out, long-standing) example of the disjunction between criminological facts and their interpretation and policy is provided by immigration. Criminologists have provided substantially inconsistent advice to policymakers over time about the connection of immigration to crime. Once viewed by sociologists as a major cause of crime, worthy of validating major sociological theories of why people engage in crime, good research now calls the effect of immigration on crime into question, even suggesting that the most likely effect is negative (i.e., in the United States, immigrants are less likely than the native population to engage in crime). As a consequence, a criminal justice system based, in part, on the expectation of high crime rates for immigrants will be wasteful of resources, disruptive for families, expensive, and will needlessly intervene in the lives of a low-risk population.

In the United States, large swings in the crime rate over many years cannot be shown to be attributable in the main to changes in the criminal justice system—not to the massive increase in rates of incarceration, not to the great escalation in criminal justice expenditures, and not to increases in policing. The failure to find substantial general deterrence effects for imprisonment is a great puzzle to many practitioners and scholars (even as police continue to credit themselves with the crime decline and blame “society” for modest increases). Still, some academics ramp up the promise for selective effectiveness in the criminal justice system, arguing that enforcement targeted to the right groups will even lower imprisonment. Yet, deterrence and incapacitation theories of imprisonment are in academic trouble (see, e.g., National Research Council 2014), even as their proponents seek ever more nuanced mechanisms to save them as an animating theory of criminal justice.

Meanwhile, public perception of crime—its trend and level, nature and distribution, and such important matters as likely victims and sensible methods for reducing exposure to risk—is unconnected to empirical reality. So, too, are public images of what the criminal justice process really is and the roles of critical actors, such as prosecutors and judges. Massive changes in the crime rate (e.g., years of decline since the 1990s) are not reflected in public perceptions. Polls indicate that representative samples of (p.134) adults routinely report that crime and violence are increasing, even as the data continue to show a persistent and large decrease in all crime, including violence, during the past 30 years in the United States. Media (including television and movies) depict a system that does not substantially exist, in which crimes are largely solved by technical experts with amazingly clever applications of “profiling” and other crime-fighting knowledge, instead of the strikingly more common banality of crime and its resolution.

These disjunctions between the facts about crime and crime policy, between academic and practitioner views about effective crime prevention, and between public images of crime and what crime is typically like are certainly not new. Indeed, modern history is replete with examples of huge changes in policies, practices, and ideologies about crime and its prevention. The reliance on the rehabilitative ideal as both a theory for the justice system and a way to structure its practices (e.g., parole and the indeterminate term in sentencing) became increasingly controversial until it was removed nearly wholesale from the criminal justice system in the 1980s and 1990s. That circumstance was supported by academicians who argued that rehabilitation did not and likely could not work and who favored a reliance on general deterrence instead (Wilson 1975). Today, rehabilitating offenders is undergoing rehabilitation as a justifying aim of the criminal law as evidence indicates some modest program effectiveness.

“Modern” policing at one time meant friendly, neighborhood policing rather than a highly selective and aggressive focus on likely offenders in “hot spots.” Increasing the age of potential juvenile court jurisdiction was once viewed as sensible, until waiver to adult court replaced it as a more fashionable way to deal with youth crime. Deterrence has been invoked in the recent past to justify substantially harsher punishments for drug use and sales and in an effort to reduce gun use in crime.

Harsh sentencing policies, such as “three strikes” laws, were justified on the idea that locking up such repetitive offenders could lower the crime rate (i.e., their selective incapacitative appeal). Such laws, however, were out of step with research and theory even at the outset of this policy trend. As was pointed out, they were contradicted by basic facts about age and crime (Gottfredson & Hirschi 1988). Today, at last, the consensus is that despite its intuitive appeal, selective sentencing is beyond the predictive capabilities of modern science (National Research Council 2014).

Theory-Based Policy

This book breaks from the tradition of justifying the nature of the criminal justice response by the presumptive deterrent, incapacitative, and (p.135) rehabilitative consequences of state-based punishment. An altogether different theory of the causes of crime and related social problems creates an altogether different set of expectations for state-based punishment for crime. In fact, there is an important, even necessary, relationship between a theory of criminal behavior and expectations for criminal justice. More important, theories of criminal behavior can indicate whether and to what extent criminal justice can effectively reduce crime and, as a consequence, can be useful in setting limits on criminal justice interventions. This is a thesis of this book—a thesis that is as old as the classical school of criminology and one that has profound consequences for the type of justice system we should have. To the extent that the criminal justice response to crime is justified by utilitarian or consequentialist theories (e.g., sanctions designed to lower the crime rate), then a true theory of crime causation can inform us about the limits of criminal justice responses. In Part II of this book, we use policy examples to illustrate the power of the theory to help us comprehend trends, effects, and the likely effects of policies.

