Theories Evaluations

Criminological_Theories_Introduction_and_Evaluatio..._----_CHAPTER_2_DETERRENCE_AND_RATIONAL_CHOICE_THEORIES.pdf

Home >Law homework help >Theories Evaluations

CHAPTER 2

DETERRENCE AND RATIONAL CHOICE THEORIES

Classical Criminology and the Deterrence Doctrine

Classical criminology refers primarily to the eighteenth century writings of Cesare Beccaria in Italy and Jeremy Bentham in Eng

land. 1 Both were utilitarian social philosophers who were primarily concerned with legal and penal reform rather than with formulating an explanation of criminal behavior. In doing so, however, they formu lated a theory of crime that remains relevant to criminology today.2

The system of law, courts, and penalties of the day that the classical criminologists wanted to change was marred in most European coun tries by arbitrary, biased, and capricious judicial decisions. It was com mon to use torture to coerce confessions and to inflict cruel punishments, including whipping, public hanging, and mutilation. The classical criminologists were intent on providing a philosophical ra tionale for reforming the judicial and legal system to make it more rational and fair. Their ideas converged with the developing interests of the rising middle classes of merchants and the economic philosophy promoting trade, commerce, and industry. They promoted reforms which many of the leading intellectuals of the day were advocating. Their arguments also fit well with developing political movements seek ing greater citizen participation and democratic control of government. Many of the law reforms proposed by classical utilitarian philosophers, such as doing away with cruel and unusual punishment and instituting the right to a speedy trial, were incorporated into the Constitution of the United States in its B i l l of Rights amendments. Others, such as a legislatively fixed scale of punishment for each type and degree of crime, were incorporated into the new legal codes of France in 1791, following the French Revolution.

CHAPTER 2

DETERRENCE AND RATIONAL CHOICE THEORIES

Classical Criminology and the Deterrence Doctrine

Classical criminology refers primarily to the eighteenth century writings of Cesare Beccaria in Italy and Jeremy Bentham in Eng

land.1 Both were utilitarian social philosophers who were primarily concerned with legal and penal reform rather than with formulating an explanation of criminal behavior. In doing so, however, they formu lated a theory of crime that remains relevant to criminology today.2

The system of law, courts, and penalties of the day that the classical criminologists wanted to change was marred in most European coun tries by arbitrary, biased, and capricious judicial decisions. It was com mon to use torture to coerce confessions and to inflict cruel punishments, including whipping, public hanging, and mutilation. The classical criminologists were intent on providing a philosophical ra tionale for reforming the judicial and legal system to make it more rational and fair. Their ideas converged with the developing interests of the rising middle classes of merchants and the economic philosophy promoting trade, commerce, and industry. They promoted reforms which many of the leading intellectuals of the day were advocating. Their arguments also fit well with developing political movements seek ing greater citizen participation and democratic control of government. Many of the law reforms proposed by classical utilitarian philosophers, such as doing away with cruel and unusual punishment and instituting the right to a speedy trial, were incorporated into the Constitution of the United States in its Bill of Rights amendments. Others, such as a legislatively fixed scale of punishment for each type and degree of crime, were incorporated into the new legal codes of France in 1791, following the French Revolution.

Akers, Ronald L.. Criminological Theories : Introduction and Evaluation, Taylor & Francis Group, 1999. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/usf/detail.action?docID=1273212. Created from usf on 2023-03-02 13:30:58.

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

16 Criminological Theories: Introduction and Evaluation

Deterrence: Certainty, Severity, and Celerity of Punishment

Severity and Fitting the Punishment to the Crime The basic premise in classical criminology is that actions are taken

and decisions are made by persons in the rational exercise of free wil l . Al l individuals choose to obey or violate the law by a rational calculation of the risk of pain versus potential pleasure derived from an act. In contemplating a criminal act, they take into account the probable legal penalties and the likelihood that they will be caught. If they believe that the legal penalty threatens more pain than the probable gain produced by the crime, then they will not commit the crime. Their calculation is based on their own experience with criminal punishment, their knowl edge of what punishment is imposed by law, and their awareness of what punishment has been given to apprehended offenders in the past. (See the discussion of specific and general deterrence below.)

A legal system that is capricious and uncertain does not guarantee sufficient grounds for making such rational decisions. Such a system is not only unjust, it is also ineffective in controlling crime. In order to prevent crime, therefore, criminal law must provide reasonable penal ties which are applied in a reasonable fashion to encourage citizens to obey rather than violate the law. The primary purpose of criminal law is deterrence. It should not be used simply to avenge the wrongs done to the state or the victim. The legislators enact laws that clearly define what is unlawful, prescribe punishment for law violation sufficient enough to offset the gain from crime, and thereby deter criminal acts by citizens. Judges should do no more than determine guilt or innocence and should use no discretion to alter penalties provided for by law.

The punishment must "fit the crime." This may be interpreted as retribution: an eye for an eye, a tooth for a tooth. But to Bentham and Beccaria, fitting the punishment to the crime meant more than making the punishment proportional to the harm caused to society. It meant that the punishment must be tailored to be just severe enough to over come the gain offered by crime. Punishment that is too severe is unjust, and punishment that is not severe enough will not deter.

The assumption behind this argument is that the amount of gain or pleasure derived from committing a particular crime is approximately the same for everyone. Therefore, making the punishment fit the crime stands in contrast to the punishment fitting the individual. The law should strictly apply the penalty called for a particular crime, and the penalty should not vary by the characteristics or circumstances of the offender. The argument also assumes that the more serious or harmful the crime, the more the individual stands to gain from it; therefore, the more serious the crime, the more severe the penalty should be to deter it. In classical criminology, this concept of proportionality meant that the legislature should enact an exact scale of crimes with an exact scale

16 Criminological Theories: Introduction and Evaluation

Deten-ence: Certainty, Severity, and Celerity of Punishment

Severity and Fitting the Punishment to the Crime The basic premise in classical criminology is that actions are taken

and decisions are made by persons in the rational exercise of free will. All individuals choose to obey or violate the law by a rational calculation of the risk of pain versus potential pleasure derived from an act. In contemplating a criminal act, they take into account the probable legal penalties and the likelihood that they will be caught. If they believe that the legal penalty threatens more pain than the probable gain produced by the crime, then they will not commit the crime. Their calculation is based on their own experience with criminal punishment, their knowl edge of what punishment is imposed by law, and their awareness of what punishment has been given to apprehended offenders in the past. (See the discussion of specific and general deterrence below.)

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

17 Deterrence and Rational Choice Theories

of threatened punishment, without regard to individual differences. This was later modified to consider that age and mental capacity may affect one's ability to reason rationally.

Certainty and Celerity of Punishment The deterrence doctrine does not rest on the severity of legal penalties

alone. It further determines that, in order to deter, punishment for crime must be swift and certain. Celerity refers to the swiftness with which criminal sanctions are applied after the commission of crime.

The more immediately after the commission of a crime a punishment is inflicted, the more just and useful it will be. . . . An immediate pun ishment is more useful; because the smaller the interval of time between the punishment and the crime, the stronger and more lasting will be the association of the two ideas of crime and punishment. (Beccaria, 1972:18-19)

Certainty refers to the probability of apprehension and punishment for a crime. If the punishment for a crime is severe, certain, and swift, the citizenry wil l rationally calculate that more is to be lost than gained from crime and wil l be deterred from violating the law. Both Beccaria and Ben- tham saw a connection between certainty and severity of punishment. Certainty is more effective in deterring crime than severity of punishment. The more severe the punishment, the less likely it is to be applied; and the less certain the punishment, the more severe it must be to deter crime.

Specific and General Deterrence There are two ways by which deterrence is intended to operate. First,

apprehended and punished offenders will refrain from repeating crimes if they are certainly caught and severely punished. This is known as "specific deterrence" or "special deterrence." Second is "general deter rence," in which the state's punishment of offenders serves as an exam ple to those in the general public who have not yet committed a crime, instilling in them enough fear of state punishment to deter them from crime (Zimring, 1971; Zimring and Hawkins, 1973).

