Sociological Theories

Social institutions are important in socializing children. In this first picture, parents are reading to their child at home. Inthe second, family members are attending church together. In the third, children are arriving by bus at their school. Theenvironment of social institutions may have a significant impact on whether a child becomes a delinquent or law-abidingperson.

Sociological Theories of Criminal Behavior II: The Social-Process Approach

CHAPTER 6

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM 42

143

CHAPTER OUTLINEKey TermsIntroductionLearning Theory

Tarde’s Imitation TheoryThe Modern Media and Imitation

Social Learning TheorySutherland’s Differential Association TheoryAkers’s Social Learning Theory

Control TheoryReiss and Nye’s Control TheoryReckless’s Containment TheoryHirschi’s Bonding TheoryGottfredson and Hirschi’s Self-Control Theory

Labeling TheoryThe Emergence and Development of Labeling TheoryThe Effects of LabelingEmpirical ResearchRecent Refinements of Labeling Theory

Braithwaite’s Restorative JusticeMatsueda’s Reflective AppraisalSampson and Laub’s Cumulative Disadvantages

Evaluation of Labeling TheoryIntegrated Theories

Akers: Conceptual AbsorptionDevelopmental and Life-Course Theories

Sampson and Laub: Age-Graded TheoryMoffitt: Developmental Theory

Cullen and Colvin: Social Support or CoercionThornberry: Interactional TheoryKrohn: Network AnalysisKaplan: Self-DerogationTittle: Control BalanceElliott, Ageton, and Canter: Integrated Strain/ControlRobinson: Integrated Systems Theory of Antisocial

BehaviorAgnew: General TheoryFarrington: Integrated Cognitive Antisocial PotentialConclusion to Integrated Theories

Conclusion to Criminal Behavior TheoriesPolicy Implications of Criminal Behavior TheoriesSummaryStudy QuestionsBrief Essay AssignmentsInternet ActivitiesNotes

KEY TERMSadolescent limitedsage-graded theoryconceptual absorptionconceptual fusioncontainment theorycontrol balance desirabilitycontrol balance theorycontrol theorycumulative disadvantagedesistancedevelopmental and life-course (DLC) theorydifferential associationdifferential association-reinforcementdisintegrative shamingintegrated cognitive antisocial potential (ICAP) theoryintegrated systems theory (IST)interactional theorylabeling theorylife courselife-course persistersreflected appraisalreintegrative shamingrestorative justiceself-conceptself-control theoryself-derogation theorysequential integrationshamingsingular theory assessmentsocial capital

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 143

Chapter 5 discussed sociological theories that empha-size the relationship of criminal behavior to thesocial structure or the organization of society.

Whether the environment is seen as a determining or afacilitating factor in the causation of crime, the emphasisis on the environment’s structure rather than on individ-ual characteristics, as in the constitutional theories dis-cussed in Chapters 3 and 4. One problem with the social-structural theories, however, is that they do not explainthe process by which individuals turn to crime. This chap-ter analyzes social-process theories, which attempt to ex-plain how individuals become criminals.

It is not possible to separate all sociological theories intothe categories of social structure and social process. Sometheories may be considered in both categories. For exam-ple, the neutralization theory of Gresham Sykes and DavidMatza, discussed in Chapter 5, explains the process bywhich a person neutralizes any inhibitions that he or shemight have against violating laws. Yet the theory is alsorelated to subcultures, an aspect of the social structure.Likewise, labeling theory is discussed in this chapterbecause it explains the process by which a person becomesa criminal—the person is labeled by those in a position tomake that determination. Others characterize labeling the-ory as a social-structural theory that has some similaritiesto conflict theory because it is through the social institu-tions that labeling occurs.

Social-process theories developed as sociologists began toanalyze the fact that not all people exposed to the same social-structural conditions respond in the same way. Some becomelaw-abiding citizens, and others become criminals. Not all

criminals respond in criminal ways in all circumstances; like-wise, not all noncriminals observe the law at all times. Toexplain these differences, sociologists suggest that humanbehavior is learned and that criminal behavior may beacquired in the same way as any other behavior. Thisapproach has also been taken by other disciplines.

Learning TheoryVarious scholars have based their research on learning the-ory. The topic was introduced in Chapter 4’s discussion ofpsychological theories. There, it was noted that learningmay be accomplished by using other people as models andthat motivations can come from biological and psycholog-ical factors. We noted the work of Hans J. Eysenk, whostressed the interrelationship between psychology and biol-ogy in explaining how humans learn to behave. This chap-ter adds the social factor, but first it looks at other contri-butions that may have influenced the development of socialconcepts of learning theory.

Although learning theory has been developed by soci-ologists and psychologists to explain a variety of behav-iors, this chapter’s discussion of the theory will focus oncriminal behavior.

Tarde’s Imitation TheoryGabriel Tarde (1843–1904), who may be considered a fore-runner of modern imitation theory, was born in southernFrance. After studying law, he was a magistrate for manyyears. He showed considerable interest in social problems

144 PART 2 / Explanations of Criminal Behavior

influence of that theory is assessed in light of recentreformulations of learning theory by sociologists andcriminologists. The discussion of learning theory closes with alook at imitation theory, the mass media, and criminal behavior.

The next section of the chapter examines controltheories, which are based on the assumption that criminalbehavior occurs when society’s normal methods ofcontrolling people break down. The chapter then discusseslabeling theory, which focuses on the process by whichpeople who engage in certain acts come to be called (orlabeled) criminal whereas others who engage in those samekinds of behavior are not labeled. A section on integratedtheories discusses numerous approaches to integratingtheories of criminal behavior. The chapter closes with aconclusion to criminal behavior theories and an analysis ofthe policy implications of the theories of criminality thathave been explored in this chapter and in Chapter 5.

KEY TERMS (continued)

social supporttheoretical integrationtheory competitiontrajectoriestransitionsturning points

INTRODUCTION

This chapter is concerned with the process by which peoplebecome criminals. Do they learn this behavior? If so, how and under what circumstances? The chapter begins with adiscussion of learning theory, which includes a look at thedevelopment of imitation theory in the 1800s, followed by abrief look at the influence that today’s media may have onimitative behavior. The chapter then turns to an analysis of the social learning theory called differential association. The

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 144

and, as a judge, proved himself to be a deep thinker and aprofound philosopher. Reacting against Cesare Lombrosoand the positivist school, Tarde argued that people are notborn criminal; they become criminal. He saw criminalbehavior as the result primarily of social factors, a belief thatconstitutes one of his greatest contributions to criminology.1

Tarde rejected the biological and physical theories ofcriminal behavior, but he did not become a social deter-minist. He thought people had some choice in their behav-ior, although he did believe that when the ability to chooseis impaired, people should not be held responsible for theircriminal acts.2

Tarde’s social-process theory of criminal behavior isreflected in his belief that all the “important acts of sociallife are carried out under the domination of example.”3

Upon this belief he formulated his theory of imitation,through which he explained the process of acquiring crim-inal as well as noncriminal behavior. In developing his the-ory, Tarde distinguished between fashion and custom,both of which are forms of imitation. Fashion is charac-teristic of the imitation that takes place in crowds or citieswhere contact is close and frequent. Custom refers to thephenomenon that occurs in small towns and rural areaswhere contact is less frequent and change occurs less fre-quently. Since both fashion and custom are forms of imi-tation, each occurs to some degree within a society; fash-ion may uproot and create a new custom. Since fashionand custom are related to the degree of social contact,Tarde formulated his first law of imitation: “Men imitateone another in proportion as they are in close contact.”4

Tarde’s second law of imitation is that the inferior imi-tates the superior. Peasants imitate royalty; small-town andrural residents imitate city residents. Tarde wrote: “Infec-tious epidemics spread with the air or the wind; epidemicsof crime follow the line of the telegraph.”5 The third lawof imitation is the law of insertion: “When two mutuallyexclusive fashions come together, one can be substitutedfor the other. When this happens, there is a decline in theolder method and an increase in the newer method.”6

Tarde illustrated this position by noting the increase in theuse of a gun rather than a knife for murder.

Tarde’s neglect of the physical, psychological, and eco-nomic influences in behavior and his oversimplification ofcausation have led most sociologists to reject his imitationtheory, but, said sociologist Stephen Schafer, Tarde’s“emphasis on the social origins of crime had a lasting impacton criminological thought in both Europe and America.”7

The recent emphasis on the influence of the mass mediaon behavior has refocused on the possible impact of imi-tation on behavior, especially criminal behavior.

The Modern Media and ImitationIt is possible that delinquent and criminal behavior islearned from the media rather than through associationwith other human beings. Whether the process is one of

identification or imitation, the result is the same: What themedia portray may influence how the audience behaves.Some of that impact may be seen in the concern overwhether television has an effect on behavior, a concern thatraises the possibility of psychological and sociologicalexplanations of the process by which a person becomeseither deviant or law-abiding.

Television and the Internet are the prime sources ofnews for most Americans, and there is evidence that manytelevision viewers believe that the news they acquire is rea-sonably accurate.8 The issue here, however, is whetherexposure to television (or other media, such as movies orthe Internet) causes crime. There are arguments pro andcon, and actual causation is difficult to prove. A Yale Uni-versity psychology professor, who in 1960 began a studyof the causes of aggression, visited his sample 10 years laterand again in another 10 years; he reported a high correla-tion between violence and the amount of televisionwatched. The professor stated that he “found that the vio-lent programming [his subjects] had watched was relatedto the seriousness of the crimes they committed, howaggressive they were to their spouses, and even to howaggressive their own kids were.”9

Writing in the 1970s, one of the most prominent of thepsychological learning theorists, Albert Bandura, referredto the “reciprocal interaction between cognitive, behav-ioral and environmental determinants.”10 According toBandura, research had shown that television, the mostinfluential of the media for adolescents, had four types ofeffects on their social behavior:

1. The teaching of aggressive styles of conduct

2. The lessening of restraints on aggression

3. Desensitization and habituation to violence

4. The shaping of images of reality on which peoplebase their actions

Bandura claimed that television can distort people’sperceptions of the real world: “Heavy viewers see the soci-ety at large as more dangerous regardless of their educa-tional level, sex, age, and amount of newspaper reading.”11

Some of the attention given to the effects that violenceon television and the Internet might have on the behaviorof adults and children is anecdotal (i.e., based on individ-ual stories rather than on empirical research). Crimes suchas those discussed in Media Focus 6.1 may, however, helpform people’s attitudes about the influence of the media.Consider those crimes carefully. How do you react to eachcase in terms of a causal relationship between the crimi-nal act and television viewing?