What is needed to sort out all of the disparate trends in crime policy and the conflicting claims about the prospect for a criminal justice system to affect the crime rate is to abandon the view that ordinary crime and related problem behaviors can be very much affected by the utilitarian aims of the criminal law as it is necessarily practiced in a liberal democracy. We do not need more research seeking to show how to make the criminal justice system more responsive to theories of deterrence and incapacitation. Rather, what we need is a stronger focus—in policy, programs, and research—on the root causes of crime in the early years of life and how these causes interact with the formal mechanism of social control embedded in our justice system. Considerable research, both within and outside of the criminal justice tradition, demonstrates that deterrence and incapacitation theories central to the criminal justice system response rely on erroneous images of most crime and violence. What crime policy needs is a true theory of crime—one that recognizes that important general causes of crime and delinquency are established early in life, well before the formal justice system comes into play; that prevention of these causes is entirely possible with appropriate policy attention; and that the means to limit the effects of crime and delinquency offered by the media, politicians, police, and other agents of the criminal justice system will always have costly and relatively ineffective consequences. A focus on opportunities for crimes, linked explicitly with an understanding of the nature of crime and the research on self control, provides another avenue to reduce many offenses. A focus on the importance of age effects and the generality principle can help target prevention activities of the state, providing a basis for both limitations on the criminal justice response and advice about plausible effectiveness.

(p.136) In fact, it is a thesis of this book that a true theory of crime tells us what the limits of the criminal justice system are and offers a reason to restrain the natural appetite of the state apparatus for ever greater control and harsher punishments. A true theory of the causes of crime tells us much about what crime is like (the nature of crime), about the plausible effects and boundaries of the criminal justice system, and how to control the amount of crime that is occurring. It can serve as an important restraint on the historical tendency of the criminal justice apparatus to grow because of putative claims that more will do what less has been incapable of showing—apparently natural tendencies of common views of the theories of general deterrence and incapacitation.

We described in detail the theory in Part I and presented an empirically grounded perspective on the causes of crime. We are now in position to examine critically several important policy issues in criminal justice as examples of how a theory-driven view of policy will allow us to better understand the limits of the criminal justice system in controlling crime, consistent with adherence to other values important in liberal democracies.

We have previously made the central argument presented in this book (e.g., Gottfredson & Hirschi 1990, 1995). At that time, much of our argument was speculative because the scope and quality of research bearing on policy aspects were not well developed. But that has since changed substantially (Gottfredson 2006, 2011b, 2013). The scholarship from within criminology and from related disciplines, both in the United States and elsewhere, confirms that the causes of crime are mainly beyond the ken of the criminal justice system and identifies the reasons that it is so difficult to demonstrate effectiveness for the many well-meaning manipulations of law and policy directed at the formal criminal justice system. At the same time, research demonstrating the role of early life experiences for subsequent problem behaviors is compelling in demonstrating that prevention of crime and delinquency should be the focus of sound crime policy. The quality and amount of research on policing, sentencing, and opportunities for crime are stronger and provide evidence largely consistent with the expectations of our general theory. We can now review evidence and arguments about guns, drugs, immigration, policing, imprisonment, and early child care and understand what works and what does not and why. We begin with the fundamental argument, advanced by scholars of the criminal law at least since the 1800s, that a true theory of crime should inform us about how much we should rely on state intervention in lowering the crime rate.

(p.137) Crime Theory and Public Policy

There has always been tension in criminology (as with all sciences) about the interplay between basic theory and public policy. On the one hand, the principal evaluative criteria applied to theories include consistency with agreed-upon facts, parsimony, internal consistency, clarity, and scope. Whether or not appropriate public policies that are effective may be derived from them is not generally the theorists’ principal concern. A true theory may yield good or bad public policy (by many criteria) or, as is often the case, the public policy implications of basic theory may not be straightforward or easily deduced. On the other hand, criminology has a long (and troubling) tradition of explicit rejection of basic research and theory as a useful guide for public policy, at least since Wilson’s (1975) influential arguments about the lack of utility of work on “root causes” of crime for public policy. Arguably, the decades-long focus in American criminology on general deterrence, incapacitation, and policing as the most “sensible” policy choices for the reduction of crime has origins in this view. Some theoretical traditions in criminology, such as the deterrence doctrine and the “broken windows” idea, do, of course, have explicit policy foci, but they are often seen as the only theories with direct, practical implications by policymakers.