Modern Deterrence Theory

Studies of Deterrence The principles of certainty, severity, and celerity of punishment, pro

portionality, and specific and general deterrence remain at the heart of modern deterrence theory (Zimring and Hawkins, 1973; Gibbs, 1975; Wright, 1993b). Furthermore, the deterrence doctrine remains the philosophical foundation for modern Western criminal law and crimi nal justice systems. The policy implications of deterrence theory evolved from the interest in changing the judicial and penal policy of

Deterrence and Rational Choice Theories 17

of threatened punishment, without regard to individual differences. This was later modified to consider that age and mental capacity may affect one's ability to reason rationally.

Certainty and Celerity of Punishment The deterrence doctrine does not rest on the severity oflegal penalties

The more immediately after the commission of a crime a punishment is inflicted, the more just and useful it will be .... An immediate pun ishment is more useful; because the smaller the interval of time between the punishment and the crime, the stronger and more lasting will be the association of the two ideas of crime and punishment. (Beccaria, 1972:18-19)

Certainty refers to the probability of apprehension and punishment for a crime. If the punishment for a crime is severe, certain, and swift, the citizenry will rationally calculate that more is to be lost than gained from crime and will be deterred from violating the law. Both Beccaria and Ben tham saw a connection between certainty and severity of punishment. Certainty is more effective in deterring crime than severity of punishment. The more severe the punishment, the less likely it is to be applied; and the less certain the punishment, the more severe it must be to deter crime.

Specific and General Deterrence There are two ways by which deterrence is intended to operate. First,

Modern Deterrence Theory

Studies of Deterrence The principles of certainty, severity, and celerity of punishment, pro

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

18 Criminological Theories: Introduction and Evaluation

the eighteenth century, and the theory continues to attract many ad herents today because of its direct applicability to policy on law en forcement, courts, and imprisonment. The most common policy reaction to crime problems is to call for increased penalties, more severe sentences, additions to the police force so more arrests can be made, and the increased certainty of conviction and sentencing. These trends are directly related to all the efforts by legislators to make criminal penalties more certain and severe, to reduce the recidivism of already punished offenders, and to deter new offenders. A policy of longer sentences, especially when selectively applied to habitual offenders, may also be based on the premise that imprisonment, even when it does not deter, will at least incapacitate offenders for a period of time (Blumstein et al., 1978). But the deterrence potential is always behind the policy on all criminal sanctions, from the death penalty on down.

In spite of the long history and continuing importance of deterrence theory, empirical research designed to test it was rare until the late 1960s. Prior to that, most discussions of deterrence revolved around the humanitarian, philosophical, and moral implications of punish ment rather than the empirical validity of the theory (Ball, 1965; Toby, 1964; Gibbs, 1975). Since 1970, however, deterrence has been one of the most frequently discussed and researched theories in criminology (see Gibbs, 1975; Tittle, 1980; Wright, 1993b; Stitt and Giacopassi, 1992).

The first studies on deterrence consisted primarily of comparisons between states which provided capital punishment for first-degree homicide and those which had no death penalty. The early studies also examined homicide rates in states before and after they abolished capi tal punishment. These studies found that the provision or absence of the death penalty in state statutes had no effect on the homicide rate (Sellin, 1959; Bedau, 1964). Research by Gibbs (1968), Tittle (1969), and Chiricos and Waldo (1970) set the stage for many of the studies that followed and have continued to this day. Their studies moved beyond the effects of the death penalty to test the deterrent effect of certainty and severity of punishment on a whole range of criminal and delinquent offenses. They did not include measures for celerity of pun ishment, however, and it has seldom been included in deterrence re search ever since.

Objective Measures of Deterrence Deterrence research measures the severity and certainty of criminal

penalties in two ways. The first approach is to use objective indicators from official criminal justice statistics. The certainty or risk of penalty, for instance, is measured by the arrest rate (the ratio of arrests to crimes known to the police) or by the proportion of arrested offenders who are prosecuted and convicted in court. The severity of punishment may

18 Criminological Theories: Introduction and Evaluation

Objective Measures of Deterrence Deterrence research measures the severity and certainty of criminal

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

19 Deterrence and Rational Choice Theories

be measured by the maximum sentence provided by law for an offense, by the average length of sentence for a particular crime, or by the proportion of convicted offenders sentenced to prison rather than to probation or some other non-incarceration program. Deterrence theory predicts an inverse or negative relationship between these official meas ures of legal penalties and the official crime rate measured by crimes known to the police. When the objective certainty and severity of crimi nal sanctions are high, according to the theory, official crime rates should be low (Gibbs, 1968; 1975; Tittle, 1969; 1980; Chiricos and Waldo, 1971; Ross, 1982).

Perceptual Measures of Deterrence The second approach is to measure individuals' subjective percep

tions of legal penalties. The objective threat of legal punishment means nothing if citizens are not aware of the official sanctions or do not believe that there is any high risk of penalty if they were to commit a crime. In fact, most people have a very limited knowledge of what the legal penalties actually are and often make very inaccurate estimations of the true odds of apprehension and incarceration. But a person's fear of punishment should have a deterrent effect on his or her decision to violate the law, even if that fear has no connection with objective reality. Ultimately, deterrence theory proposes that it is what people believe about the certainty, severity, and swiftness of punishment, regardless of its true risks, that determines their choice of conformity or crime.

Recognizing this crucial cognitive dimension of deterrence, re searchers have utilized "subjective" measures of the risks and severity of legal penalties as perceived by individuals. This is measured, for example, by asking respondents on questionnaires or in interviews questions such as, "How likely is it that someone like you would be arrested if you committed X?" Most research on deterrence since the 1970s has used these perceptual measures, typically relating individu als' perceptions of risk and severity of penalties to their self-reported delinquency and crime. 3 The higher the risks of apprehension and the stiffer the penalties for an offense perceived by individuals, the theory predicts, the less likely they are to commit that offense.

Do Criminal Sanctions Deter? If there were no criminal justice system and no penalties provided

by law for harmful acts against others and society, it would be obvious that laws prohibiting certain behavior would carry no threat for viola tion. The laws could maintain some moral suasion, and most would probably still obey the law and refrain from predatory acts. But law lessness would be more rampant than it is now. Indeed, a formal control system of laws and government is essential to social order in a modern political state. In this sense, the mere existence of a system that provides

Deterrence and Rational Choice Theories 19

Perceptual Measures of Deterrence The second approach is to measure individuals' subjective percep

Recognizing this crucial cognitive dimension of deterrence, re searchers have utilized "subjective" measures of the risks and severity of legal penalties as perceived by individuals. This is measured, for example, by asking respondents on questionnaires or in interviews questions such as, "How likely is it that someone like you would be arrested if you committed X?" Most research on deterrence since the 1970s has used these perceptual measures, typically relating individu als' perceptions of risk and severity of penalties to their self-reported delinquency and crime.3 The higher the risks of apprehension and the stiffer the penalties for an offense perceived by individuals, the theory predicts, the less likely they are to commit that offense.

Do Criminal Sanctions Deter? If there were no criminal justice system and no penalties provided

by law for harmful acts against others and society, it would be obvious that laws prohibiting certain behavior would carry no threat for viola tion. The laws could maintain some moral suasion, and most would probably still obey the law and refrain from predatory acts. But law lessness would be more rampant than it is now. Indeed, a formal control system of laws and government is essential to social order in a modem political state. In this sense, the mere existence of a system that provides

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

20 Criminological Theories: Introduction and Evaluation

punishment for wrongdoing deters an unknown amount of crime. This effect of the chance of punishment versus no punishment at all has been referred to as absolute deterrence (see Zimring and Hawkins, 1973; Gibbs, 1975; see also Wright, 1993b).