Researchers remind us that media exposure does notoccur in a vacuum. Although most of the surveys of reac-tions to a particular television show, for example, occurright after the show is televised, we cannot assume a directcause-and-effect relationship. The behavior must be exam-ined in the total context in which it occurs, and the facts

CHAPTER 6 / Sociological Theories of Criminal Behavior II: The Social-Process Approach 145

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 145

can be misleading. For example, women watch more tele-vision than men, yet crime rates are higher among men.Teens view less television than adults do, yet most prop-erty crimes are committed by teens and young adults(under 25). Clearly something other than television isinvolved. A National Institute of Justice review of researchon the issue concluded that there is “no clear evidence ofcausal links.”12 James Q. Wilson and Richard J. Herrnsteinconcluded that most of the research does not show a causallink between television and crime.13

The critical question is to determine the processes thatare involved in reactions to media violence. Psychological

and sociological theories of behavior must be considered.14

Many people who view the programs or movies in ques-tion do not engage in antisocial behavior. Why do theyreact in appropriate ways, whereas others who viewed thesame material react in antisocial, even violent, ways? Tan-nis MacBeth Williams suggested this explanation:

The empirical and theoretical evidence suggeststhat . . . the effects of television’s content depend inpart on the extent to which contradictory messages areavailable, understood, and consistent. In the case of sex-role attitudes, messages from television are consistent


6.1

ME

DIA

FO

CU

S

The following crimes were reported in the media during recentyears. Each case stated that the media were the cause or amajor cause of the crime. What questions would you raise inreaction to these cases?

1. T. J. Soloman, age 15, was accused of shooting six class-mates (one of whom died) at his high school in Conyers, Geor-gia. Police found a letter written by Soloman in which he statedhis allegiance to the “brothers and sisters related to the trenchcoat mafia.” That was a group in which Littleton, Colorado’s,Columbine High School shooters, Eric Harris and Dylan Kle-bold, were fringe members. Just one month previously, Harrisand Klebold had gone on a killing rampage at their high school,killing 14 students and one teacher before killing themselves.Soloman was reported to have boasted to other students thathe could do a better job than the Columbine killers and thathis school should have had a shooting long before. It was sug-gested that Soloman’s knowledge of the Columbine shootingscame from his exposure to television.1

2. A 15-year-old boy, who was accused of sexually abus-ing his 8-year-old half-sister for the previous three years, saidhe got the idea for the crime from watching television. Theyouth (along with his 13-year-old brother) testified that hehad oral, anal, and vaginal sex with the child after theirmother left for work in the mornings and before she returnedin the afternoons. When asked by police how he learnedthese behaviors, the boy said, “I watched ‘The Jerry SpringerShow.’” When asked what that meant, he said he watcheda show on incest. The program was characterized by lewdlanguage and physical violence, but only one or two episodeshad dealt with incest. Still, one of those shows was entitled“I’m Pregnant by My Brother.”2

3. After arresting a 17-year-old accused of murdering hisstepmother and half-sister, police claimed the suspect wasobsessed with the movie Natural Born Killers. Friends saidthe youth had been unhappy, had threatened to kill his fam-ily, and, after seeing the movie, had shaved his head andwore tinted glasses similar to those worn by the offender inthe movie.3

4. A 17-year-old boy was convicted of murdering a woman,whom he stabbed and strangled. His defense was that he had

Crime as Imitation: Are the Media Responsible?

watched the movie A Clockwork Orange several times and asa result “did bad things.” The movie portrays a “maraudingyoung Briton with a vacant stare who acts on his violentimpulses.” The movie character and his gang associatesengaged in random rapes and murders on London’s streets.The movie was nominated for several Academy Awards butwas criticized for its portrayal of violence. After he receivedseveral death threats, the director withdrew the movie from cir-culation in Britain.

At his trial, the defendant testified that he did “badthings” only when he wore a black Clockwork Orange T-shirt, which he was wearing when he killed the victim.4

5. A 4-year-old boy accidentally shot and killed his 2-year-old sister. Authorities say the boy learned how to load a gunby watching television and learned to shoot by playing withhis friend’s air gun.5

6. The mother of a 5-year-old boy who started a fire thatkilled his younger sister claimed the boy got the idea fromthe MTV cartoon “Beavis and Butt-Head,” which promotedburning as fun. Shortly after the boy watched the cartoon,his mother caught him playing with matches. According tothe fire chief who investigated the fire, “The children admit-ted they saw it on TV and thought they could do it, too.”6

7. A 16-year-old, assisted by his 14-year-old cousin,stabbed his mother to death. The teen stated that they wereinfluenced by the “slice-and-dice” Scream movies.7

1. “Colorado Killers Inspired Youth in Georgia Shooting, Note Says,” NewYork Times (10 August 1999), p. 8.

2. “Youth Says He Got Idea for Sexual Abuse from Springer Show,” NewYork Times (8 January 1999), p. 10.

3. “Police Seize Suspect Obsessed by a Movie,” New York Times (11November 1994), p. 9.

4. “Teen Found Guilty of Murder, Blames A Clockwork Orange,” MiamiHerald (23 June 1990), p. 3B.

5. “Boy, Four, Learned How to Load Gun from TV—He Kills Sister,”Miami Herald (28 August 1991), p. 3B.

6. “Mother Blames a Deadly Fire on an MTV Cartoon,” New York Times(10 October 1993), p. 14.

7. “Teenager: ‘Scream’ Films Inspired Me to Kill Mom,” St. PetersburgTimes (15 January 1998), p. 14.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 146

and either absent or reinforced in real life, whereas inthe case of aggressive behavior, most viewers receivecontradictory messages from both sources. All viewersmay learn aggression from television, but whether theyact aggressively will depend on a variety of factors.15

The influence of television on crime has been measuredin the context of publicity about capital punishment (dis-cussed in Chapter 13). William C. Bailey found no rela-tionship between television viewing of news about capitalpunishment and the incidence of homicide. The evidencedid not support the assumption that publicizing execu-tions deters homicides, nor did it support the assumptionthat publicity encourages and increases homicides. Baileynoted that such publicity may be communicating the factthat a person convicted of homicide is not likely to be exe-cuted, thus reducing any possible deterrent effect of capi-tal punishment. But it seems clear that “the current levelof executions and media practices regarding executions inthis country neither discourage nor promote murder.”16

Social scientist Ray Surette’s analysis of the impact ofthe media on crime covers many aspects of the relation-ship between the media, especially television, and criminalbehavior. His research suggests that we are more aggressivesocially because of the mass media, but that social aggres-sion is not always criminal, and most crime is not violent.Research implies that the relationship between the mediaand crime may be indirect and that the relationship isgreater in property crimes than in violent crimes, with thepossible exception of the impact of pornography. Further,there is some evidence of short-term imitation of mediaviolence by children, but “researchers who have looked foran incorporation of violent behavior into the viewer’s over-all behavior pattern or a subsequent willingness amongchildren to use violence as a problem-solving method havereported mixed results.”17

In 1998, after visiting and working with faculty and stu-dents who had witnessed the shootings at the WestsideMiddle School in Jonesboro, Arkansas, psychology pro-fessor Dave Grossman related his beliefs concerning theviolence by young people in that incident and others.According to Grossman, the behavior of the accused killersat Jonesboro, like many others, can be explained in partby the phenomenon of media-induced violence. The boys,who killed four students and a teacher and wounded 10other people, were characterized by a lack of self-esteemand a sense of inferiority that, combined with media vio-lence, can provoke violent behavior in young people whoare “wannabes”: “They want to be tough, they want toimpress people, they want to make a bold statement andthey don’t know how. And then the media tells them how.”Grossman suggested that the media message is that “killingis the route to greatness. Killing is the route to fame.”Grossman also believes that specific types of role modelson television may precipitate similar types of violence. Healleged that school shootings by white, adolescent boys in1997 and 1998 may have been influenced by the 1995

movie The Basketball Diaries. In that movie, the characterplayed by Titanic star Leonardo DiCaprio entered aschoolroom and shot numerous teachers and children. “Indoing so,” Grossman said, “he became a role model thatother white males desire to emulate.” Movie producers andothers reject that analysis.18

The consideration of a possible impact of the media oncrime, especially among young people, is not limited tothe United States, as Global Focus 6.1 notes.

In a review of three books on the relationship betweenthe media and crime, Jana Bufkin explored the books’ cov-erage of the impact of the media on crime victimization andthe fear of crime as well as the commission of crime. Severalconclusions were drawn, some of which are noted here:

■ “[D]ata presented in the news, even when providedby seemingly unbiased sources, may be faulty andshould be approached with caution. . . .

■ [T]the news tends to focus on white, female vic-tims, suggesting to their white, female viewers thattheir crime victimization potential is greater than isactually the case. . . .

■ The media’s coverage of drug problems in our soci-ety serves as a prop for the ideology of control andit has done so throughout its history. . . .

■ Not only do these accounts project representationsthat suggest that street crimes are the greater prob-lem in Western societies, but they also portrayoffenders as crazed strangers who interact with thevictims exclusively for the purpose of attacking orharming them.”19

Whatever the impact of the media might be in terms ofimitation or copycat behavior, criminologists look at sociallearning theory as crucial to explaining the process bywhich people learn criminal behavior.

Social Learning TheorySocial learning theory in criminology is generally traced tothe contributions of noted criminologist Edwin H. Suther-land and his differential association approach.

Sutherland’s Differential AssociationTheoryAs an explanation of criminal behavior, differential asso-ciation is based on the premise that criminal behavior islearned in the same way that any other behavior is learned.The term was introduced in 1939 by Edwin H. Sutherland,sometimes referred to as the dean of American criminol-ogy, whose work has had a significant impact on sociolo-gists and criminologists in the United States.

When Sutherland was asked to write a criminology textin 1921, his primary interest was in the controversy that wasraging between environmental and hereditary theories ofbehavior. Sutherland wanted to analyze criminal behavior


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 147

by utilizing some of the prevailing sociological concepts.He was also interested in finding concrete causes of crime.As he examined the facts and issues, he decided that noconcrete variables could explain crime. For example, thecondition associated most frequently with crime is gender;that is, most people apprehended for crimes are men. ButSutherland said that it was obvious that gender was notthe cause of crime. He turned to abstract explanations anddecided that a learning process involving communicationand interaction must be the principle that would explainall types of crime.20

In the 1939 edition of his text, Sutherland introducedthe concept of differential association, which he developedinto a theory to explain the process by which an individ-ual engages in criminal behavior. Sutherland’s theory hasnine propositions.21 They are discussed here, but for easyreference to the propositions alone, see Exhibit 6.1.

First, “criminal behavior is learned.” In this proposi-tion, Sutherland emphasized his belief that criminalbehavior is not inherited. Nor did he believe that a personwho has not been trained in criminal behavior invents thatbehavior. It must be learned.


6.1

GL

OB

AL

FO

CU

S

The possibility of imitation leading to copycat crimes wasraised in 2007, after a teenager killed six students, a nurse,and the principal at his high school in Finland before com-mitting suicide. After the shooting rampage, police stated thatPekka-Eric Auvinen, 18, may have been in contact throughthe Internet with 14-year-old Dillon Cossey, in Philadelphia,Pennsylvania, who had recently been arrested for planning anattack at his school. Both were alleged to have communicatedthrough a myspace.com site that focused on the ColumbineHigh School (Littleton, Colorado) killers, Eric Harris and DylanKlebold, who killed 12 students and one teacher andwounded over 20 others before killing themselves in 1999.1

According to one media account,

More than eight years after the bloodiest high schoolmassacre in U.S. history, Columbine has become a tow-ering symbol of retribution and martyrdom amongwould-be imitators far beyond Littleton, Colo. . . .Eric Harris and Dylan Klebold have ascended to the sta-tus of folk heroes.2

Shooting Rampage in Finland: A Copycat Crime?