At the same time, theorists are justifiably wary of critiques based on the alleged policy implications of their basic work. Such implications are rarely straightforward, nearly always multifaceted, and need to be drawn using criteria other than simply effectiveness. Putative connections may be illegitimately drawn, based on false logic, or simply wrong. And in this arena, it is most likely that a variety of policy implications can be drawn, each with differing implications for choices that properly invoke criteria other than solely effectiveness. As a result, there is ample room for caution in thinking about the relation between basic theory and policy advice in the field.

Classical Criminology and Theory-Driven Public Policy

Our thesis for this book is the following: A valid theory of criminal behavior delimits the role of the state in the suppression of crime and the proper treatment of criminals. Such has been the assumption of major reformers of the criminal law for centuries. From their ideas about the causes of crime and delinquency, reformers from the classical school of criminology, such as Cesare Beccaria and Jeremy Bentham, derived the principle of legality, a (p.138) focus on the certainty, celerity, and severity of punishment and the concept of proportionality between harm and punishment. In their view, the aim of the criminal law was to facilitate the greatest happiness for the greatest number—an outcome that could be enhanced by instilling the fear of state punishments for undesirable behavior. Such an aim could be enhanced, given the assumption of rational calculation informed by hedonic calculus (maximize pleasure/avoid pain), by clear identification of the behaviors that were unwanted and by the provision of punishments in which the pain delivered by the state outweighed the pleasure derived from the behavior. The crime causation theory underscoring the state-punishment model was, of course, general deterrence.

In American literature, there have been many proponents of criminal law based on general deterrence, from legal scholars such as Norval Morris (1974), who applied the principle to sentencing, to political scientists such as James Q. Wilson (1975), who applied the ideas to policing, and economists such as Gary Becker (1974), who argued for enhanced deployment of deterrence theory generally in decision-making about crime policy. As Pratt et al. (2006) note, deterrence theory possesses two features that increase its appeal: (1) It offers an easily understood, straightforward explanation of crime (individuals commit crime when the benefits outweigh the costs); and (2) it offers an easily understood, straightforward solution to crime—heighten the costs of illegal conduct.

This book began with the premise that a portion of the classical view is correct—that the causes of crime must be connected to the ideas of criminal justice in fundamental ways if we are not to be disappointed in and misled about the consequences of the criminal justice process for crime reduction. Today, practitioners assume this connection, even though they seldom make explicit the crime causal theory involved in their policy proposals. So, leaders of policing claim responsibility for crime increases (not enough resources) or crime decreases (“smart policing” or informed deployment of resources); prosecutors and other politicians seek to appeal to the polity by complaining that sentencing changes could better protect the community from predation and violence; and correctional advocates point to manipulations in treatment regimens, “restorative” and reintegration mechanisms that might reduce recidivism. Policies ranging from individualization of treatment to substantial discretion for prosecutors and judges, separate systems for juveniles and adults, mandatory minimum or fixed prison terms, habitual offender statutes, criminal restrictions on alcohol and other drug use, selective or “focused” policing strategies, police in schools, and anti-gang programs are predicated (at least implicitly) on ideas about the causes of criminal behavior. It can be said that for much of (p.139) modern history, theory about the role of the state in crime control has been justified by criminological theory—by a belief that there is a fundamental connection between state response to antisocial behavior and the causes of crime.

Many of these ideas seem commonsensical: They are frequently taken for granted or seen to be obvious on their face. Everyone knows that fear can affect behavior if it is clear enough and sure enough (deterrence), and it is now well known that a relatively small proportion of offenders in any population will account for a very large share of all offenses committed. Why not selectively focus criminal justice resources on the chronic few, isolate those who repeatedly commit offenses, and punish violence harshly enough to persuade others away from it? The logic of these crime policies and the theory of crime they presuppose are so compelling that it is not coincidental that state involvement in criminal justice is, today, at unprecedented levels in modern history.

Crime Theory, Deterrence Theory, and the Classical Tradition in Criminology

Utilitarian or consequentialist social theorists, including the scholars in the “classical school” of criminology such as Jeremy Bentham and Cesare Beccaria, concerned themselves with the essential question of how society could best both maximize personal liberty and ensure individual safety. Because one major threat to personal liberty is crime, it is necessary to empower the government to put limits on the use of force and fraud in the pursuit of self-interest—that is, to sanction criminal conduct. However, they recognized the threats to individual liberty attendant on granting government such powers—powers designed to maximize the happiness of the greatest numbers via restrictions on individual harmful conduct. Once authorized to limit some individual liberties in order to protect the safety of the many, the central problem becomes the means to limit the extent of intervention by the state. In which behaviors should the state be authorized to intervene and with what force?