However, absolute deterrence is not the relevant issue in deterrence research. Most people, most of the time and under most circumstances, conform to the law because they adhere to the same moral values as those embodied in the law, not because they are worried about impris onment. We do not steal and kil l because we believe it is morally wrong. We have been educated and socialized to abhor these things. Our so cialization comes from the family, church, school, and other groups and institutions in society; and partly from the educative effect of the law itself, simply by its formal condemnation of certain acts (Andenaes, 1971; see Gibbs, 1975, for a review of other preventive effects of law beyond deterrence). Therefore, the important question that research on deterrence attempts to answer is, does the actual or perceived threat of formally applied punishment by the state provide a significant mar- ginal deterrent effect beyond that assured by the informal control sys tem? (Gibbs, 1975; Zimring and Hawkins, 1968)

The best answer seems to be yes, but not very much. Studies of both objective and perceptual deterrence often do find negative correlations between certainty of criminal penalties and the rate or frequency of criminal behavior, but the correlations tend to be low. Severity of pun ishment has an even weaker effect on crime, whether among the main body of criminal offenders (Smith and Akers, 1992) or among a special category such as those convicted of white-collar offenses (Weisburd et al., 1995). Neither the existence of capital punishment nor the certainty of the death penalty has ever had a significant effect on the rate of homicides. Findings show that there is some deterrent effect from the perceived certainty of criminal penalties, but the empirical validity of deterrence theory is limited.

Deterrence and the Experiential Effect Paternoster et al. (1983; 1985) take a more critical stance toward

deterrence theory. They maintain that even a conclusion of modest support for deterrence theory is a misinterpretation of the findings of negative correlations between the perceptions of sanctions and the deterrence of offenses. They support their argument by referring to the fact that much of deterrence research is "cross-sectional." Cross-sec tional research means that the perceptions of risk and the self-reported offenses are measured during the same time period. Indeed, in some studies, the respondents have been asked about current perceptions of deterrence, while questions about offenses have included violations that occurred in the past. Consequently, this research often does not

20 Criminological Theories: Introduction and Evaluation

However, absolute deterrence is not the relevant issue in deterrence research. Most people, most of the time and under most circumstances, conform to the law because they adhere to the same moral values as those embodied in the law, not because they are worried about impris onment. We do not steal and kill because we believe it is morally wrong. We have been educated and socialized to abhor these things. Our so cialization comes from the family, church, school, and other groups and institutions in society; and partly from the educative effect of the law itself, simply by its formal condemnation of certain acts (Andenaes, 1971; see Gibbs, 1975, for a review of other preventive effects of law beyond deterrence). Therefore, the important question that research on deterrence attempts to answer is, does the actual or perceived threat of formally applied punishment by the state provide a significant mar ginal deterrent effect beyond that assured by the informal control sys tem? (Gibbs, 1975; Zimring and Hawkins, 1968)

Deterrence and the Experiential Effect Paternoster et al. (1983; 1985) take a more critical stance toward

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

21 Deterrence and Rational Choice Theories

ascertain whether the respondents' perceptions of risk came before or after they committed the offenses. Sometimes, the offenses preceded the reported perceptions of risk.

Paternoster et al.'s own longitudinal study, which measured percep tions of the certainty of legal punishment before and after offenses, was able to show that "the effect of prior behavior on current perceptions of the certainty of arrest, the experiential effect, is stronger than the effect of perceptions of certainty on subsequent behavior, the deterrent effect" (Paternoster et al., 1983:471; emphasis added). Those respon dents with "little prior experience in committing an offense have higher estimates of the certainty of punishment than those with experience" (Paternoster et al., 1985:429). In other words, the more frequently re spondents have been involved in law violations in the past, the lower their perceived risk of sanctions in the present. Paternoster et al. (1983) concluded that findings of negative associations between perceived risks and criminal behavior reflect more the "experiential" effect of behavior on the perceptions of risk than the "deterrent" effect.

Paternoster et al. (1983) are correct in concluding that the correlation they found between the perceptions of risk and subsequent offenses are too weak to validate deterrence theory. They fail to recognize, how ever, that their findings on the experiential effect may not in fact con tradict the principle of specific deterrence. If respondents had previously committed offenses, but had not been punished for them, the principle of specific deterrence would subsequently predict a low level of perceived certainty. Individuals who are involved in repeated crimes without suffering punishment should have lower perceptions of risk, since they have gotten away with it so often. This in turn should be related to repeating offenses in the future (Stafford and Warr, 1993). Specific deterrence is supposed to operate based on persons getting caught and punished for criminal acts. If they are not, the theory argues, then they will come to believe that the certainty of punishment is low. In this sense, then, deterrence theory predicts the very experiential effect that Paternoster et al. found.

It would be contrary to the principle of specific deterrence if the research had found that respondents who reported frequent arrests in the past still had perceptions of a low risk of criminal sanctions. Pater noster et al. asked only about prior behavior. They did not measure past experience with arrest and punishment, so we do not know the extent to which their finding of an experiential effect contradicts deterrence theory.

Modifications and Expansions of Deterrence Concepts Another aspect of this study by Paternoster et al. (1983; 1985) points

to the movement by many researchers to expand deterrence concepts

Deterrence and Rational Choice Theories 21

ascertain whether the respondents' perceptions of risk came before or after they committed the offenses. Sometimes, the offenses preceded the reported perceptions of risk.

Paternoster et al.'s own longitudinal study, which measured percep tions of the certainty of legal punishment before and after offenses, was able to show that "the effect of prior behavior on current perceptions of the certainty of arrest, the experiential effect, is stronger than the effect of perceptions of certainty on subsequent behavior, the deterrent effect" (Paternoster et al., 1983:471; emphasis added). Those respon dents with "little prior experience in committing an offense have higher estimates of the certainty of punishment than those with experience" (Paternoster et aI., 1985:429). In other words, the more frequently re spondents have been involved in law violations in the past, the lower their perceived risk of sanctions in the present. Paternoster et al. (1983) concluded that findings of negative associations between perceived risks and criminal behavior reflect more the "experiential" effect of behavior on the perceptions of risk than the "deterrent" effect.

Paternoster et al. (1983) are correct in concluding that the correlation they found between the perceptions of risk and subsequent offenses are too weak to validate deterrence theory. They fail to recognize, how ever, that their findings on the experiential effect may not in fact con tradict the principle of specific deterrence. If respondents had previously committed offenses, but had not been punished for them, the principle of specific deterrence would subsequently predict a low level of perceived certainty. Individuals who are involved in repeated crimes without suffering punishment should have lower perceptions of risk, since they have gotten away with it so often. This in tum should be related to repeating offenses in the future (Stafford and Warr, 1993). Specific deterrence is supposed to operate based on persons getting caught and punished for criminal acts. If they are not, the theory argues, then they will come to believe that the certainty of punishment is low. In this sense, then, deterrence theory predicts the very experiential effect that Paternoster et al. found.

Modifications and Expansions of Deterrence Concepts Another aspect of this study by Paternoster et al. (1983; 1985) points

to the movement by many researchers to expand deterrence concepts

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

22 Criminological Theories: Introduction and Evaluation

beyond legal penalties. Paternoster et al. included variables from social bonding theory (i.e., moral beliefs and attachment to parents and peers) and social learning theory (i.e., the perceived risk of informal sanctions from family and friends and association with offenders). (See Chapters 4 and 5.) When these other variables are taken into account, the already weak relationship between the perceptions of risk of legal penalties and offense behavior virtually disappears.