Shortly after the shootings in Finland, two teenagers, ages16 and 17, arrested in Stockholm, Sweden, and accused ofconspiring to murder their principal and a janitor, were thoughtto have been inspired by the Finnish shootings.3

In Australia, Jason James Cousins, 30, was accused of post-ing a threat against students at Cambridge Park High School,which he once attended, stating that he would give the Finnishkiller, whom he allegedly admired, four days of media atten-tion before he began his own rampage.4

1. “School Shooters Use MySpace to Network,” UPI Release (12 Novem-ber 2007), n.p.

2. “Columbine Both Symbol, Obsession,” Philadelphia Inquirer (15November 2007), p 1.

3. “School Murder Plot,” Sunday Times (Perth, Australia) (11 Novem-ber 2007), p. 37.

4. “Net Threat to Shoot Students,” Daily Telegraph (Australia) (14November 2007), p. 13.

Crime as a function ofimitation was raised afterPekka-Eric Auvinen, an 18-year-old Finnish student,began a shooting rampage athis high school, killing sixstudents, a nurse, theprincipal, and himself.Auvinen reportedly had beenin contact with an Americanstudent concerning otherhigh school killings,including that of Columbinein Littleton, Colorado.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 148

Second, “criminal behavior is learned in interactionwith other persons in a process of communication.” Thatprocess involves gestures as well as verbal interactions.Third, “the principal part of the learning of criminalbehavior occurs within intimate personal groups” (e.g.,families). Sutherland did not believe that the media playan important role in the process. His approach should beunderstood in the context of pre-television times.

Fourth, “the learning includes (a) techniques of com-mitting the crime, which are sometimes very complicated,sometimes very simple; (b) the specific direction ofmotives, drives, rationalizations, and attitudes.” Fifth, “thespecific direction of motives and drives is learned fromdefinitions of the legal codes as favorable or unfavorable.”Sutherland pointed out that in American society the def-initions of legal codes are mixed: Some favor violatingthose codes, while others support compliance. The mix-ture creates cultural conflict with respect to legal codes.

Sixth, “a person becomes delinquent because of anexcess of definitions favorable to violation of law over def-initions unfavorable to violation of law.” This statementis Sutherland’s principle of differential association. Thosewho engage in criminal behavior do so because they havecontacts with that type of behavior and are isolated fromanticriminal behavior.

Seventh, “differential associations may vary in fre-quency, duration, priority, and intensity.” This statementis a crucial one in Sutherland’s theory, and it means thatassociations with criminal and noncriminal behavior varyin terms of those four elements. Sutherland stated that thefirst two, frequency and duration, did not need to bedefined; their meaning is obvious. He suggested that pri-ority is important in that behavior learned early in lifemight persist for a long time, but that is not always thecase. In his view, “priority seems to be important princi-

pally through its selective influence.” Finally, intensity “hasto do with such things as the prestige of the source of acriminal or anti-criminal pattern and with emotional reac-tions related to the associations.” Sutherland acknowledgedthat it would be difficult to develop a precise formula tomeasure all of these four modalities.22

Eighth, “the process of learning criminal behavior byassociation with criminal and anticriminal patternsinvolves all of the mechanisms that are involved in anyother learning.” To Sutherland, the same learning theorythat explains noncriminal behavior explains criminalbehavior.

Ninth, “while criminal behavior is an expression of gen-eral needs and values, it is not explained by those generalneeds and values, since noncriminal behavior is an expres-sion of the same needs and values.” To Sutherland, theattempt to explain criminal behavior in terms of a person’sneeds or desires was as futile as trying to explain the behav-ior in terms of respiration: Breathing (like a need or desire)does not distinguish criminal from noncriminal behavior,although it is necessary for both.

Sutherland has had a “massive impact on criminology,”according to some critics, such as George B. Vold andThomas J. Bernard.23 In one of his efforts to test a portionof Sutherland’s theory, James F. Short Jr. called the theory“the most truly sociological of all theories which have beenadvanced to explain criminal and delinquent behavior.”24

Short, however, pointed out some of the problems withthe theory, which he said was not testable in its generalterms; some reformulations were necessary. He attemptedto measure the terms frequency, duration, priority, andintensity. Within the limitations of his study, Short foundstrong support for Sutherland’s theory, although in a laterstudy he concluded that a reformulation might be neces-sary before the theory’s concepts could be measured.25


6.1E

XH

IBIT 1. Criminal behavior is learned.

2. Criminal behavior is learned in interaction withother persons in a process of communication.

3. The principal part of the learning of criminalbehavior occurs within intimate personal groups.

4. When criminal behavior is learned, the learningincludes (a) techniques of committing the crime,which are sometimes very complicated, sometimesvery simple; (b) the specific direction of motives, drives, rationalizations, and attitudes.

5. The specific direction of motives and drives islearned from definitions of the legal codes as favor-able or unfavorable.

6. A person becomes delinquent because of an excessof definitions favorable to violation of law over def-initions unfavorable to violation of law.

7. Differential associations may vary in frequency,duration, priority, and intensity.

8. The process of learning criminal behavior by asso-ciation with criminal and anticriminal patternsinvolves all of the mechanisms that are involved inany other learning.

9. While criminal behavior is an expression of generalneeds and values, it is not explained by those gen-eral needs and values since noncriminal behavior isan expression of the same needs and values.

Source: Edwin H. Sutherland, Principles of Criminology, 4th ed. (Phila-delphia: Lippincott, 1947), pp. 6-7 (emphasis omitted).

The Nine Propositions of Sutherland’s Differential Association Theory

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 149

One early attempt to test differential association bymeasuring actual delinquency as reported by best friendsled Albert J. Reiss Jr. and Albert L. Rhodes to conclude:“We are led to question the postulate that differential asso-ciation is a necessary and sufficient condition explainingdelinquency.”26 In 1944, Sutherland considered this andother criticisms in a paper that was not published untilafter his death. Sutherland acknowledged that some of thecriticisms were valid, and he concluded that criminal asso-ciations alone do not explain criminal behavior: “Rather,it is those associations plus tendencies toward alternateways of satisfying whatever needs happen to be involvedin a particular situation.”27

Donald R. Cressey, who revised the text after Suther-land’s death in 1950, acknowledged problems with differ-ential association. The theory, he wrote, “is neither precisenor clear. . . . Most significantly, the published statementgives the incorrect impression that there is little concernfor accounting for variations in crime and delinquencyrates. This is a serious error in communication on Suther-land’s part.”28 Cressey believed that the theory needed tobe reformulated, but he made no attempt to do so in sub-sequent editions of the text, stating: “The theory is presentlyin a period of great popularity. . . . It would be inappro-priate to modify the statement in such a way that researchwork now in progress would be undermined.”29

Cressey analyzed the criticisms and defended the theoryagainst some of the attacks. He stated that one result of thetheory’s ambiguity and the critics’ failure to read the theorycarefully was the assumption that people become criminalsbecause of their association with criminals or criminal pat-terns of behavior and attitudes. But, observed Cressey, thetheory is that people become criminals because of an over-abundance of associations with criminal, as compared withanticriminal, behavior patterns.30

Some critics argued that a person can become a criminalwithout associating with criminals; therefore, differentialassociation does not apply. That was not the point, saidCressey. One may be exposed to criminal attitudes andbehavior without being exposed to criminals. For example,parents who teach their children not to steal may suggest thatit is permissible to steal a loaf of bread if they are starving.

Differential association has produced extensive researchfindings; one of the most consistent among these in delin-quency literature is the relationship between peer associ-ations and delinquency. The influence of Sutherland isseen in empirical research concerning this relationship.31

Mark Warr reported that the relationship between peerassociations and delinquency involves several aspects ofpeer relations, all of which are consistent with Sutherland’sposition as well as with explaining (at least in part) therelationship between age and criminal behavior:

1. Differential exposure to delinquent peers, meaningthe number of delinquent peers reported byrespondents at different ages.

2. Time spent in the company of peers.

3. The importance of friends to respondents.

4. Respondents’ commitment or loyalty to their ownparticular set of friends.32

Mark Warr and Mark Stafford emphasize that althoughpeer attitudes affect delinquency, peer behavior is evenmore important, suggesting that delinquency does notresult primarily from the influence of attitudes acquiredfrom peers: “Rather, it more likely stems from other sociallearning mechanisms, such as imitation or vicarious rein-forcement, or from group pressures to conform.” It is sug-gested that for purposes of analysis, attitudes and behav-ior should be separated, in that “what peers do appears tobe at least as important as what they think.”33

Earlier reactions focused on the theory’s developmentand refinement; more recently, scholars have focused ontesting it or rejecting it and replacing the theory with con-trol or integrated theories, discussed later in this chapter.Some argue that differential association theory is soundbut that further developments are needed. For example, itis difficult to measure the content of definitions favorableto law violations. In general, according to Ross L. Mat-sueda, the theory “appears supported, but requires addi-tional research to specify the concrete content of itsabstract principles.”34

Of the suggestions for revising Sutherland’s theory, themost often cited and tested is Ronald L. Akers’s learningtheory, which Akers developed more thoroughly after hiswork with Robert L. Burgess.

Akers’s Social Learning TheoryIn an effort to provide a “more adequate specification ofthe learning process” required in the theory of differentialassociation, Burgess and Akers developed a concept of dif-ferential association-reinforcement. Their purpose was tointegrate Sutherland’s theory with the principles of mod-ern behavior theory. They assumed that in doing so, theycould render differential association propositions moretestable and at the same time make the learning processeseasier to understand.35

According to Akers, his social learning theory hasundergone some revision. Today, he views it as based onfour basic concepts, which he and Christine Sellers defineas follows:

1. “Differential reinforcement refers to the balance ofactual or anticipated rewards and costs, both socialand nonsocial, that follow a behavior.

2. Imitation refers to engagement in behavior afterobserving similar behavior in others who havemeaning to the individual.

3. Definitions refer to one’s own attitudes that definethe commission of an act as right or wrong, justi-fied or unjustified.