In other words, the problem for all criminal justice theories is to decide where the boundaries of intervention lie and when state force could be said to be excessive. For the classical scholars, who did not specifically question the effectiveness of state sanctions,1 the solution to the problem of limits was to be found in human nature itself, in the principles of utility. People act in such a way to maximize their pleasures and avoid pains. For Cesare Beccaria, the influential Italian classical theorist, the limits of state (p.140) punishment could be derived from the balance of advantages in criminal behavior:

For a punishment to attain its end, the evil which it inflicts has only to exceed the advantage derivable from the crime; in this excess of evil one should include the certainty of punishment and the loss of the good which the crime might have produced. All beyond this is superfluous and for that reason tyrannical. (Beccaria 1764/1995, 43)

Beccaria did not believe we should sacrifice

all liberty to the Leviathan in return for the protection it offered us. . . . On the contrary, he contended that we give up only the smallest portion of our liberty, i.e., that portion necessary for us to enjoy the remaining part in peace and tranquility. As a result, law was justified only to the extent that it was limited to what was required to preserve the maximum amount of individual liberty possible. (Bellamy 1995, 17–18)

Indeed, “He insisted that punishments should never be more painful than was necessary to prevent a given crime or outweigh in suffering the harm done to society by the misdemeanour they aimed to prevent” (Bellamy 1995, 25).

Jeremy Bentham, the great British utilitarian philosopher and social critic, agreed with Beccaria’s premise:

The general object which all laws have, or ought to have, in common, is to augment the total happiness of the community; and therefore, in the first place, to exclude, as far as may be, everything that tends to subtract from that happiness: in other words, to exclude mischief. But all punishment is mischief: all punishment in itself is evil. Upon the principle of utility, if it ought at all to be admitted, it ought only to be admitted in as far as it promises to exclude some greater evil. (Bentham 1798, 158)

Contrary to the way Bentham is sometimes interpreted, he did not rely solely on the notion of formal, legal systems of control but, rather, included elements of the broad range of human motivations for action, and his restraints included those central to the self-control perspective of our general theory of crime:

By utility is meant that property in any object, whereby it tends to produce benefit, advantage, pleasure, good, or happiness (all this in the present case comes to the same thing) or (what comes again to the same thing) to prevent the happening of mischief, pain, evil, or unhappiness to the party whose interest is considered. (Bentham 1798, 12)

(p.141) Thus, not suffering the confines of school is sometimes sufficient cause to truant, and the immediate cessation of an unpleasant argument can stimulate aggravated assault. The truant finds it easier to stay home and watch television or to sleep a little longer than to be bored by another day of school. The abusive adult gets his way (quieting a spouse or a child) more quickly with force than with reason. In every case, an opportunity for immediate gratification presents itself, and the actor is overwhelmed by the concerns of the moment. Our control theory focuses attention on the structure of these opportunities and on the development of the properties that shield people from yielding to temptations of the moment. Notice that the skill or training (or learning) required to engage in (and hence benefit from) the general run-of-the-mill antisocial behavior is minimal. As Bellamy (1995, 15) states, “The foundation of Beccaria’s theory was human nature and in particular ‘ineradicable human sentiments.’ ” From this, Bentham went on to derive the properties of punishments that affect the probability of crime (e.g., “variability,” “equability,” and “commensurability”) and behaviors unfit for criminal punishments. The theory embedded in this classical view has come down to us in criminology as deterrence theory—as the theory that crime will occur in the absence of governmental punishments. In this theory, the commission of crimes is evidence that something is wrong in the balance between the benefits of crime to the offender and the characteristics of state punishments—that the punishments are too uncertain, are too long delayed, are not well connected to the crime, or are too lenient.

To most people, the logic of consequentialism is compelling. All people calculate advantages and all people fear punishments (or at least can be made to fear them). The advantages of crime are, to most people, obvious. Thus, if crime rates are too high, or are increasing, modification of state penalties must be in order. But the deterrence theory derived from the classical school is only one of the theories implicit in the classical formulation. All classical theories begin with several strong assumptions: (1) All people are capable of criminal acts; (2) criminal acts require no special motivation, skills, or competencies; (3) self-interest motivates human conduct; (4) self-interest comprises both the pursuit of pleasure and the avoidance of pain; and (5) all groups value social conduct and condemn antisocial conduct (including crime). Thus, in classical theories, people conform their conduct to sanctions (do not act antisocially) if (1) they are aware of the sanctions, (2) they perceive of them as sanctions (i.e., they indeed are sanctions to the actor), and (3) the sanctions outweigh the benefits of the act.