The research by Paternoster et al. followed up previous studies (An derson et al., 1977; Akers et al., 1979; Grasmick and Green, 1980) in which the concept of deterrence is expanded beyond the strictly legal or formal sanctions to include "informal deterrence." Informal deter rence means the actual or anticipated social sanctions and other con sequences of crime and deviance that prevent their occurrence or recurrence. This research has found that the perceptions of informal sanctions, such as the disapproval of family and friends or one's own conscience and moral commitments, do have deterrent effects. Indeed, they have more effect on refraining from law violations than the per ceived certainty of arrest or severity of penalties (Green, 1989; Grasmick and Bursik, 1990).

Zimring and Hawkins (1973) have argued that formal punishment may deter most effectively when it "sets off" or provokes these informal social sanctions. An adolescent may refrain from delinquency, not only merely out of fear of what the police will do, but of what his or her parents will do once they learn of his or her arrest. Williams and Hawkins (1989) expand on this notion of the deterrent effects of informal sanc tions that may be triggered by the application of formal criminal justice sanctions. They found in their study that the arrest of an abusing hus band or boyfriend may have a deterrent effect, in part because of a concern over the negative reactions of friends, family, neighbors, or employers toward him based on their knowledge that he has been ar rested. In this instance, fear of arrest may be a deterrent, not only because of the negative experience of the arrest itself, but because of other negative consequences invoked by the arrest. These may include the informal costs of severed relationships, damage to one's reputation, and the possible loss of current or future employment. Williams and Hawkins argue that the general concept of deterrence should be ex panded to include these informal negative sanctions. Subsequent re search by Nagin and Paternoster (1991b) does not support this argument, however, when it is applied to delinquency. They find a very small deterrent effect from the perceptions of formal sanctions, and this effect is not increased at all as a result of informal costs that may be related to the formal sanction. Instead, the informal sanctions have an independent effect on delinquent behavior that is stronger than the effect of perceived formal sanctions.

22 Criminological Theories: Introduction and Evaluation

beyond legal penalties. Paternoster et al. included variables from social bonding theory (Le., moral beliefs and attachment to parents and peers) and social learning theory (Le., the perceived risk of informal sanctions from family and friends and association with offenders). (See Chapters 4 and 5.) When these other variables are taken into account, the already weak relationship between the perceptions of risk of legal penalties and offense behavior virtually disappears.

Zimring and Hawkins (1973) have argued that formal punishment may deter most effectively when it "sets off' or provokes these informal social sanctions. An adolescent may refrain from delinquency, not only merely out of fear of what the police will do, but of what his or her parents will do once they learn of his or her arrest. Williams and Hawkins (1989) expand on this notion of the deterrent effects of informal sanc tions that may be triggered by the application of formal criminal justice sanctions. They found in their study that the arrest of an abusing hus band or boyfriend may have a deterrent effect, in part because of a concern over the negative reactions of friends, family, neighbors, or employers toward him based on their knowledge that he has been ar rested. In this instance, fear of arrest may be a deterrent, not only because of the negative experience of the arrest itself, but because of other negative consequences invoked by the arrest. These may include the informal costs of severed relationships, damage to one's reputation, and the possible loss of current or future employment. Williams and Hawkins argue that the general concept of deterrence should be ex panded to include these informal negative sanctions. Subsequent re search by Nagin and Paternoster (1991 b) does not support this argument, however, when it is applied to delinquency. They find a very small deterrent effect from the perceptions of formal sanctions, and this effect is not increased at all as a result of informal costs that may be related to the formal sanction. Instead, the informal sanctions have an independent effect on delinquent behavior that is stronger than the effect of perceived formal sanctions.

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

23 Deterrence and Rational Choice Theories

That the informal sanction system may be more effective in control ling crime than legal sanctions should come as no surprise. But does research evidence that informal sanctions on criminal and delinquent behavior have a deterrent effect on crime increase the empirical validity of deterrence theory? In my opinion, it does not. Deterrence theory refers only to the threat of legal punishment.

[T]he proper definition [of deterrence]... is narrow. In a legal context, the term "deterrence" refers to any instance in which an individual con templates a criminal act but refrains entirely from or curtails the com mission of such an act because he or she perceives some risk of legal punishment and fears the consequences. (Gibbs, 1986:325-36)

There is no room in deterrence theory for variations in the rewards for crime, the social consequences of actions, individual or group pro pensities toward crime, and the whole range of other variables. The question to be answered about deterrence theory is not whether pun ishment of any kind from any source deters, but whether the threat of punishment by law deters. The more the deterrence theory is expanded to include informal sanctions and other aspects of the social environ ment beyond the law, the less it remains a deterrence theory and the more it begins to resemble other theories that already include these variables. It is more appropriate, therefore, to interpret positive find ings on informal sanctions and similar variables as supporting the other theories (e.g., social learning and social bonding), from which the vari ables have been borrowed, than it is to conclude that such findings support an expanded deterrence theory.

Rational Choice Theory

Deterrence and Expected Utility The expansion of the concept of deterrence has been most associated

with the introduction into criminology in the 1980s of "rational choice" theory.4 Rational choice theory is based on the "expected utility" prin ciple in economic theory. The expected utility principle simply states that people will make rational decisions based on the extent to which they expect the choice to maximize their profits or benefits and mini mize the costs or losses. This is the same general assumption about human nature made in classical criminology.

The obvious affinity between deterrence and rational choice theories stems from the fact that they both grew out of the same utilitarian philosophy of the eighteenth century (see Gibbs, 1975). The former was applied to the law, and the latter to the economy. Despite this long historical connection, rational choice theory of crime has only recently been introduced in criminology. Except for the use of such concepts as

Deterrence and Rational Choice Theories 23

[T]he proper definition [of deterrence] ... is narrow. In a legal context, the term "deterrence" refers to any instance in which an individual con templates a criminal act but refrains entirely from or curtails the com mission of such an act because he or she perceives some risk of legal punishment and fears the consequences. (Gibbs, 1986:325-36)

There is no room in deterrence theory for variations in the rewards for crime, the social consequences of actions, individual or group pro pensities toward crime, and the whole range of other variables. The question to be answered about deterrence theory is not whether pun ishment of any kind from any source deters, but whether the threat of punishment by law deters. The more the deterrence theory is expanded to include informal sanctions and other aspects of the social environ ment beyond the law, the less it remains a deterrence theory and the more it begins to resemble other theories that already include these variables. It is more appropriate, therefore, to interpret positive find ings on informal sanctions and similar variables as supporting the other theories (e.g., socialleaming and social bonding), from which the vari ables have been borrowed, than it is to conclude that such findings support an expanded deterrence theory.

Rational Choice Theory

Deterrence and Expected Utility The expansion of the concept of deterrence has been most associated

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

24 Criminological Theories: Introduction and Evaluation

"aleatory risk" in delinquency research by sociologists (Short and Strodtbeck, 1965), rational choice was introduced to criminology pri marily through the analyses of crime by economists (see Becker, 1968; Heineke, 1978; Crouch, 1979). Gibbs (1975:203) notes that "shortly after the revival of interest in the deterrence question among sociologists, economists were drawn to the subject in large number."

Some criminologists, who had been conducting deterrence research for some time, began in the 1980s to refer to the economic model of rational choice as an expansion of the deterrence doctrine beyond legal punishment. However, rational choice theorists claim much more than just an expansion of deterrence theory. The theory is proposed as a general, all-inclusive explanation of both the decision to commit a spe cific crime and the development of, or desistance from, a criminal career. The decisions are based on the offenders' expected effort and reward compared to the likelihood and severity of punishment and other costs of the crime (Cornish and Clarke, 1986).

Research on Rational Choice Theory Do offenders calculate that the effort and costs of crime are less than

the expected reward before the act in the way predicted by rational choice theory? The answer depends on whether one believes this theory assumes that pure or partial rationality operates in crime. Does the theory hypothesize that each person approaches the commission of a crime with a highly rational calculation of pleasure versus pain before acting on or refraining from the crime? Does an offender choose to commit a crime with full knowledge and free will , taking into account only a carefully reasoned, objectively or subjectively determined set of costs and benefits? If it is this kind of pure rationality that rational choice theory assumes, then the theory has virtually no empirical va lidity. The purely rational calculation of the probable consequences of an action is a rarity even among the general conforming public. More over, even offenders who pursue crime on a regular, business-like basis typically do not operate through a wholly rational decision-making process.