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 150

4. Differential association refers to direct and indirectexposure, via associations with others, to patterns ofbehavior as well as to patterns of norms and values.”36

Social learning occurs first in a process of differentialassociation. The person interacts and identifies withgroups (peer groups, family, schools, churches, and so on)that provide models for social reinforcements and behav-ior. The individual develops attitudes toward the behav-iors and attaches definitions to those behaviors, definingthem as good or bad, acceptable or unacceptable. Thesedefinitions help reinforce behavior and serve as cues forbehavior. The more positive the definitions people have ofa given behavior, the more likely they are to engage in it.According to Akers and Sellers, some definitions that arefavorable to violating the law are so strong (e.g., radicaldefinitions held by terrorists) that they essentially requireone to follow them. But, in most cases, definitions that arefavorable to violating the law do not act as direct motiva-tions to commit crimes. “Rather, they are conventionalbeliefs so weakly held that they provide no restraint or arepositive or neutralizing attitudes that facilitate law viola-tion in the right set of circumstances.”37

A person may associate with numerous groups, and thenorms, values, and behaviors of the groups may differ; somemay be deviant, others conforming. The direction chosenby an individual is determined by the “relative balance ofreinforcements, definitions, and behavioral models.”38

Akers and his students and colleagues have conductedmany tests of this social learning theory, including such areasas smoking among adolescents, deviant drinking among theelderly, and rape and sexual coercion. According to Akersand Sellers, “the findings . . . demonstrated that the sociallearning variables of differential association, differential rein-forcement, imitation, and definitions, singly and in combi-nation, are strongly related to the various forms of deviant,delinquent, and criminal behavior studied.”39

Akers and Sellers recognize that social learning theoryhas its critics and its limitations, but they contend that itis the “most strongly and consistently supported” of thesocial psychological theories advanced to explain delin-quency and crime.40

In recent years, Akers has expanded his social learningtheory into the social structure and social learning (SSSL)model, which combines social structure and social process.He advances four dimensions of social structure. First isdifferential social organization, which refers to the demo-graphic variables in a given area—such as age and popu-lation density—that affect crime rates. Second is differen-tial location in the social structure, which encompasses thecharacteristics of individuals, such as race and ethnicity,class, gender, marital status, and age, that determine theirlocations within groups. The third dimension of the socialstructure consists of theoretically defined structural vari-ables, which includes anomie, group conflict, social disor-ganization, and other concepts that have been used by

other theorists to identify social-structural conditions thatmay be related to high delinquency and crime rates. Thefourth, and final, dimension is differential social location,which refers to memberships in primary, secondary, andreference groups such as the family or peer groups. Akersand Sellers conclude:

The differential social organization of society and com-munity, as well as the differential location of persons inthe social class, race, gender, religion, and other struc-tures in society, provides the general learning contextsfor individuals that increase or decrease the likelihoodof their committing crime.41

Akers and Sellers reported their findings after reviewingthe research on social learning theory, concluding that “thegreat preponderance of research conducted on social learn-ing theory has found strong relationships in the theoreti-cally expected direction between social learning variablesand criminal, delinquent, and deviant behavior.” Akers andSellers cite numerous studies that demonstrate the impor-tance of peers in a youngster’s development of delinquentand criminal behavior, including the following:

■ “There is abundant evidence to show the significantimpact on criminal and deviant behavior of differ-ential association in primary groups such as familyand peers. . . .

■ [I]n general, parental deviance and criminality arepredictive of the children’s future delinquency andcrime.

■ Moreover, youngsters with delinquent siblings(especially same-sex older siblings) in the family aremore likely to be delinquent, even when parentaland other family characteristics are taken intoaccount. . . .

■ Other than one’s own prior deviant behavior, thebest single predictor of the onset, continuance, ordesistance [ceasing, stopping] of crime and delin-quency is differential association with conformingor law-violating peers. . . .

■ Virtually every study that includes a peer associa-tion variable finds it to be significantly and usuallymost strongly related to delinquency, alcohol anddrug use and abuse, adult crime, and other forms ofdeviant behavior.”42

Akers and Sellers noted the importance of associationwith gangs to delinquent and criminal behavior and con-cluded that the importance of social learning through dif-ferential association on delinquent behavior “is among themost fully substantiated and replicated findings in crimi-nology,” with only the factors of age and gender so con-sistently associated with crime and delinquency.43

Akers’s social learning theory continues to generateresearch and reassessment. Despite its popularity, how-ever, there are problems, as he and his collaborators have


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 151

noted. Among the questions the theory raises is, whichcomes first? Do young people become delinquent becausethey are reinforced in that type of behavior by delinquents,or are they delinquent and seek the company of others whobehave similarly? But, state Akers and Sellers, the researchaccounts for the reciprocal relationship between the con-duct of a person and his or her association with others.And social learning theory purports that through associa-tions, one learns approved as well as disapproved behav-ior. Social learning theory would come into serious ques-tion only if it were found that “the onset of delinquencyalways or most often predates interaction with peers whohave engaged in delinquent acts and/or have adhered todelinquency-favorable definitions.” The theory would alsonot be supported “if the research evidence showed thatwhatever level of delinquent behavioral involvement pre-ceded association with delinquent peers stayed the sameor decreased rather than increased after the association.”This has not been the case, however, according to Akersand Sellers.44

Other researchers agree. According to Mark Warr,“Although many investigations offer evidence of recipro-cal effects, no study yet has failed to show a significanteffect of peers on current and/or subsequent delin-quency.”45 But Warr also offers this criticism: “[T]he gen-erality of the theory has yet to be fully demonstrated. Tobe sure, the evidence of social learning is extensive andimpressive . . . but it is concentrated disproportionatelyon tobacco, alcohol, and other drug use, and on relatively

minor forms of deviance.”46 Akers and Sellers react to thatstatement by arguing that, with the possible exception ofself-control theory (discussed later in this chapter), sociallearning theory has a broader scope than other crimino-logical theories.47

This discussion of learning theories is summarized inExhibit 6.2.

Control TheoryThis section explores approaches that take a quite differentapproach from those discussed previously. Control theoryrefers to a group of theories or approaches that focus onexplaining why people obey the law. The scholars devel-oping this area of concern believe that it is natural for peo-ple to want to commit crimes; we have to be motivated notto do so. People who obey the law do so because they areresponding to appropriate social controls. They are social-ized to obey rather than to disobey the law, and this social-ization process requires a lot of work. Those who violatethe law do so because the social controls are not working;their bonds to law-abiding persons have been broken orwere never developed. Thus, criminals are not differentkinds of people; they are not people who learned the wrongkind of behaviors; they are people who live in the wrongneighborhoods; they are people for whom social controlshave, for whatever reason, been ineffective. Crime resultswhen social controls are weakened or broken down; whencontrols are strong, crime does not occur. The problem is


6.2

EX

HIB

IT Tarde’s Imitation TheoryGabriel Tarde was a forerunner in developing theoriesabout what are today called copycat crimes. He rejectedthe biological and psychological theories but did notbelieve in social determinism; rather, he said, behavior isthe process of imitation. Tarde articulated three laws ofimitation: (1) People imitate each other in proportion totheir close contact; (2) the inferior imitates the superior;and (3) when two mutually exclusive behaviors occurtogether, one can be substituted for the other, resulting inthe decline of the use of the older and an increase in theuse of the newer.

Sutherland’s Differential AssociationBased on nine propositions (see Exhibit 6.1), Edwin H.Sutherland’s theory advocates that all criminal behavior islearned, not inherited. It is learned through social interac-tion with others in the same way that noncriminal behav-ior is learned. The learning occurs primarily in small groupsand includes techniques as well as motives and drives.Criminal behavior basically occurs when the individual has

more definitions favorable to a deviant life than to a law-abiding life.

Akers’s Social Learning TheoryRonald L. Akers (joined in some of his early work by RobertL. Burgess) expanded differential association theory toinclude the principles of modern behavior theory. Behavioris strengthened by positive reinforcement and weakened bynegative reinforcement—thus his differential association-reinforcement theory. Individuals learn these reinforcementsfrom peers, families, schools, and other institutions.

Akers’s social learning theory develops four concepts:differential reinforcement, imitation, definitions, and dif-ferential association. More recently, Akers expanded hisposition to include four social-structural dimensions: dif-ferential social organization (such as age); differential loca-tion in the social structure (such as social class); theoret-ically defined structural variables (such as anomie or groupconflict); and differential social location (e.g., membershipsin primary, secondary, and reference groups, such as thefamily or peer groups).

Learning Theory: A Summary

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 152

to try to explain what can be done in a positive way to elicitappropriate behavior. The issue is not “why do they do it?”but, rather, “why don’t we do it?” According to TravisHirschi, whose social control approach is discussed here,“There is much evidence that we would if we dared.”48

The question, then, is not how to prevent criminalbehavior but how to train people to engage in law-abidingbehavior. Social control theorists take a variety of ap-proaches.

Reiss and Nye’s Control TheoryOne of the first to apply control theory to delinquent orcriminal behavior was Albert J. Reiss Jr., who maintainedthat such behavior results from the failures of personal andsocial controls. Personal controls are internalized, whilesocial controls result from formal controls (such as laws)and informal controls (such as social sanctions from par-ents, schools, etc.).49

Shortly after Reiss’s work was published, sociologist F. IvanNye analyzed delinquency using three control categories:

1. “Direct control, by which punishment is imposedor threatened for misconduct and compliance is re-warded by parents.

2. Indirect control, by which a youth refrains fromdelinquency because his or her delinquent act mightcause pain and disappointment for parents or otherswith whom the youth has a close relationship.

3. Internal control, by which a youth’s conscience orsense of guilt prevents him or her from engaging indelinquent acts.”50

Although Nye recognized that formal institutions, suchas law, could influence social control, he believed that theinformal institutions, such as the family, were the mostinfluential. Thus, the more an adolescent’s needs are metwithin the family, the less likely the adolescent will be tolook for satisfaction in unacceptable ways outside the fam-ily. Nye reported that he found associations between delin-quent behavior and such family situations as lack of dis-cipline, broken or disintegrated homes, lack of sufficientparental affection toward the children, and even rejection.These associations were weak, however, and Nye has beencriticized for not clearly articulating exactly “how a par-ticular aspect of family relationships included in his studywas connected to the concepts in his control theory.” Fur-thermore, for some of the family attributes, such as socialclass, Nye did not show any connection to delinquency.51

Reckless’s Containment TheoryPerhaps the best known of the early control theories is Wal-ter C. Reckless’s containment theory, which stresses thatwe live in a society that provides a variety of opportunitiesfor conformity or nonconformity. Not everyone choosesthe illegal opportunities; thus social-structural theories that

stress the availability of illegal and legal opportunities, theexistence of a subculture, the location of goods and serviceswithin the city, population density, and other variables can-not explain all criminal behavior. What we need to knowis why those phenomena affect some people and not oth-ers. That is, why are some of us immune to such influencesin that our exposure to them is not followed by criminalbehavior? Reckless suggested that the answer lies in con-tainment theory, which he defined as follows:

The assumption is that there is a containing externalsocial structure which holds individuals in line and thatthere is also an internal buffer which protects peopleagainst deviation of the social and legal norms. The twocontainments act as a defense against deviation from thelegal and social norms, as an insulation against pressuresand pulls, as a protection against demoralization andseduction. If there are “causes” which lead to deviantbehavior, they are negated, neutralized, rendered impo-tent, or are paired by the two containing buffers.52

There are two types of containment: outer containmentand inner containment. Outer containment (or externalcontrol) might be called social pressure, and in simple soci-eties this kind of social control works well. The commu-nity’s social norms are taught to new members, who inter-nalize them and are restrained by the community’sreaction to any violation of these norms. Social ostracismmay be the most effective social control in simple societiesor communities, but as societies become more complex,outer containment loses its effectiveness. People mustdevelop inner containment mechanisms by which to con-trol their own behavior.