Although these elements of classical theory are simple, seem reasonable, and are straightforward, important differences in the causal principles (p.142) underlying classical theory create different theories of crime and delinquency. The pleasure principle (seek pleasure, avoid pain) can be deployed to create different theories depending on what “pains” are pertinent to regulation of behavior and whether or not there are differences among people in the extent to which the pains are relevant. If the relevant pains are criminal justice sanctions (arrest, fines, and prison), the theory is deterrence; if the pains are disapproval of family and friends or educational or employment success, then the theory is what has come to be referred to as control theory (Gottfredson & Hirschi 1990; Hirschi 1969).

Both the classic school in criminology (e.g., Beccaria 1764/1993; Bentham 1798) and the deterrence perspective more recently (Packer 1968; Pratt et al. 2006) adopt the assumptions of the rational, calculating actor, weighing costs and benefits of contemplated acts. In the deterrence model, violations of the criminal law are an option for action, constrained by the potential consequences. The important features of these consequences are properties of the criminal sanction—especially the likelihood and the severity of the penalties imposed. In the most influential modern theoretical exposition of this view, economist G. Becker (1974) argues that people should be influenced by these in deciding whether to commit a crime, just as they are influenced in all other courses of action by consideration of costs and benefits. Variability in sanctions matter, much more than do variations among people in their attention to long-term costs and benefits. The criminal law defines crime for this view by definition, because once behavior is a violation of the criminal law, it is subject to penalties, and it is characteristics in these penalties that cause crime. Such a perspective has a good deal of commonsensical appeal and is the dominant view in the criminal justice process (Packer 1968).

Comparing Control Theory and Deterrence Theory

For control theories, on the other hand, individual differences in the tendency to weigh the long-term consequences of acts, whether violation of the criminal law or not, are highly significant. Penalties provided by the criminal justice system, such as arrest, fines, and incarceration, are typically considerably removed in time from the temptations of the acts themselves. Persons with high self control might well be influenced by such remote consequences (or, likely, so well controlled by social sanctions so as not to attend to criminal justice sanctions), but individuals relatively low in self control are not. If crimes provide quick, easy, and immediate benefits and little immediate cost, then, like other such acts, those with low self control will be more likely to undertake them, whatever the severity of the ultimate (p.143) punishment. At the same time, individuals with high self control attend to the noncriminal justice costs of crimes and delinquencies—shame, embarrassment, and loss of affection and regard by family, friends, colleagues, and community—more so than do those with relatively low self control. Because crimes and delinquencies bear such costs, they are likely to be more consequential to those with high self control than are criminal penalties, whatever their characteristics. Those with high self control are easily deterred, and thus the severity of sanctions would not be expected to matter much for them either.

This all means, of course, that the theory of deterrence by the criminal justice system and modern control theory make altogether different predictions about the effectiveness of (long-term) criminal justice sanctions or crime prevention efforts. For control theory, far-removed criminal sanctions, such as provided by criminal justice processing and punishment, whether certain or not, should have very little consequence for offending. Variations in saturation policing should be understood, in control theory terms, as efforts to effect short-term costs. Their effects should be limited to highly visible, intensive and intrusive “at the scene” or “hot spot” methods, and they should generally produce no lasting deterrence effects. In other words, deterrence by the criminal sanctions might work best where they are needed the least, and they might work least well where offending is most likely. They should work best when the costs are made highly intrusive and present in the moment, rather than remote and theoretical. They also are likely to run counter to other critical values for sanctions, such as fairness and proportionality. These predictions seem consistent with the evidence, as demonstrated in subsequent chapters.

Although the two theories share the elements of classical theory, they have a somewhat different image of the criminal actor; deterrence theory makes very little room for individual differences in responsiveness to sanctions, preferring instead to focus on aspects of sanctions that make them more or less effective (e.g., certainty and variability). In effect, all actors can be persuaded by long-term sanctions, such as those available to the state in a liberal democracy (i.e., subsequent to elaborate procedural rules designed to ensure fairness). In control theory, individual differences play a fundamental role in the effectiveness of various types of sanctions; there is great variation in the extent to which effective socialization has been accomplished, such that immediate costs and benefits dominate among some individuals much of the time. These people are less able to defer action in the interest of long-term benefits and costs. Focusing criminal justice sanctions on them will be both ineffective in reducing crime and inconsistent with values of fundamental fairness and proportionality.