For instance, in a study of repeat property offenders, Tunnell (1990; 1992) found that the threat of re-imprisonment did not deter their recommission of crimes. Offenders thought that they would gain in come from their crimes and would not be caught, or they believed that they would not serve much prison time if they did get caught. Further more, they were not afraid to serve time in prison because life in prison was not threatening to them. These findings would seem to be in line with rational choice theory, since the expected benefits were perceived as outweighing the expected costs of the crime; hence, the decision was made to commit the crime. However, the process whereby offenders

24 Criminological Theories: Introduction and Evaluation

Research on Rational Choice Theory Do offenders calculate that the effort and costs of crime are less than

the expected reward before the act in the way predicted by rational choice theory? The answer depends on whether one believes this theory assumes that pure or partial rationality operates in crime. Does the theory hypothesize that each person approaches the commission of a crime with a highly rational calculation of pleasure versus pain before acting on or refraining from the crime? Does an offender choose to commit a crime with full knowledge and free will, taking into account only a carefully reasoned, objectively or subjectively determined set of costs and benefits? If it is this kind of pure rationality that rational choice theory assumes, then the theory has virtually no empirical va lidity. The purely rational calculation of the probable consequences of an action is a rarity even among the general conforming public. More over, even offenders who pursue crime on a regular, business-like basis typically do not operate through a wholly rational decision-making process.

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

25 Deterrence and Rational Choice Theories

reached a decision to attempt another crime did not fit the model of a purely rational calculation of costs and benefits. They did try to avoid capture, but their actions and assessments of the risks were very unre alistic, even to some extent irrational. They were unable to make rea sonable assessments of the risk of arrests, did little planning for the crime, and were uninformed about the legal penalties in the state where their crimes were committed. Moreover, all of the offenders in the study:

[R]eported that they (and nearly every thief they knew) simply do not think about the possible legal consequences of their criminal actions before committing crimes Rather than thinking of possible negative consequences of their actions, those offenders reported thinking pri marily of the anticipated positive consequences. . . . They simply be lieved that they would not be caught and refused to think beyond that point.

The decision-making process appears not to be a matter of rational evaluation or calculation of the benefits and risks . . . [R]isks (1) are thought about only rarely or (2) are considered minimally but are put out of their minds. (Tunnell, 1990:680-81)

Similarly, in an ethnographic study of burglars, Paul Cromwell and his associates found that "a completely rational model of decision mak ing in residential burglary cannot be supported" (Cromwell et al., 1991:43). Rather, professional burglars engage in only partially rational calculation of gains and risks before deciding to burglarize a house, and "research reporting that a high percentage of burglars make care fully planned, highly rational decisions based upon a detailed evalu ation of environmental cues may be in error" (Cromwell et al., 1991:42).

Most of our burglar informants could design a textbook burglary.... [T]hey often described their past burglaries as though they were ration ally conceived and executed. Yet upon closer inspection, when their previous burglaries were reconstructed, textbook procedures frequently gave way to opportunity and situational factors (Cromwell et al., 1991:42).

The empirical validity of a purely rational explanation of crime may not be important, however, because rational choice theorists seldom put forth such pure models. Instead, they have developed models of partial rationality that incorporate limitations and constraints on choices through lack of information, moral values, and other influences on criminal behavior. Although rational choice theorists often refer to the "reasoning criminal" and the "rational component" in crime, they go to great length to point out how limited and circumscribed reasoning and rationality are. The empirically verified models in the literature are based on the assumptions of a fairly minimal level of rationality.

Deterrence and Rational Choice Theories 25

[RJeported that they (and nearly every thief they knew) simply do not think about the possible legal consequences of their criminal actions before committing crimes .... Rather than thinking of possible negative consequences of their actions, those offenders reported thinking pri marily of the anticipated positive consequences .... They simply be lieved that they would not be caught and refused to think beyond that point.

The decision-making process appears not to be a matter of rational evaluation or calculation of the benefits and risks ... [RJisks (1) are thought about only rarely or (2) are considered minimally but are put out of their minds. (Tunnell, 1990:680-81)

Similarly, in an ethnographic study of burglars, Paul Cromwell and his associates found that "a completely rational model of decision mak ing in residential burglary cannot be supported" (Cromwell et al., 1991 :43). Rather, professional burglars engage in only partially rational calculation of gains and risks before deciding to burglarize a house, and "research reporting that a high percentage of burglars make care fully planned, highly rational decisions based upon a detailed evalu ation of environmental cues may be in error" (Cromwell et al., 1991 :42).

Most of our burglar informants could design a textbook burglary .... [TJhey often described their past burglaries as though they were ration ally conceived and executed. Yet upon closer inspection, when their previous burglaries were reconstructed, textbook procedures frequently gave way to opportunity and situational factors (Cromwell et al., 1991:42).

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

26 Criminological Theories: Introduction and Evaluation

(See, for example, Piliavin et al., 1986; Cornish and Clarke, 1986; Ben nett, 1986; Carroll and Weaver, 1986; Harding, 1990.)

Proponents often contrast rational choice theory with what they label "traditional criminology." They believe that this theory differs from other theories because, in their view, all other criminological theories assume that criminal behavior is irrational. But they are mistaken about other criminological theories. In reality, except for psychoanalytic the ory and some versions of biological theory (see Chapter 3), all other criminological theories assume no more or less rationality in crime than do most rational choice models.

Furthermore, the rational choice models that have been supported by research evidence do not stick strictly with measures of expected utility. They incorporate various psychological and sociological back ground and situational variables taken from other theories, to such an extent that there is little to set them apart from other theoretical models. In fact, some of the studies purporting to find evidence favoring rational choice theory actually test models that are indistinguishable from other, supposedly non-rational choice theories. The clearest example of this is the research by Paternoster (1989a; 1989b). He tested the effect on delinquent behavior of several variables in what he calls a "deter rence/rational choice" model. This model consists of the following vari ables: affective ties, costs of material deprivation, social groups and opportunities, informal social sanctions, perceptions of formal legal sanctions, and moral beliefs about specific delinquent acts. There is nothing in this set of variables that distinguishes it as a rational choice model. Al l the variables are taken from social learning and social bond ing theories. Paternosters finding that these variables are related to delinquent behavior, therefore, tells us little about the empirical validity of rational choice theory. However, it does tell us about the validity of social learning and social bonding theories.

The broadening of rational choice theory has the same consequence as the expansion of deterrence theory—it becomes a different theory. When rational choice theory is stated in its pure form, it does not provide an adequate explanation of criminal behavior. It provides a more empirically verified explanation of crime when it is expanded to include variables beyond rationally expected utility. However, when rational choice theory is modified in this way, the level of rationality it assumes is indistinguishable from that expected in other theories, and it incorporates explanatory variables from them. When the modifica tions reach this point, it is no longer appropriate to call the result rational choice theory (Akers, 1990).

26 Criminological Theories: Introduction and Evaluation

(See, for example, Piliavin et aI., 1986; Cornish and Clarke, 1986; Ben nett, 1986; Carroll and Weaver, 1986; Harding, 1990.)