Inner containment (or internal control) refers to the abil-ity to direct oneself, which is related to one’s self-concept.According to Reckless, “One of the components of capabil-ity of self is a favorable self-image, self-concept, self-perception. The person who conceives of himself as aresponsible person is apt to act responsibly.”53 A high goallevel, especially regarding societal goals, and a high aspira-tion level geared to society’s expectations are essential com-ponents of the self. Frustration tolerance and identificationwith society’s values and laws are important. The oppositeof this is alienation—the release of inner containment.

Reckless emphasized that the components of externaland internal containment are buffers, not causes. Theyoperate to help the individual refrain from succumbing topressures to violate laws. If the buffers are strong, the indi-vidual is law-abiding; if they are weak, he or she commitsa crime. The buffers help to neutralize the norms.

Despite the claims of Reckless and his associates thatcontainment theory explains most delinquency and crime;that the theory might bring psychologists and sociologiststogether in the study of crime because it involves both dis-ciplines; and that the theory, unlike many others, can beused in an individual case history,54 the theory has beencriticized severely. It cannot explain why people who do


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 153

the same things are labeled differentially; it is limited in itspredictive ability; and it includes a questionable measureof self-concept, along with a lack of control groups in someof the early works. The difficulty of measuring the strengthor weakness of external and internal containment is aproblem. Further, the theory does not explain why somechildren with bad self-concepts are not delinquent.

Containment theory may be most useful when com-bined with other approaches. In a comparison of con-tainment theory with differential association theory, Har-win L. Voss concluded that the two theories are similar,except that differential association emphasizes the processof differential association, whereas containment theoryemphasizes the product of socialization—the self-concept.Both of these theories can account for delinquency morefully than either can separately.55

Travis Hirschi refined the elements of control theory,as noted later. But first it is important to recall one othertheory of delinquent behavior that is considered by someas a control theory: neutralization and drift. This theorywas discussed in Chapter 5 (see again p. 111) as a socialstructure theory and should be reviewed here.

Hirschi’s Bonding TheoryIn their review and critique of control theories of criminalbehavior, Akers and Sellers stated: “All of the earlier con-trol theories were superseded by the version proposed byTravis Hirschi, who remains today the major control the-orist.” Hirschi’s theory “has come to occupy a central placein criminological theory. Indeed, it is the most frequentlydiscussed and tested of all theories in criminology.”56

Hirschi’s control theory focuses on social bonding.According to Hirschi, it is conforming behavior, notdeviance, that we need to explain. The real question is why,

with so many opportunities and pressures to commitcrimes, most of us are law-abiding citizens most of thetime. The basic concept of control theory, asserts Hirschi,is the individual’s bond to society; delinquency is morelikely when that bond is weakened. The bond has fourcomponents:

■ Attachment to conventional persons

■ Commitment to conventional behavior

■ Involvement with conventional people

■ Belief in conventional norms57

Attachment refers to the feelings we have toward oth-ers. If we have close ties to them, we are more likely to carewhat others think of our behavior; likewise, if we do nothave close ties, we care less. The controlling of delinquencyis tied to the attachments young people have to their par-ents. But Hirschi emphasized that attachment to peers mayalso control delinquency, even when those peers are notalways law-abiding. It is the attachment that is mostimportant; it is the lack of attachment that is conduciveto delinquency.

Commitment refers to the investment one has in activ-ities, such as getting an education. A person must measurethe extent to which that investment would be lost bydeviant behavior before he or she engages in deviance.Thus, commitment to conventional behavior, such asgoing to school, might be sufficient to cause someone toavoid jeopardizing that education by engaging in delin-quent behavior. According to Hirschi, most people acquiregoods and services that they do not want to lose; the riskof losing those is “society’s insurance that [people] willabide by the rules.”58

Most people are engaged in conventional activities, andthis involvement does not permit them sufficient time to


Happy family gatherings areimportant in the positivesocialization of children. In thispicture four generations celebrate the little girl’s birthday.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 154

engage in delinquent or nonconventional activities.Hirschi notes: “Many people undoubtedly owe a life ofvirtue to a lack of opportunity to do otherwise.”59 Finally,by belief Hirschi means that a person accepts society’s con-ventional norms and rules, believing that those generalrules and laws (although not necessarily each specific one)are correct and should be obeyed. But individuals vary inthe extent to which they have these beliefs, and “the less aperson believes he should obey the rules, the more likelyhe is to violate them.”60

Hirschi tested his theory on a sample of Californiayouths using the self-report method of collecting data (dis-cussed and analyzed in Chapter 2). The 4,000 junior andsenior high school students were given questionnairesdesigned to measure their attitudes toward friends, neigh-borhood, parents, school, teachers, and human relations.The students were asked to respond to the following sixquestions, indicating whether they had (1) never commit-ted the offense, (2) committed the offense more than oneyear ago, (3) committed the offense during the past year,or (4) committed the offense during the past year as wellas more than a year ago:

1. Have you ever taken little things (worth less than$2) that did not belong to you?

2. Have you ever taken things of some value (between$2 and $50) that did not belong to you?

3. Have you ever taken things of large value (worthover $50) that did not belong to you?

4. Have you ever taken a car for a ride without theowner’s permission?

5. Have you ever banged up something that did notbelong to you on purpose?

6. Not counting fights you may have had with abrother or sister, have you ever beaten up on any-one or hurt anyone on purpose?61

The responses to these questions were used as an indexof self-reported delinquency. In addition, questions wereasked about work, money, expectations, aspirations, par-ticipation in school activities, and use of leisure time.School records, including grades, and police records werealso used as sources of data for the study.

The high association between low socioeconomic classand crime, found in the earlier studies using official crimedata, had been questioned by self-report studies. Hirschifound strong evidence that this traditional association doesnot exist. Indeed, he noted very little association betweensocial class, as measured by the father’s occupation, andadmitted or official delinquency, with the exception of alow incidence, by both measures, of delinquency amongthe professionals’ sons. Hirschi did find, however, thatboys “whose fathers have been unemployed and/or whosefamilies are on welfare are more likely than children fromfully employed, self-sufficient families to commit delin-quent acts.” Hirschi’s research also disclosed that positiveattitudes toward teachers and school were related to non-delinquent behavior. Hirschi suggested that the closer theties to parents, the less likely it was that the youths wouldengage in delinquent acts. It was not the parent’s statusthat was important but, rather, the child’s attachment tothe parent.62

Hirschi concluded that young people who are not veryattached to their parents and to school are more likely to bedelinquent than are those who have these kinds of attach-ments. He also found that youths who have positive atti-tudes toward their own accomplishments are more likely tobelieve in the validity and appropriateness of conventionallaws and the moral rules of society than are youths who arenegative about their own accomplishments.

Although Hirschi’s California study was based on asample of urban respondents, a study of a rural sample inNew York state, conducted by Michael Hindelang, found


The impact of schools goes beyondthe academic education of students.These high school students arediscussing family and health issues inan informal school setting.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 155

support for Hirschi’s control theory.63 However, Hinde-lang “failed to replicate a positive relationship betweenattachment to parents and attachment to friends. . . .[H]e failed to show that attachment to friends increasesthe likelihood of delinquent behavior.” In fact, Rand D.Conger noted that Hindelang “found a slightly positiverelationship between identification with peers and delin-quency which is unexplainable in terms of control theory.”This does not mean that the theory is incorrect—only thatit is incomplete. We need to go beyond attachment to peersand discover the type of peer to whom the individual isattached before the analysis has validity for the predictionof delinquent behavior.64

Delbert S. Elliott and Harwin L. Voss found that thetype of peer did not matter,65 while Eric Linden and JamesC. Hackler reported that the likelihood of delinquentbehavior seems to be decreased by attachment to conven-tional peers but increased by attachment to delinquentpeers.66 Bill McCarthy, Diane Felmlee, and John Haganfound that the gender of friends matters in deterring prop-erty crimes, with the friendships showing more influenceover females than males. Females are more likely to havethe type of friendships that Hirschi and Matza believe pro-vides the bonding that deters delinquent behavior: Theirfriendships are more intimate than those of males, they aremore likely to occur in settings with more social control,they engage in fewer risky behaviors than males, and theyare more likely to discourage offending.67

Hirschi recognized that his control theory could notescape unscathed. In the first place, his theory “underes-timated the importance of delinquent friends; it overesti-mated the significance of involvement in conventionalactivities.” Hirschi decided that probably one should lookat the relationship between delinquent activities and theperson’s self-concept or self-esteem. That relationshipmight be important in explaining “the potency of theadult-status items, such as smoking, drinking, dating, anddriving a car.” Although control theory can help us under-stand these relations, Hirschi noted, it leaves a lot unex-plained. However, he concluded this early work on anoptimistic note with regard to his theory: “I am confidentthat when the processes through which these variablesaffect delinquency are spelled out, they will supplementrather than seriously modify the control theory, but thatremains to be seen.”68

Perhaps a more serious challenge to Hirschi’s theoryhas been raised by those who emphasize that most tests ofthe theory have used cross-sectional data at a particulartime rather than measuring the relationship of variablesover time. Robert Agnew sampled young men at the begin-ning of their sophomore year in high school and again nearthe end of their junior year and suggested that Hirschi’scontrol theory is more limited than he claims. Agnew’sstudy showed that the theory does not predict delinquencyamong middle to older adolescents, although it might beimportant in predicting delinquency among younger ado-

lescents.69 Similar findings were reported by Randy L.Lagrange and Helene Raskin White, who concluded thatadolescence should not be viewed as a unitary period but,rather, should be refined further by the factor of age.70

Another criticism of Hirschi’s control theory, discussedhere as a social-process theory, comes from one ofHirschi’s former doctoral students, Thomas J. Bernard,who argued that the four elements of the theory are notall social-psychological variables, as Hirschi suggests.According to Bernard, commitment is a social-structuralvariable that differs from the other three variables ofHirschi’s theory—attachment, involvement, and belief—which are social-psychological factors. Commitment“confuses the definition and the explanation of confor-mity” and “is properly conceived as an antecedent variablethat measures the role of social structure in the origin ofdelinquency.”71

Bernard argues that viewing commitment as a social-structural rather than a social-process variable would elim-inate the current perception in Hirschi’s theory that “theindividual’s bonding to society is independent of socialstructural characteristics, and therefore can be enhancedwithout broader societal changes.”72

Despite his suggestions for revision of Hirschi’s controltheory, Bernard emphasizes that none of his ideas should“challenge the substantial contribution of Hirschi’s theoryto criminology [which has] generated a host of studies thathave advanced criminology as a science, whether or notthey support Hirschi’s conclusions.”73