(p.144) A second, but highly related, difference between deterrence and control theories relates to the power that different forms of sanctions have with regard to governing individual conduct. Deterrence theory focuses on long-term costs and thus favors political and economic sanctions. Indeed, it might be said that deterrence theory, at least as it has come to be practiced, is a governmental control theory. Control theories focus on short-term costs and thus favor natural (or physical) sanctions (barriers) and personal and informal sanctions (e.g., those available to parents and friends).

We argue that control theories are able to predict the research literature on crime prevention, whereas deterrence theories cannot. This is so because, in the language of control theory, effective crime prevention techniques simply take advantage of the power of physical or natural sanctions and thus dissuade those with little self control from crime. Criminal justice sanctions are ineffective, save those provided by saturation policing or other highly intrusive methods.

The general similarities between deterrence and control theories have resulted in confusion about the predictions each makes concerning public policy about crime. Having abandoned the idea that long-term costs are effective sanctions, control theories have divorced themselves from many of the policy prescriptions of the classical theories. The implication of this fact has not been previously explored.

We are convinced that the elements of control theory are essentially correct and, as a result, the implications for the criminal sanction are profound. We are also convinced that a correction concerning the direction that utilitarian theory should take is timely. The emphasis on state sanctions and power has been misplaced. Indeed, as discussed later, the empirical work doubts the influence of such sanctions. In the policy world, in which the basic theory rings true, the consequence is to do more of the same—much more. In the academic world, the consequence is to dispute theories of rational choice and self-interest, or to severely limit their scope. But when appropriately conceived, more rather than less behavior falls within the province of control theories, and they provide strong guidance about the limits of the criminal sanction.

We are thus in position to review research on specific public policies about crime, contrasting the criminal justice view with the control theory view. Our aim in Part II is to show that many of the puzzles about the effectiveness or lack thereof of the criminal justice system are anticipated by the theory of self control, and we can reconcile a large literature by applying its principles and logic to the results of research. We think the scope for such an exercise is very large, spanning the range of juvenile justice and criminal justice policies and practices. Although we argue that control theory provides guidance to most contemporary policies about juvenile and (p.145) criminal justice, within the scope of this book, we seek to illustrate the applicability of this approach with a few important criminal justice examples in the chapters that follow.2 We hope thereby to show that application to other policies can be straightforward.

The majority of research studies centered on self-control and social control theories have been basic in nature, and the issues studied have been quite wide in scope. However, in addition to this basic research, it has been argued that high-quality evaluation studies may be providing evidence about the validity of control theories (Gottfredson 2006). The central focus of modern control theories on early childhood and the role of socialization has provided a connection between control theories, research on families, and suggestions that crime policy could be more effective by placing a focus on early prevention (Gottfredson & Hirschi 1990, 1995; Junger-Tas 1997).

The past several years have witnessed an increase in the number of high-quality studies of prevention, using random allocation designs, suitable follow-up periods, and multiple indicators for outcomes. These span a range of efforts, from systematic interventions very early in life aimed at reducing problem behaviors to large-scale studies of variation in styles of policing. Many of these have strong implications for the validity of basic theory. At the same time, considerable evidence has accumulated about the consequences of the very substantial increases in incarceration in the United States on crime rates and about the connection between long-term sanctions and individual behavior.

Elements of Control Theory Suggest Prevention: A Brief Review for Policy

Control theories assume rational actors, seeking to maximize benefit and minimize harm. They are restraint theories, such that motivation to engage in delinquency is not seen to be problematic; rather, differences in the probability of delinquent conduct are seen to be a function of differential controls. The number and intensity of these controls affect the probability that an actor will pursue short-term benefits in the face of subsequent negative consequences. The relations between family and friends, schooling, and employment are viewed as essential in this calculus. Caring about the wishes of family and friends and also caring about with wishes of teachers and others stand as restraints to delinquency and crime. It is theorized that these restraints must be taught and that they are most effectively taught early in life, at which time they become a strong basis for self control. Variation in self control is then associated with many behaviors that affect life chances (p.146) throughout the life course, including delinquency, problem behaviors, and school and employment success.