Furthermore, the rational choice models that have been supported by research evidence do not stick strictly with measures of expected utility. They incorporate various psychological and sociological back ground and situational variables taken from other theories, to such an extent that there is little to set them apart from other theoretical models. In fact, some of the studies purporting to find evidence favoring rational choice theory actually test models that are indistinguishable from other, supposedly non-rational choice theories. The clearest example of this is the research by Paternoster (1989a; 1989b). He tested the effect on delinquent behavior of several variables in what he calls a "deter rence/rational choice" model. This model consists of the followingvari abIes: affective ties, costs of material deprivation, social groups and opportunities, informal social sanctions, perceptions of formal legal sanctions, and moral beliefs about specific delinquent acts. There is nothing in this set of variables that distinguishes it as a rational choice model. All the variables are taken from social learning and social bond ing theories. Paternoster's finding that these variables are related to delinquent behavior, therefore, tells us little about the empirical validity of rational choice theory. However, it does tell us about the validity of social learning and social bonding theories.

The broadening of rational choice theory has the same consequence as the expansion of deterrence theory-it becomes a different theory. When rational choice theory is stated in its pure form, it does not provide an adequate explanation of criminal behavior. It provides a more empirically verified explanation of crime when it is expanded to include variables beyond rationally expected utility. However, when rational choice theory is modified in this way, the level of rationality it assumes is indistinguishable from that expected in other theories, and it incorporates explanatory variables from them. When the modifica tions reach this point, it is no longer appropriate to call the result rational choice theory (Akers, 1990).

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

27 Deterrence and Rational Choice Theories

Routine Activities Theory

Felson and Cohen: Offenders, Targets, and Guardians Elements of deterrence and rational choice are also found in routine

activities theory.5 In order for a personal or property crime to occur, there must be at the same time and place a perpetrator, a victim, and/or an object of property. The occurrence can be facilitated if there are other persons or circumstances in the situation that encourage it, or it can be prevented if the potential victim or another person is present who can take action to deter it. Lawrence Cohen and Marcus Felson (1979) have taken these basic elements of time, place, objects, and persons to develop a "routine activities" theory of crime events. They do so by placing these elements into three categories of variables which increase or decrease the likelihood that persons will be victims of "direct contact" predatory (personal or property) crime.

The three main categories of variables identified by Cohen and Fel son are: (1) motivated offenders, (2) suitable targets of criminal victimi zation, and (3) capable guardians of persons or property. The main proposition in the theory is that the rate of criminal victimization is increased when there is a "convergence in space and time of the three minimal elements of direct-contact predatory violations" (Cohen and Felson, 1979:589). That is, the likelihood of crime increases when there is one or more persons present who are motivated to commit a crime, a suitable target or potential victim that is available, and the absence of formal or informal guardians who could deter the potential offender. The relative presence or absence of these elements is variable, and "the risk of criminal victimization varies dramatically among the circum stances and locations in which people place themselves and their prop erty" (Cohen and Felson, 1979:595). The theory derives its name from the fact that Cohen and Felson begin with the assumption that the conjunction of these elements of crime are related to the normal, legal, and "routine" activities of potential victims and guardians. "[T]he spa tial and temporal structure of routine legal activities should play an important role in determining the location, type, and quantity of illegal acts occurring in a given community or society" (Cohen and Felson, 1979:590).

Routine activities are defined by Cohen and Felson as "recurrent and prevalent activities which provide for basic population and individual needs . . . formalized work, as well as the provision of standard food, shelter, sexual outlet, leisure, social interaction, learning, and child- bearing" (Cohen and Felson, 1979:593). They hypothesize that changes in daily activities related to work, school, and leisure since World War II have placed more people in particular places at particular times which both increase their accessibility as targets of crime and keep

Deterrence and Rational Choice Theories 27

Routine Activities Theory

Felson and Cohen: Offenders, Targets, and Guardians Elements of deterrence and rational choice are also found in routine

activities theory.s In order for a personal or property crime to occur, there must be at the same time and place a perpetrator, a victim, and/or an object of property. The occurrence can be facilitated if there are other persons or circumstances in the situation that encourage it, or it can be prevented if the potential victim or another person is present who can take action to deter it. Lawrence Cohen and Marcus Felson (1979) have taken these basic elements of time, place, objects, and persons to develop a "routine activities" theory of crime events. They do so by placing these elements into three categories of variables which increase or decrease the likelihood that persons will be victims of "direct contact" predatory (personal or property) crime.

Routine activities are defined by Cohen and Felson as "recurrent and prevalent activities which provide for basic population and individual needs ... formalized work, as well as the provision of standard food, shelter, sexual outlet, leisure, social interaction, learning, and child bearing" (Cohen and Felson, 1979:593). They hypothesize that changes in daily activities related to work, school, and leisure since World War II have placed more people in particular places at particular times which both increase their accessibility as targets of crime and keep

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

28_ Criminological Theories: Introduction and Evaluation

them away from home as guardians of their own possessions and prop erty.

In recent applications of the theory, Felson (1994) downplays the significance of formal guardians because "crime is a private phenome non largely impervious to state intervention." Rather, he emphasizes the crime prevention and deterrence that naturally occurs in the infor mal control system, the "quiet and natural method by which people prevent crime in the course of daily life. This control occurs as people interact and bring out the best in one another" (Felson, 1994:xii-xiii). The police are not the only capable guardians. Indeed, guardians who prevent or deter crime are more likely to be ordinary citizens, oneself, friends, family, or even strangers. The vulnerability of property to theft is affected by a number of physical features, such as its weight and ease of mobility and how much physical "target hardening" (e.g., installing better locks) has been done. But sights and sounds, being in dangerous and risky places, routines of the family and household, and one's per sonal characteristics have an effect on the risk of victimization for both violent and property crime. Felson also extends the theory beyond predatory crimes to such offenses as illegal consumption and sales of drugs and alcohol.

Felson's emphasis on the informal control system does not distin guish routine activities theory from the general sociological view (dis cussed above and in Chapter 8) that conformity to the law comes more from the informal system of socialization and control than from the formal control system. This general sociological view has also been applied to drug use. "The general reduction in drug use in American society [from the late 1970s to the early 1990s] may be the result of changes in social norms and the informal control system unrelated to conscious and deliberate prevention, treatment, or law enforcement efforts" (Akers, 1992:183). The validity of routine activities theory, therefore, does not rest on the relative importance of the informal and formal control systems in crime but on how well hypotheses about the effect of the three main elements of the theory on crime are supported.

Empirical Validity of Routine Activities Theory Cohen and Felson argue that a change in any one of these elements

would change the crime rates, but that the presence of all three would produce a multiplier effect on crime rates. Their research (Cohen and Felson, 1979), however, focused on only two of the three elements: suitable targets and the absence of capable guardians. They do not rule out that the "routine activity approach might in the future be applied to the analysis of offenders and their inclinations as well" (Cohen and Felson, 1979:605).

28 Criminological Theories: Introduction and Evaluation

them away from home as guardians of their own possessions and prop erty.

Felson's emphasis on the informal control system does not distin guish routine activities theory from the general sociological view (dis cussed above and in Chapter 8) that conformity to the law comes more from the informal system of socialization and control than from the formal control system. This general sociological view has also been applied to drug use. "The general reduction in drug use in American society [from the late 1970s to the early 1990s] may be the result of changes in social norms and the informal control system unrelated to conscious and deliberate prevention, treatment, or law enforcement efforts" (Akers, 1992: 183). The validity of routine activities theory, therefore, does not rest on the relative importance of the informal and formal control systems in crime but on how well hypotheses about the effect of the three main elements of the theory on crime are supported.

Empirical Validity of Routine Activities Theory Cohen and Felson argue that a change in anyone of these elements

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

29 Deterrence and Rational Choice Theories

They present data on post-war trends (into the 1970s) in family ac tivities, consumer products, and businesses that seem to be compatible with the trends in type and rate of crimes in the United States. Their main findings, however, relate crime rates to a "household activity ratio," the percentage of all households that are not husband-wife fami lies or where the wife is employed in the labor force. Such households are more vulnerable to crime victimization because their members are away from home more and less able to function as guardians of their property. They are more likely to possess more desirable goods to be stolen, and they are more exposed to personal crime away from home. Controlling for age composition and unemployment, Cohen and Felson found that the changes in household activity were correlated with changes in the rates of all major predatory violent and property crimes. They recognize that these were not the direct measures of the concepts in the theory, but they conclude that the findings are consistent with the theory.