Stephen A. Cernkovich and Peggy C. Giordano foundsupport for Hirschi’s approach. For example, control the-ory suggests that the greater the degree of school bonding,the less likihood there is of delinquency. It is argued thatthis occurs despite the race, gender, and socioeconomicstatus of the juveniles. This proposition was examinedamong African-American juveniles, with the result thatbonding was found to be important and not influencedsignificantly by race. Cernkovich and Giordano concluded,however, that it was premature to exclude race as a factorin an analysis of school bonding and delinquency. Theystated that more research is needed before we concludethat race is or is not a relevant factor in explaining deli-quency. These conclusions challenged the belief that theschool environment of African-American juveniles is moredysfunctional than that of whites: “The black youths in oursample are not isolated and alienated from school; ratherthey are just as committed and attached as whites, if notmore so.”74

The influence of Hirschi’s bonding theory can be seenin an analysis that found that communal school organiza-tion and student bonding are predictive of the lack of stu-dent disorder, delinquency, and victimization. Allison AnnPayne, Denise C. Gottfredson, and Gary D. Gottfredsondefined communal school organization as the “organizationof a school as a community, as indicated by supportive rela-tionships between and among teachers, administrators, and


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM 6

students, a common set of goals and norms, and a sense ofcollaboration and involvement.” Schools with high com-munal organization have better morale among both stu-dents and faculty, and students bond with the faculty to agreater extent. Such schools have fewer behavior problemsamong students, less delinquency, and fewer victimizationsof teachers and students by other students. The researchersmade the analogy of the communal school organization tothat of Sampson’s concept of neighborhood collective effi-cacy, which was discussed in Chapter 5.75

A final tribute to Hirschi is one noted in the 2003 pres-idential address to the American Society of Criminology,in which John H. Laub examined what he called the fiveturning points in the development of criminology. One ofthose was the contributions of Travis Hirschi in his 1969publication, Causes of Delinquency. According to Laub,“Over the last 25 years, more than [the works of] any otherscholar, Travis Hirschi’s work has dominated intellectualdiscussion and substantially formed the research agendafor the field of criminology.” This work, noted Laub, hasbeen referred to by others as a “turning point in deviancyresearch,” “a highly significant piece of research,” and “awork of major consequence.”76

Gottfredson and Hirschi’s Self-ControlTheoryIn recent years, Hirschi has moved from his original socialbonding theory and collaborated with Michael Gottfred-son in developing a general theory of crime that focuses onself-control and purports to explain all acts of deviancy andcriminal behavior by all persons. In their 1990 book devel-oping their general theory, Gottfredson and Hirschi re-ferred to self-control as the “differential tendency of peopleto avoid criminal acts whatever the circumstances in whichthey find themselves.”77 In a later publication, they weremore specific, referring to the “tendency to avoid actswhose long-term costs exceed their momentary advan-tages.”78 The implication apparently is that the new defi-nition places an emphasis on all consequences rather thanon just long-term consequences.79

Gottfredson and Hirschi reject the traditional theoriesof crime, arguing that in an attempt to answer the ques-tion of what causes crime each discipline looks to its owncentral concepts. “Thus sociology looks to social class, cul-ture, and organization; psychology looks to personality;biology looks to inheritance; and economics looks toemployment or work.”80

According to Gottfredson and Hirschi, all these expla-nations are incompatible with the nature of crime, whichthese researchers define as “an act of force or fraud under-taken in pursuit of self-interest.”81 Crime is distinguishedfrom criminality, which refers to a person’s predispositionto commit a crime. Criminality is a part of the personal-ity trait that is central to the thesis of this approach, whichis the concept of low self-control.

Gottfredson and Hirschi view behavior in terms of theclassical view discussed in Chapter 3. People seek pleasure andavoid pain. Crime is a fast and easy way to attain quick plea-sure; the issue is not what causes crime but what constrainsit, and, say these scholars, the answer is low self-control.

Persons with low self-control have several characteris-tics. First, they seek immediate gratification, with a “here-and-now” attitude, in contrast to persons with high self-control, who are able to defer gratification. Second,persons with low self-control look for easy or simple waysto gratify their desires. Crime provides “money withoutwork, sex without courtship, revenge without courtdelays.” Third, they find in crime their need for acts thatare exciting, risky, or thrilling. Criminal acts involve“stealth, danger, speed, agility, deception, or power.” Incontrast, persons with high self-control “tend to be cau-tious, cognitive, and verbal.” Fourth, persons with low self-control have little stability in their lives, for crimes do notprovide stability in the form of a job or a career, friend-ships, or family ties. Little skill or planning is required topursue crime; thus, persons with low self-control do notneed or value cognitive or academic skills. Fifth, peoplewith low self-control are “self-centered, indifferent, orinsensitive to the suffering and needs of others.” Thus, theycan handle the fact that crime causes pain or discomfortfor its victims. This does not mean that those with low self-esteem are “routinely unkind or antisocial. On the con-trary, they may discover the immediate and easy rewardsof charm and generosity.” Finally, persons with low self-control have little tolerance for frustration and tend torespond to situations physically rather than verbally.82

Gottfredson and Hirschi maintain that low self-controlis primarily a product of poor parenting. On the contrary,when parents are supportive, show affection, and invest intheir children, they may produce “a child more capable ofdelaying gratification, more sensitive to the interests ofothers, more independent, more willing to accept re-straints on his activity, and more unlikely to use force orviolence to attain his ends.”83

Although self-control may improve over time as theresult of better parenting or the influence of other socialinstitutions, in most individuals, low self-control persiststhroughout life. Persons with low self-control are muchmore likely than persons with high self-control to engagein criminal activities throughout their lives.84 Low self-control does not cause crime, but it interacts with crimeopportunities. When an opportunity to commit a crimeexists, a person with high self-control is not likely to com-mit the act; a person with low self-control may succumb.

Gottfredson and Hirschi claim that their theory is ageneral one in that it “explains all crime, at all times, and,for that matter, many forms of behavior that are not sanc-tioned by the state.” That is a big claim, but the authorsnote that “modesty per se is not a virtue of a theory.”85

Self-control theory has generated extensive researchand theory development, with most studies supporting the


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 157

theory. For example, Christine S. Sellers found some evi-dence that self-control may explain variables, such as drugand alcohol abuse, academic probation, excessive party-ing, and friendship loss, that are associated with courtshipviolence. These variables may be indicators of low self-esteem and low self-control. In addition, the theory mayexplain the fact that many persons who are involved incourtship violence are from homes in which domestic vio-lence was experienced.86

In a test of self-control theory, Carter Hay examinedparenting and its role in the development of a child’s self-concept. Hay emphasized that Gottfredson and Hirschi“argue that the principal cause of individuals’ low self-control is ineffective child rearing.” In particular, the chil-dren of parents who do not watch them closely and whodo not recognize and punish the child’s deviant behaviorwill have low self-control. Those children are more likelyto become deviant or delinquent than are their counter-parts with higher self-control. But few studies have beenconducted on the role of parenting in the development ofself-control. In seeking to remedy this situation, Hay basedhis study on the theory of an early psychological approachto self-control, that of authoritative parenting, referring toparents who are both demanding and responsive. Hetested two hypotheses regarding parenting and concludedthat there is support for self-control theory but that thesupport was not overwhelming.87

Other scholars have also looked at parenting and self-control theory. James D. Unnever, Francis T. Cullen, andTravis C. Pratt selected a sample of 2,472 students andexamined the relationship of parental management, atten-tion deficit hyperactivity disorder (ADHD) (for a discus-sion, see Chapter 4) and delinquent behavior. They foundthe following:

■ “First, low self-concept was a strong predictor ofboth self-reported delinquency and self-reportedarrests.

■ Second, parental monitoring not only increasedself-control, but had direct effects on both measuresof delinquency.

■ Third, the effects of ADHD on delinquency werelargely through low self-control.”88

Carter Hay and Walter Forrest recently tested the sta-bility thesis of social-control theory—that is, that low self-control not only is developed early in life but also persistsover time. They also looked at the role that parent social-ization (according to Gottfredson and Hirschi, the mainexplanation for differences in self-control) has in the self-control of adolescents. With regard to the first issue, theseresearchers found moderate stability of self-control in theshort run but less so in the long run, for most but not allof their sample. Thus, one’s self-control at age 7 is moresimilar to the self-control that one has at age 9 than toone’s degree of self-control at age 15. The researchers also

found that “parenting still matters for self-control beyondchildhood.” They concluded that overall, there is moder-ate support for self-control theory.89

Gottfredson and Hirschi’s self-control theory of crimehas also been used to explain criminal victimization. EricA. Stewart and his colleagues conducted research on a sam-ple of African-American female offenders who had com-mitted a wide range of offenses; were considered to behigh-rate offenders; were drug users (used illegal drugs atleast three times in the previous 30 days); and were not indrug treatment, jail, prison, or any other institution. Afterinterviewing the women and analyzing the data, theseresearchers concluded as follows:

The results of this study corroborate that women whodisplay low levels of self-control have a higher risk of vio-lent victimization. Even after commonly applied demo-graphic and lifestyle correlates of victimization were con-trolled, low self-control remained significant.90

Self-control theory, although well recognized, remainscontroversial among researchers, with Raymond Pater-noster and Robert Brame noting the disagreement overhow to test the theory but adding that this lack of con-sensus “is solid evidence of the value of [Gottfredson andHirschi’s] theoretical framework for contemporary crim-inological thought.”91

Some scholars have taken the position that for self-control theory to be complete, other dimensions must beconsidered. Richard P. Wiebe, for example, examined therelationship between psychopathy and low self-control.Psychopathy “was developed to account not for criminalbehavior in general, but for antisocial behavior that was‘adequately motivated.’”92 Both constructs, psychopathyand low self-control, reflect “a self-centered, antisocial per-sonality” but had not been studied together. Indeed, Got-tfredson and Hirschi “rejected psychological theories ofcrime, including psychopathy, and many theorists of psy-chopathy have explicitly rejected any application of theirconstruct to predict crime within the general population.”93

Wiebe explained the similarities between psychopathyand low self-control: “Each maps out a path to an antiso-cial life, in which immediate and selfish desires are grati-fied at the expense of both the rights and feelings of otherpeople and one’s own long-term interests.”94 Wiebe devel-oped the construct of self-direction, “comprised of ele-ments of self-control and psychopathy, that appears to besignificantly related to offending.” Emphasizing the needfor this “fresh look” at crime, Weibe concluded, “Pre-venting crime may require attention not only to the causesof antisocial behavior, but to the causes of self-directedbehavior as well.” He raised the possibility that, in hiswords, “low self-control, as currently conceived and mea-sured, is not the last word in predicting delinquency fromintrapersonal variables.”95

In 2004, Hirschi articulated a new definition of self-control, calling it “the tendency to consider the full range of


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 158

potential costs of a particular act [which] moves the focusfrom the long-term implications of the act to its broaderand often contemporaneous implications.” In what theycall the first attempt to test this redefinition, Alex R.Piquero and Jeff A. Bouffard published their research in2007 and concluded that future studies of self-controlshould consider the situational aspects of the concept.According to these investigators, the new definition of selfcontrol means that, at the point of making decisionsregarding behavior, some people will consider potentialcost factors and their prominence, while other people willconsider a fewer number of factors and their prominence.Thus, the situational aspect of self-control is important,with some situations resulting in reactions of high self-control and others in low self-control. This approachmight explain such phenomenon as why, for example, apolice officer might engage in domestic battering or a bankofficial might embezzle. These individuals exhibit signifi-cant self-control in one situation but not in others.96