A focus on early childhood socialization and on the family provides this version of control theory with a clear set of alternative expectations for public policy to theories of deterrence and policing. General deterrence presumes that fear of criminal justice sanctions, which are most often long removed in time from the act, can affect the calculus to offend. The focus on variation in long-term sanctions places deterrence theory at odds with the control theories described here. In addition, the continuing role thought by control theory to be played by early childhood socialization and the development of self-control stands as a significant difference between the rational actor expectations of general deterrence theory and those of control theory.

To the extent that a key causal factor in delinquency and crime is a relative lack of focus on costs associated with behaviors that provide immediate gains, sanctions provided by the juvenile justice and criminal justice system cannot be expected to impact much on the rates of crime and delinquency. That is, control theories that focus on the restraining influences created by interpersonal relationships in childhood suggest that the generally far-removed sanctions of the justice system are largely irrelevant—they are unnecessary for those who already are restrained by social bonds and are irrelevant to those for whom long-term sanctions of any type are not the focus of attention. Control theories thus provide a clear basis for predicting that deterrence provided by police and that provided by incarceration are problematic methods of crime prevention. Although the certainty, and especially the celerity, of sanctions should play a role, variation in punitiveness in the mostly time-removed sanctions of the juvenile and criminal justice systems does not. Whether a sentence to incarceration is for 6 months or 6 years may well not matter if there is little impact of distant sanctions in the first place. Thus, studies of the effectiveness of these prevention methods are, in effect, also studies pertinent to the validity of control theory.

Deterrence Research on Criminal Justice Sanctions Is Consistent with Control Theory

Reviews of the deterrence literature studying the actual general effect of policing and imprisonment on crime show that it is difficult to discover consistent, strong relations between actual deterrence variables and crime rates, particularly associated with the severity of sanctions (for general discussion (p.147) of this point, see Chapters 8 and 9, this volume, and Gottfredson 2011b). For example, Blumstein and Wallman’s (2000, 480) conclusion based on their review of the empirical data about the deterrent and incapacitative effects of imprisonment is unambiguous: “There has been a massive growth in prison populations since the early 1970’s with no comparable effect on crime rates” (see also Walker 2015; Webster, Doob, & Zimring 2006). The findings of the National Research Council’s (2014) report on incarceration are consistent with the conclusion of negligible effects on crime of criminal justice sanctions (see our discussion in Chapter 8, this volume).

An excellent comprehensive review of the perceptual deterrence literature is provided by Pratt (2009), with results also consistent with control theory expectations. Based on a meta-analysis of 40 studies on perceptual deterrence, Pratt concluded that deterrence-based predictors were “at best, only weakly associated with criminal behavior” and that “support for the deterrence perspective is most likely to be found in studies that are methodologically the weakest of the bunch” (71).

Studies of the Effects of Childhood Interventions Are Consistent with Control Theory

A burgeoning research literature based on strong research designs now clearly supports the idea that substantial and lasting prevention effects can be achieved by affecting early childhood experiences in ways designed to enhance socialization and monitoring. In stark contrast to the difficulty in finding prevention effects from criminal justice sanctions, the literature on early childhood prevention is encouraging. A good deal of systematic research on the preventative effects of enhancing early childhood environments has been accumulating across disciplines, from education to criminology. For example, Webster-Stratton and Taylor (2001) reviewed childhood prevention programs designed to promote parent and teacher competencies to prevent conduct problems that have evaluations published in peer-reviewed journals, with effects demonstrated in randomized controlled trials with alternative treatments. They discovered a number of different successful programs, including intensive home visiting for poor mothers and behavioral parent skill training.

Schindler and Black (2015) reviewed programs that focus on the first 5 years of life and found programs that intensively target children’s social skills and self-regulation and those that target adult caregivers’ skills in behavior management are particularly promising. Greenwood’s (2006) careful review provides a classification of six types of effective programs, (p.148) ranging from home visits by nurses to parent training (see also Eckenrode et al. 2001; Olds et al. 1998).

Piquero and colleagues (2009, 2016) performed meta-analyses of studies of parenting undertaken with children younger than age 5 years. In the 78 studies meeting their criteria for inclusion, and using self-report criteria for delinquency, they report a mean effect size of .37. In a companion review, Piquero, Jennings, and Farrington (2010) focused on self-control training in random design studies (N = 34) that sought self-control improvement among young children. They conclude that not only was it possible to systematically alter self control but also these interventions reduced delinquency.