Cohen et al. (1981) present the theory in a more formalized fashion, renaming it "opportunity" theory and testing its propositions with data from the national crime victimization surveys. The formal theory refers to exposure, proximity, guardianship, and target attractiveness as vari ables that increase the risk of criminal victimization. But these are not measured directly. They are assumed from variations in age, race, and income, household composition, labor force participation, and resi dence in different areas of the city. Although inconclusive on some, their findings are consistent with most of the hypotheses; consequently, they conclude that the theory is supportable.

Messner and Tardiff (1985) use the routine activities approach to interpret their findings on the correlations between the social charac teristics of Manhattan homicide victims, the time and location of the homicides, and the relationship between victims and offenders. They do not attempt to account for the rate or number of homicides, but only for the place and type of homicide. They contend that "sociode- mographic and temporal characteristics structure routine activities and, in so doing, affect both the location of potential victims in physical space and the 'pool' of personal contacts from which offenders are ultimately drawn" (Messner and Tardiff, 1985:243). These sociode- mographic and temporal variables should be related to where and with whom one carries out the normal round of activities. When one's routine activities are concentrated in and around the home, victims and offend ers are less likely to interact; whereas, spending more time away from the house increases the chances of victimization by strangers. Messner and Tardiff found weak support for the expectations about family versus stranger homicides, but no relationship between time and location of homicides.

Deterrence and Rational Choice Theories 29

Messner and Tardiff (1985) use the routine activities approach to interpret their findings on the correlations between the social charac teristics of Manhattan homicide victims, the time and location of the homicides, and the relationship between victims and offenders. They do not attempt to account for the rate or number of homicides, but only for the place and type of homicide. They contend that "sociode mographic and temporal characteristics structure routine activities and, in so doing, affect both the location of potential victims in physical space and the 'pool' of personal contacts from which offenders are ultimately drawn" (Messner and Tardiff, 1985:243). These sociode mographic and temporal variables should be related to where and with whom one carries out the normal round of activities. When one's routine activities are concentrated in and around the home, victims and offend ers are less likely to interact; whereas, spending more time away from the house increases the chances of victimization by strangers. Messner and Tardiff found weak support for the expectations about family versus stranger homicides, but no relationship between time and location of homicides.

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

30 Criminological Theories: Introduction and Evaluation

Sherman et al. (1989) also report findings consistent with routine activities theory in their study of the "hot spots" of predatory crime. They note that prior research on routine activities used data on the characteristics of individuals or households as measures of lifestyles that affect the convergence of victim, offender, and guardians. Their research focused on the "criminology of place" by using Minneapolis police "call data" (i.e., crimes reported to the police by telephone) in Minneapolis to locate concentrations (i.e., hot spots) of such calls at certain addresses, intersections, parks, and hospitals. They found that most crime reports came from only 3 percent of all the locations in the city and that reports of each of the major types of predatory crime were concentrated only in a few locations. Sherman et al. do not know what it is about these places that make them hot spots, but they believe that there is something about them that relates to the convergence of victims and offenders in the absence of guardians.

Kennedy and Forde (1990) also reported support for routine activi ties theory based on both property and violent crime data from a tele phone victimization survey. They found that victimization varies by age, sex, and income, but also varies by the extent to which persons stay at home or go out at night to bars, work, or school. They concluded from this that the routine activity of leaving home at this time renders these persons more vulnerable as victims and less capable as guardians over their property.

Findings from qualitative research on the responses of the formal and informal control systems to the devastation of Hurricane Andrew in Florida in 1992 are generally consistent with routine activities theory. The natural disaster temporarily increased the vulnerability of persons and property as crime targets. For a short time, there was nearly a complete loss of formal guardianship in the form of police protection in some of the neighborhoods. Motivated offenders with previous re cords were attracted to the areas in the aftermath of the storm, and some local people took criminal advantage of the situation. However, there was little looting in the neighborhoods and crime rates actually went down during the time when the community was most vulnerable (but then increased again after the initial impact period). This was most likely the result of stepping into the void by competent guardians in the form of neighbors watching out for neighbors, citizens guarding their own and others property (sometimes with firearms), citizen pa trols, and other steps taken to aid one another in the absence of gov ernment and formal control (Cromwell et al., 1995).

The fact that some may be motivated to commit crime when targets are made vulnerable by such events as natural disasters raises questions about the concept of motivated or likely offender. Does the concept of motivated offender in routine activities theory refer only to someone

30 Criminological Theories: Introduction and Evaluation

Sherman et al. (1989) also report findings consistent with routine activities theory in their study of the "hot spots" of predatory crime. They note that prior research on routine activities used data on the characteristics of individuals or households as measures of lifestyles that affect the convergence of victim, offender, and guardians. Their research focused on the "criminology of place" by using Minneapolis police "call data" (i.e., crimes reported to the police by telephone) in Minneapolis to locate concentrations (Le., hot spots) of such calls at certain addresses, intersections, parks, and hospitals. They found that most crime reports came from only 3 percent of all the locations in the city and that reports of each of the major types of predatory crime were concentrated only in a few locations. Sherman et al. do not know what it is about these places that make them hot spots, but they believe that there is something about them that relates to the convergence of victims and offenders in the absence of guardians.

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

31 Deterrence and Rational Choice Theories

with a pre-existing set of crime-prone motivations or does it include anyone who is enticed by the opportunity for quick gain itself, even though he or she may not have previously existing criminal intentions? In the former case, the situation provokes motivation to action but does not create it; in the latter, the situation both creates and provokes the motivation. Since all persons are thus potentially motivated to commit crime, can the presence of a motivated offender simply be assumed from the presence of any person? If so, how does the theory distinguish between circumstances in which a motivated offender is present and those in which one is not? There is ambiguity on this point in routine activities theory (Akers et al., 1994).

Jensen and Brownfield (1986) point to another variable which is seldom controlled for in studies of routine activities: the deviant or non-deviant nature of the activities in which victims are engaged. They found that the activities most strongly related to adolescents becoming victims of crime are not the normal conforming routine activities (dat ing, going out at night, shopping, or going to parties), but rather the deviant activity of committing offenses. In other words, those who commit crimes are more likely to be victims of crime. Engaging in offense behavior itself, of course, does not fit Cohen and Felson's defi nition of "routine" activity. Moreover, as Jensen and Brownfield (1986) point out, since criminal behavior is correlated with victimization, vari ables taken from theories that explain criminal behavior should also be correlated with victimization.

In fact, most of the variables in the opportunity [routine activities] model of victimization have appeared in one form or another in tradi tional etiological theories of crime or delinquency. Exposure and prox imity to offenders is central to differential association and social learning theories of criminality. Cohen et al. propose that exposure and proximity to offenders increase the risk of victimization, while differen tial association and social learning theories propose that the same vari ables increase the chances of criminal behavior. . . . In short, "victimogenic" variables have been introduced in earlier theories as "criminogenic." (Jensen and Brownfield, 1986)

Even though it draws upon etiological theories, routine activities theory is only indirectly a theory of the commission of criminal behav ior. It is primarily a theory of criminal victimization. That is, it does not offer an explanation of why some persons are motivated to develop a pattern of crime or commit a particular crime. It simply assumes that such persons exist and that they commit crimes in certain places and times at which the opportunities and potential victims are available. Routine activities theory does not explain why informal crime precau tions may or may not be exercised by individuals in their homes or elsewhere, nor does it explain formal control exercised by law and the

Deterrence and Rational Choice Theories 31

In fact, most of the variables in the opportunity [routine activities] model of victimization have appeared in one form or another in tradi tional etiological theories of crime or delinquency. Exposure and prox imity to offenders is central to differential association and social learning theories of criminality. Cohen et al. propose that exposure and proximity to offenders increase the risk of victimization, while differen tial association and social learning theories propose that the same vari ables increase the chances of criminal behavior .... In short, "victimogenic" variables have been introduced in earlier theories as " criminogenic." (Jensen and Brownfield, 1986)

Even though it draws upon etiological theories, routine activities theory is only indirectly a theory of the commission of criminal behav ior. It is primarily a theory of criminal victimization. That is, it does not offer an explanation of why some persons are motivated to develop a pattern of crime or commit a particular crime. It simply assumes that such persons exist and that they commit crimes in certain places and times at which the opportunities and potential victims are available. Routine activities theory does not explain why informal crime precau tions mayor may not be exercised by individuals in their homes or elsewhere, nor does it explain formal control exercised by law and the

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

32 Criminological Theories: Introduction and Evaluation

criminal justice system. It simply assumes that, if informal or formal guardians are not present or able to prevent crime, then crime wil l occur.