In concluding their analysis of Gottfredson’s andHirschi’s self-control theory, Akers and Sellers emphasizedthat the theory “has generated enormous interest andattention in criminology and may supplant social bond-ing as the principal control theory. . . . While someresearch reports contradict the theory, and the broadclaims of being the explanation of criminal and deviantbehavior cannot be sustained, on balance the empiricalevidence supports the theory.”97

Although praising Gottfredson and Hirschi’s theory,some scholars have called for alterations of the theory. Somehave found that parenting influences a child’s risk for delin-quency in more ways than social control does. Parentalbehavior may create anger in children, leading to delin-quency.98 In addition, there is some evidence that “low self-control is a condition that can be improved through appro-priate intervention and social experiences.”99

The research linking self-control and crime is extensive,but another aspect of Gottfredson’s and Hirschi’s generaltheory is that self-control occurs through parental social-ization, thus rejecting any influence of other fields, suchas genetics. Recall Chapter 4’s emphasis on integrating thefindings of disciplines other than sociology into explana-tions of criminal behavior. John Paul Wright and KevinM. Beaver take just that approach in testing the role of par-enting in the development of self-control. They examinedtwins and found a weak and inconsistent effect of parent-ing on self-control. They suggested that self-control the-ory should be revised to include the large body of researchthat links behavior to biological factors, such as the effectof smoking and the use of drugs on the fetuses of preg-nant women. They also found that the type of methodol-ogy used to measure social variables (such as self-concept)influences findings. Wright and Beaver concluded that

it is time for criminologists to come to grips with theconfluence of genetic similarities and individually spe-

cific experiences. There is now a voluminous literaturethat links biological features to human growth andsocial development. To the extent that insights fromthis body of knowledge remain excluded, criminologywill remain on the periphery of other, more establisheddisciplines.100

Terri E. Moffitt has also emphasized the importance ofgoing beyond self-control to explain delinquency andcrime. She and others who base their research on neu-ropsychological and developmental psychological theorieshave found support for biological and psychological influ-ences on delinquency that cannot be explained by self-control.101 This research was supported by ElizabethCauffman and her colleagues, who compared public highschool students in California with adolescents who wereincarcerated in that state. These researchers found supportfor Gottfredson and Hirschi’s position that self-control, asexemplified by future orientation, was a significant pre-dictor of offending, but they also found the biological fac-tor of heart rate and the neuropsychological factor of spa-tial span to distinguish the two groups. The “take-homemessage is that proponents of Gottfredson and Hirschi’stheory may wish to revisit the role of neurobiology in thedeterminants of self-control.”102

Finally, in a 2007 publication, Ronald L. Simons andfour collaborators studied several hundred African-American families to examine psychological factors thatmediate the association between parenting practices anddelinquency. They found that steady exposure to threeparental patterns of behavior were predictive of childproblems: high expressions of hostility and rejection, lowlack of monitoring and disciplining the child, and involve-ment in antisocial activity. The first two factors were alsopredictive of a child’s increased involvement and identifi-cation with deviant peers. The researchers found supportnot only for general control theory but also for social learn-ing theory and bonding theory.103

A summary of the control theories discussed here iscontained in Exhibit 6.3. Other control theories exist andare discussed elsewhere. Developmental and life-coursetheories combine social control and social learning theo-ries and are discussed in a later section in this chapter,along with other attempts to integrate criminology theo-ries. Power-control theories were discussed in Chapter 5.

Labeling TheoryA final single theoretical approach to explaining crime isdifferent from the others discussed. Most theories andexplanations of criminal behavior look for the cause of thebehavior, or at least the process by which the behavior cameabout. Why did the individual commit the crime, and whatcan be done to prevent future criminal acts? The answermay be found in the individual physique, body build,chemical imbalance, hormones, or chromosomes; it may


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 159

be found in the environment; or it may be the result ofsome type of social process. The emphasis is on finding outwhy the person engaged in the behavior.

In contrast, labeling theory asks why the person wasdesignated deviant. The critical issue is not the behavioritself but why the behavior is labeled deviant. Not all whoengage in certain kinds of behavior are labeled deviant, butsome are. What is the reason for this distinction? Sociol-ogist Kai T. Erikson described this approach as follows:

Some men who drink heavily are called alcoholics andothers are not, some men who behave oddly are com-mitted to hospitals and others are not . . . and the dif-ference between those who earn a deviant title in soci-ety and those who go their own way in peace is largelydetermined by the way in which the community filtersout and codes the many details of behavior which cometo its attention.104

If criminal behavior is to be explained according to theresponses of others rather than the characteristics of theoffender, the appropriate subject matter is the audience,not the individual actor; for it is the existence of the behav-ior, not why it occurred, that is significant. Only the au-dience’s response determines whether that behavior isdefined as deviant. And, those determinations may resultin serious negative reactions. Most of the popular reformmovements discussed in this text, such as diversion, dein-

stitutionalization, decriminalization, due process for all,and many of the treatment modalities, are based on thebelief that if people are not treated fairly and positivelybut, rather, are labeled negatively after being subject toquestionable procedures, the results will be negative.

For example, some states, such as Florida, have statu-tory provisions for withholding adjudication of a felonyfor persons who have been found guilty and are sen-tenced to probation. During that period, the person doesnot suffer any of the negative effects of being labeled afelon, such as losing voting rights, being prohibited fromrunning for public office, and so on. If the individualdoes not violate probation for a specified period, the indi-vidual does not have a felony record. A study of 95,519men and women in Florida, conducted by Ted Chiricos,Kelle Barrick, and William Bales, examined the subse-quent convictions of these offenders. They found that thefelon label “significantly and substantially” increased thelikelihood of repeat offenses and that the effect wasstronger for whites and females, compared to blacks, His-panics, and males.105

This section considers the history and development ofthe labeling perspective along with recent developmentsand a critique of this approach. In some respects, labelingis a social-structural approach, like conflict theory, but itis primarily a social-process theory because it attempts toexplain labeling as a process by which some people who


6.3

EX

HIB

IT Reiss and Nye’s Control TheoryAlbert J. Reiss Jr. maintained that criminal behavior resultsfrom the failures of two types of controls: personal(internalized) and social (formal controls, such as laws,and informal controls, such as social sanctions). F. IvanNye articulated three social control categories, which hebelieved could occur in formal institutions but took placeprimarily within the family: (1) direct control (parentsthreaten punishment for bad behavior and give rewards forgood behavior); (2) indirect control (youth refuses toengage in delinquent behavior because it might hurt fam-ily or friends); and (3) internal control (youth’s conscienceor guilt prevents participation in delinquent acts).

Reckless’s Containment TheoryAccording to Walter C. Reckless, there are two forms ofcontainment that control behavior. First is outer contain-ment, or external control, such as social pressure. Secondis inner containment, or internal control, which refers tothe ability to direct oneself. This ability is related to one’sself-concept or self-image. These two control componentsare not causes of behavior; rather, they are buffers thatoperate to help the person refrain from engaging in illegal

acts. If the buffers are strong, the person is law-abiding; ifthey are weak, he or she commits a crime. These buffersneutralize society’s norms.

Hirschi’s Bonding TheoryTravis Hirschi looks at the question of why most peopledo not commit crimes. He believes the answer lies in socialbonding, which includes four components: (1) attachmentto conventional persons, (2) commitment to conventionalbehavior, (3) involvement with conventional people, and(4) belief in conventional norms.

Gottfredson and Hirschi’s Self-ControlTheoryLow self-control is the key element of the theory TravisHirschi and Michael Gottfredson developed to explaincrime. Persons with low self-control have several charac-teristics in common. They (1) seek immediate gratification;(2) look for easy or simple ways to gratify their desires; (3)find in crime their need for exciting, risky, or thrilling acts;(4) have little stability in their lives; (5) are self-centeredand indifferent or insensitive to the needs of others; and(6) have little tolerance for frustration. Self-control arisesthrough parental socialization.

Control Theory: A Summary

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 160

commit deviant acts come to be known as deviantswhereas others do not.

The Emergence and Development of Labeling TheoryA 1938 statement by Frank Tannenbaum described thelabeling concept as follows:

The process of making the criminal is a process of tag-ging, defining, identifying, segregating, describing,emphasizing, making conscious and self-conscious; itbecomes a way of stimulating, suggesting, emphasizing,and evoking the very traits that are complained of.

The person becomes the thing he is described asbeing.106

The labeling concept was further developed by EdwinLemert, who distinguished primary and secondary deviance(or deviation) as follows:

Primary deviation is assumed to arise in a wide varietyof social, cultural, and psychological contexts, and atbest has only marginal implication for the psychicstructure of the individual; it does not lead to symbolicreorganization at the level of self-regarding attitudesand social roles.

Secondary deviation is deviant behavior or socialroles based upon it, which becomes a means of defense,attack or adaptation to the overt and covert problemscreated by the societal reaction to primary deviation.107

To labeling theorists, primary deviance is relativelyunimportant; it is secondary deviance that is most impor-tant, for it is the interaction between the person labeleddeviant and the labeler that counts. This approach is calledinteraction theory.

Another contributor to the early development of label-ing theory was Howard S. Becker, who pointed out thatbecause only some of the many people who break rules areconsidered deviant, we must distinguish between rulebreaking and deviance. Rule breaking describes the behav-ior, but deviance describes the reaction of others to thatbehavior; thus, rule breaking is defined as deviant whenengaged in by some, but not all, people. It is important tofind out who is and who is not labeled deviant.108

According to Walter R. Gove, whether or not labelingoccurs depends on (1) the time the act is committed, (2)who commits the act and who is the victim, and (3) soci-ety’s perception of the consequences of the act.109

The Effects of LabelingLabeling theory is based on the assumption that peoplerespond to others in an informal and unorganized way untilthese others are placed in categories that lead to stereotyp-ing, which causes corresponding responses. The responsemay become what sociologists call a self-fulfilling prophecy.

The labeled person develops a self-concept consistent withthe deviant label and acquires the knowledge and skills of thelabeled status. Once labeled, the labelee has an almostimpossible task of shedding that status. The effects of label-ing may snowball, too, in that once a person is stigmatizedby the label, new restrictions are placed on legitimateopportunities and the labelee’s probabilities of furtherdeviance are increased. Such a vicious circle is reinforcedby the tendency of the public to believe that one who com-mits a crime will always be a criminal.