Some important experiments in education have focused on early interventions with parents, providing enhanced training associated with the creation of self control. Some of these efforts find substantial effects not just for schooling but also for the reduction of delinquency. An example from education for adolescents is provided by Fosco et al. (2013), who employed a “family checkup model” designed to improve educational attainment. Their randomized controlled trial using family management training for adolescents showed increases in self-regulation and lower levels of antisocial behavior, deviant friendships, and substance abuse. Heckman (2006) found an array of early childhood education research to bolster his argument that family environments variously foster skills helpful to avoiding delinquency and crime, achieving better health, as well as achieving school and workplace success. Heckman reports analysis from randomized controlled experiments associated with Head Start programs that provided very young children from disadvantaged backgrounds (aged 3 or 4 years) with nurturing environments and then followed them to age 40 years. The treated group had greater high school graduation rates, higher salaries, more home ownership, fewer out-of-wedlock births, and fewer arrests. He calculated the benefits:cost ratio as 8:1, with advantages in education, criminal justice, and health. His general review leads him to stress early intervention, concluding that “on average, the later remediation is given to a disadvantaged child, the less effective it is” (Heckman 2007, 13251).3

Basic Theory and Crime Policy

Policy work in criminology now strongly shows the hazards implicit in eschewing basic research and theory in the design and study of policy about delinquency and crime—and that, once again, nothing may be more practical than a good theory.

(p.149) A major assumption of control theory is that it will never do to turn to the state criminal justice apparatus to ameliorate the problems of antisocial conduct. For control theory, evidence of such massive state intervention in the crime problem—whether the form is more police, greater use of imprisonment, “alternative sanctions,” or more intrusive involvement in personal tastes and habits (e.g., alcohol and drugs)—is evidence that natural and powerful social control methods have been attenuated. At the social or “macro” level, control theory anticipates a negative relationship between the presence of effective informal control systems and the size and scope of the criminal justice system. When informal, family and friendship systems appear weakened; when crime and violence appear to increase unabated, state systems of sanctions grow. In other words, wherever it exists, state criminal justice is evidence of weakened social control, not a desirable route by which it can be achieved. In addition, from the control theory perspective, whether they acknowledge the theory on which they operate or not, policy scholars who turn to the state to reduce the level of delinquency, crime, and related problem behaviors do so at the peril of ignoring basic causes of crime and violence.

Modern control theories of crime are not complete theories of justice. That is, they are not designed to provide guidance about all appropriate uses of the criminal sanction. Certainly, the criminal sanction can be justified on numerous grounds, from retribution and reprobation to deterrence, rehabilitation, restoration, and incapacitation. All societies choose to condemn and to control some conduct by means of the criminal sanction, but at some times and in some places, there is an attempt to control some conduct by the criminal law that is not so controlled at other times and in other places, such as alcohol and drug use. For this reason, among others, our control theory defines “crime” (the dependent variable for the theory) independent of the criminal law. Much of the behavior proscribed by the criminal law will fall within the theory’s definition and some will not; some behavior not criminalized will be within the intended scope of the theory. By definition, then, much of what goes on in the criminal law and in the reform of the criminal law is not particularly relevant to the theory, and the theory is not particularly relevant to every practice or reform.

For example, the justice system continuously adjusts the penalties for criminal acts. When such adjustments are undertaken for the purpose of enhancing the deterrent effects of the law or to increase the incapacitative effect of sanctions, the domains of control theory and of the criminal law overlap and the effort may be said to be either correct or mistaken from the viewpoint of the theory. But often, penalties in the criminal law are adjusted without reference to crime control purposes—to condemn or to express reprobation. In such instances, the domains do not directly overlap. The (p.150) action is not right or wrong, according to the theory; it is not evidence about the theory one way or another; and it is therefore not legitimate to invoke the theory in reference to the action. However, the role of the state in crime control simply cannot be delimited by means other than a valid theory of the causes of crime. Classical theories of criminal behavior—those relying on the image of self-interested, hedonic actors who limit the pursuit of personal pleasure when forced to do so by negative sanction—were formulated precisely for such a purpose. Modern control theory, formulated on a similar image and nurtured on five decades of behavioral research, has important, untapped potential to serve the very same purpose.


(1) Although we read Bentham as decidedly hesitant about the effectiveness of both state and religious sanctions in favor of social sanctions.

(2) We have done this in the past, such as in Gottfredson and Hirschi (1993), in which we argued that application of our theory to the distinction between juvenile and adult justice systems would favor treating adults more akin to how we treat juveniles, with an attendant massive reduction in adult incarceration (see the discussion in Chapter 3). An excellent essay cerntered on the implications of the general theory of crime for public policy is provided by Piquero (2010).

(3) For a general discussion of interventions creating skills that improve self control, see Heckman and Kautz (2014).