We have long known that vulnerability to criminal victimization is related to social characteristics such as age, sex, and race, and that unguarded or easily available property is more apt to be stolen or van dalized. Ordinary precautions, of course, decrease the chances of vic timization. Common sense tells us that, if one is sitting at home watching television rather than out on the streets, one's home is not likely to be burglarized and one has a zero chance of being the victim of a street mugging. Possessing social characteristics correlated with a higher-risk lifestyle obviously makes one more vulnerable as a crime victim. But Felson and others have taken these common sense and empirical realities and woven them into a coherent framework for un derstanding the variations in criminal victimization by time and place. The theory is well-stated, logically consistent, and has clear policy im plications and powerful potential for understanding the impact of nor mal, even desirable, social structural changes on predatory crime.

Its empirical validity has not yet been well-established, however. As we have seen, several researchers have reported findings that are con sistent with routine activities theory (see also Stahura and Sloan, 1988; Massey et al., 1989; Miethe et al., 1987; Cromwell et al., 1991). But that research has not really tested full models of the theory. With a few exceptions (Stahura and Sloan, 1988), researchers have not measured variations in the motivation for crime or variations in the presence of motivated offenders. Thus, at least one of the three major categories of variables in this theory is usually omitted. Even when included, offender motivation is not directly measured, but rather assumed from vari ations in the demographic correlates of crime.

Similarly, the other two major categories of suitable targets of crime and absence of capable guardians are usually not directly measured. The original research by Cohen, Felson, and associates used no direct measures of the routine activities of victims or suitable guardians. They were only assumed from labor force participation, household compo sition, and so on. In subsequent research, victim vulnerability and guardianship have usually been assumed from the social characteristics of victims, although some activities of victims (e.g., their presence at home or their going out at night) have been directly measured (Kennedy and Forde, 1990). As Sherman et al. noted, "most tests of routine ac tivities theory lack independent measures of the lifestyles in question and substitute presumed demographic correlates for them" (Sherman et al., 1989:31). The research on routine activities has reported numer ous findings that are consistent with the assumptions in routine activi-

32 Criminological Theories: Introduction and Evaluation

criminal justice system. It simply assumes that, if informal or formal guardians are not present or able to prevent crime, then crime will occur.

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

33 Deterrence and Rational Choice Theories

ties theory. More work needs to be done to devise direct empirical measures of its key concepts.

Summary Deterrence theory states that if legal penalties are certain, severe,

and swift, crime will be deterred. In empirical studies severity is seldom found to have a deterrent effect on crime. Neither the existence of capital punishment nor the certainty of the death penalty have had an effect on the rate of homicides. A negative correlation between objective or perceived certainty and illegal behavior is a common research find ing, but the correlation tends to be weak.

There is more empirical support when deterrence concepts are ex panded to take into account the informal social processes of reward, punishment, and moral beliefs. Rational choice theory is another type of expansion or modification of deterrence theory. When rational choice theory is stated in its pure form, it does not stand up well to empirical evidence. However, when this theory is modified so that a relatively low level of rationality is assumed and explanatory variables from other theories are added, it is more likely to be upheld by the data. When deterrence and rational choice theories are so modified, they resemble more the modern social bonding or social learning theories than the classical deterrence or pure rational choice models. Therefore, positive research findings on these modified versions are more appropriately viewed as validating these other theories from which the more powerful explanatory variables are taken, rather than validating deterrence or rational choice theories alone.

The main proposition in routine activities theory is that the rate of criminal victimization is increased when there is one or more persons likely to commit a crime, a vulnerable target or victim is present, and formal or informal guardians to prevent the motivated offender are absent. Research has not tested full models of the theory, and the major variables are usually measured indirectly. Its empirical validity has not yet been firmly established, but most of the research done so far reports findings consistent with the theory.

Deterrence and Rational Choice Theories 33

ties theory. More work needs to be done to devise direct empirical measures of its key concepts.

Summary Deterrence theory states that if legal penalties are certain, severe,

There is more empirical support when deterrence concepts are ex panded to take into account the informal social processes of reward, punishment, and moral beliefs. Rational choice theory is another type of expansion or modification of deterrence theory. When rational choice theory is stated in its pure form, it does not stand up well to empirical evidence. However, when this theory is modified so that a relatively low level of rationality is assumed and explanatory variables from other theories are added, it is more likely to be upheld by the data. When deterrence and rational choice theories are so modified, they resemble more the modem social bonding or social learning theories than the classical deterrence or pure rational choice models. Therefore, positive research findings on these modified versions are more appropriately viewed as validating these other theories from which the more powerful explanatory variables are taken, rather than validating deterrence or rational choice theories alone.

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed

34 Criminological Theories: Introduction and Evaluation

Notes

1. For Beccarias writings, see Beccaria (1963; 1972) and Monachesi (1973). For Benthams writings, see Bentham (1948) and Geis (1973). For general discussions of the classical criminology of both, see Vold (1958), Vo;d and Bernard (1986), and Wright (1993b).

2. Piers Beirne (1991) argues that Beccaria's main purpose was neither legal reform nor a rational explanation of crime. Rather, it was to intro duce a deterministic "science of man" which ran contrary to the assump tions of free-will, volitional acts. Therefore, his theory was just as positivistic as subsequent theories.

3. See, for instance, Jensen, (1969), Waldo and Chiricos (1972), Anderson et al. (1977), Jensen et al. (1978), Tittle (1980), Paternoster et al. (1983), Klepper and Nagin (1989), Nagin and Paternoster (1994), Miller and Iovanni (1994).

4. For general and specific rational choice models, many of which are basically expansions on deterrence theory, see Cornish and Clarke (1986), Piliavin et al. (1986), Klepper and Nagin (1989), Paternoster, (1989a; 1989b), Williams and Hawkins (1989), Grasmick and Bursik (1990). For general critiques of rational choice theory, see Gibbs (1989) and Akers (1990).

5. Since routine activities theory stresses the ecological distribution of victims, crime opportunities, and motivated offenders, it could well be classified with social disorganization as an ecological theory of crime (see Chapter 7). It is not rational choice or deterrence theory. However, the concept of guardianship includes formal actions by police to deter crime and incorporates elements of the deterrence doctrine. Also, it makes the assumption that motivated offenders choose to commit a crime after assessing the presence of guardians and the vulnerability of crime targets. Therefore, it is often interpreted as a rational choice the ory. For these reasons, a discussion of it is included in this chapter.

34 Criminological Theories: Introduction and Evaluation

Notes

1. For Beccaria's writings, see Beccaria (1963; 1972) and Monachesi (1973). For Bentham's writings, see Bentham (1948) and Geis (1973). For general discussions of the classical criminology of both, see VoId (1958), VoId and Bernard (1986), and Wright (1993b).

C op

yr ig

ht ©

1 99

9. T

ay lo

r &

F ra

nc is

G ro

up . A

ll rig

ht s

re se

rv ed