One act may not be enough to result in a negative label,but whether or not a person is labeled brings into playpower differentials. Certain types of groups may be morelikely than others to be labeled deviant: groups that do nothave political power, groups that are seen to threaten thepersons in power, and groups that have low social status.The last is particularly important. Even when middle- andupper-class persons are suspected of committing conven-tional crimes, they are less likely to be labeled criminal thanare lower-class persons.110

The upper classes have greater access to attorneys andare more likely to know their legal rights. They are less likelyto negotiate a plea or admit guilt. They have the symbolsof the middle and upper classes, and those symbols are notassociated with criminal status. For example, they are morelikely to have strong family ties, to have a job, to have suf-ficient income, to speak fluently and knowledgeably, to bepoised, to be able to rationalize their behavior, to have arecord of continuous employment, and to have the respectof the community and of law enforcement officials. Theyare more likely to have friends who may intervene at anystage in the legal process. They live in areas that are unlikelyto be the target of drug raids. They are more likely to getprobation if a probation condition is that they obtain psy-chiatric or other professional services.111

Crime visibility is a factor in determining whether aperson is labeled criminal. People who live in ghettos aremore likely to be visible in committing crimes. Because oftheir greater contact with public services, they are alsomore likely to be visible after crimes are committed. Forexample, most statutory rape cases are brought to theattention of the police through referrals from public wel-fare agencies. Thus, statutory rape is punished primarilyamong the poor, whose victims become visible by apply-ing for maternity aid from welfare authorities.

Empirical ResearchAttempts have been made to test labeling theory empiri-cally. In a classic study, Richard D. Schwartz and JeromeH. Skolnick measured the reaction of 100 employers to apotential employee with a criminal record. The employerswere divided into four groups, and each group was showna different folder on the prospective employee. Stated sim-ply, the conclusion was that employers would not offer ajob to a person with a criminal record. The second phase


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 161

of this study included 58 doctors who had been sued formedical malpractice (a civil wrong rather than a criminalact) in Connecticut. There was no evidence that these doc-tors had lost patients because of the malpractice suits. Mostof the doctors reported no change, and five specialistsreported an increase in the number of their patients.112

Schwartz and Skolnick warned of the problems of com-paring these two phases of their study. The doctors had aprotective institutional environment that did not exist forthe prospective employee. The doctors were permitted tocontinue using hospital facilities and had no difficulty get-ting malpractice insurance, although often at a higher rate.This protective institutional environment may eliminatethe negative labeling process that normally occurs after oneloses a court battle. Another possible reason for the dif-ferent reactions in the two phases in the study was thatphysicians were in short supply and unskilled laborerswere not. But most probably, the difference in reactionwas due to the doctors’ occupational status and the pro-tection they got from their profession. An interesting ques-tion left unanswered by this study is how a doctor wouldbe labeled if he or she had been acquitted of a charge ofassault and battery or some other criminal offense.

One example of the effect of labeling gained nationalattention in the 1970s. Eight sane researchers of variedbackgrounds sought admission to the psychiatric wards of12 hospitals throughout the United States. The hospitalswere of different types, some with excellent treatment facil-ities and others with poor ones. The researchers called thehospitals for appointments, and upon their arrival for theinitial interviews, they feigned mental illness, stating thatthey heard voices. When asked about the significant eventsin their backgrounds, all related the events accurately;none had a history of pathological experiences. Afteradmission to the hospital, all the pseudopatients acted likesane persons. All except one had been labeled schizo-phrenic, and none of the doctors or staff suspected theresearchers’ pseudopatient status. Once labeled insane,they were presumed insane by the staff, who interactedwith them daily. Their behavior did not identify them asinsane; the identity came from a label given to them uponadmission. Thus, they differed from sane persons only inthe label.113

A study by Craig Haney and his collaborators involvedthe simulation of prison life with student volunteersassigned to the roles of correctional officers and inmates.The experiment had to be terminated because of seriousproblems. Those in the experiment who were assigned therole of correctional officers began behaving as if theyderived pleasure “from insulting, threatening, humiliatingand dehumanizing” those who were inmates. The oneswho were assigned to be inmates became depressed,despondent, and helpless and acted in self-deprecatingways.114

Much of the more recent research on labeling theoryfocused on juvenile delinquency. This research calls for

refining the approach. For example, gender roles shouldbe considered. Male and female juveniles do not neces-sarily respond to labels in the same way.115 Social class andrace should be taken into account, along with whether theperson is a first-time or a persistent offender.116 The dif-ferences in reactions to negative and positive labelingshould be measured.117

Recent Refinements of LabelingTheoryIn recent years, scholars have developed refinements of tra-ditional labeling theory. This section looks at three of those.

Braithwaite’s Restorative JusticeJohn Braithwaite emphasizes that labeling per se is not theissue. Labeling must be examined in the context of the totalsocial structure. Labeling might cause some to abandoncriminal activity; it might cause others to continue. Whatare the characteristics of the broader social context in whichthe behavior occurs that might explain whether the personbecomes law-abiding or law-breaking? Braithwaite answeredthe question with his concept of shaming, defined as socialdisapproval that has the “intention or effect of invokingremorse in the person being shamed and/or condemnationby others who become aware of the shaming.”118

In the view of Braithwaite, shaming is important forsocial control, but the shaming should be of the criminalact, not the criminal. Shaming that goes far enough to stig-matize the person labeled is a process Braithwaite refers toas stigmatization or disintegrative shaming. This processcan propel the individual into delinquent or criminalbehavior. Disintegrative shaming does not involve anyattempt to reintegrate the shamed person into society; itstands in contrast to reintegrative shaming, in whichattempts are made at reintegration of the offender into thecommunity “through words or gestures of forgiveness orceremonies to decertify the offender as deviant.”119 Thesereactions are more likely to reduce the negative effect ofshaming, thus reducing the chances that the shamed per-son will continue in crime. According to Braithwaite, soci-eties that demonstrate forgiveness and show respect butstill take crime seriously have lower crime rates than thosethat humiliate and degrade criminals. Likewise, familiesthat show love and respect for their children who misbe-have are more likely to have law-abiding children than arefamilies that degrade and demean their children.120

Braithwaite believes a theory of crime should explainthat most offenders are young, minority, male, and unem-ployed. They live in large cities and have low educationalachievements and goals. According to Braithwaite, theseare the persons most likely to be targeted for disintegra-tive shaming, and, because they do not have strong socialbonds to society, they are also stigmatized, which pushesthem toward criminal subcultures. In that environment,


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 162

they have an opportunity to gain self-respect by rejectingthose who reject them; thus, they are more likely to engagein deviant behavior. “In contrast, the consequence of rein-tegrative shaming is that criminal subcultures appear lessattractive to the offender.”121 In short, reintegrative sham-ing will reduce and disintegrative shaming will increase thechances of future offending. Akers and Sellers maintain,however, that the limited empirical studies do not providesupport for this position.122 Braithwaite acknowledges thatnot as much empirical testing as might have been expectedhas occurred, but although he recognizes the researchresults are mixed, and he and his colleagues are workingon a revision of his approach, he notes the support he andone of his colleagues found in the first test of his theory.In his words:

Australian nursing home inspectors with a reintegra-tive shaming philosophy were successful in substan-tially improving compliance with regulatory laws in thetwo years after inspections, but compliance substan-tially worsened when inspectors adopted a stigmatizingphilosophy.123

One might, however, question whether that fact patterninvolving administrative law applies to juveniles or adultsviolating the criminal law.

Braithwaite’s theory of reintegrative shaming has pro-vided the basis for an important policy approach to crimeprevention, and that is restorative justice, a concept thatbegan in New Zealand but has spread throughout theUnited States. The restorative justice approach providesboth adults and juveniles a wide range of programs thatare designed to restore crime victims and the community.Examples are community work service, apologies to crimevictims, restitution, and other programs to aid crime vic-

tims.124 Exhibit 6.4 reproduces the Vermont statute onrestorative justice. Note the leeway given to local jurisdic-tions to develop programs that fit within the spirit of thestatute.

The obvious goal of restorative justice is to reducedelinquency and crime. Francis T. Cullen and his collab-orators have advocated that restorative justice be linkedwith rehabilitation (discussed in Chapter 3) in an effort atcrime reduction.125 But Akers and Sellers concluded thatthere has been insufficient research on which to makeinformed conclusions concerning the impact of restora-tive justice and other so-called faith-based programs onrecidivism rates.126

Finally, restorative justice has become importantenough in the United States to generate special collegecourses and texts. Clifford K. Dorne’s 2008 publication ofRestorative Justice in the United States is an example. Thistext contains an overview of the approach, along withresearch in the area.127

Matsueda’s Reflective AppraisalAnother contributor to labeling theory in recent years isRoss L. Matsueda.128 Matsueda based his approach on themuch earlier theory of George Herbert Mead, whose clas-sic Mind, Self, and Society was published in 1934.129 Meadexplored the concept of self, which he said is composed ofmany dimensions and is developed through symbolic inter-actions with other individuals, some of whom are more sig-nificant than others. In the process of interactions, indi-viduals often take the role of the other person and try todiscern how that person is evaluating his or her behavior.Later in their development, they learn to generalize the rolesof others, such as a group, a community, or a society. The


Some jurisdictions permit the transfer to adult criminalcourts of juveniles accused of committing violent acts;others retain juvenile court jurisdiction. Sean Patrick, right,shown here with his attorney, Carl Hueber, left, wasprocessed through the juvenile court in Spokane County(Washington) after he was accused of taking a gun toschool and pointing it at police. Patrick, who allegedlywanted police to kill him, was severally wounded whenpolice shot him in the face, arm, and stomach, but hisinjuries were not life-threatening. Patrick had reportedlytaken 30 pain killers that day. According to his attorney, he had been diagnosed as paranoid schizophrenic and had been hearing voices, some good, some bad, for twoyears prior to the crime. At his sentencing, the defenserecommended, and the prosecutors agreed, to leniency, andPatrick was placed on probation with electronic monitoringand ordered to talk to other teens about depression. Hewas also ordered to write letters of apology to the teacherand three classmates at whom he waived the gun. Just sixdays after his sentencing, Patrick allegedly attemptedsuicide by taking a large dose of pills.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 163

behavior children exhibit may initially be labeled as play ormischief by family members (and his theory is read by someto suggest that such informal labeling has a greater impactthan formal labeling does), but a larger group or the soci-ety may label that same behavior as bad, evil, deviant, oreven delinquent. According to Matsueda:

Eventually, this spiraling labeling process can leave theyouth in the hands of juvenile justice officials—cut offfrom conventional society, stigmatized by parents andteachers, and left with a delinquent self-image. Thus, aself-fulfilling prophecy is set up: through this processof deviance amplification . . . an otherwise conform-ing child may eventually respond to the initial labelingof harmless acts by confirming the delinquent label.130

Matsueda, along with Dawn J. Bartusch, studied data onjuvenile delinquency and found support for their conceptof reflected appraisal, which refers to a person’s percep-tion of how others view him or her. Matsueda and Bar-tusch advocate that those who believe that others viewthem as behaving in a delinquent manner in certain sit-uations are more likely to engage in such behavior than

are those who view others as expecting them to be law-abiding citizens.131

Sampson and Laub’s CumulativeDisadvantagesLater in this chapter, we discuss developmental and life-course theories, but at this point it is important to note thatthe life-course approach embraces labeling theory. Accord-ing to Robert J. Sampson and John H. Laub, labeling the-ory is “truly developmental in nature, because of its explicitemphasis on processes over time.”132 Labeling theory leadsto structural disadvantages, or what Sampson and Laub callcumulative disadvantages, in future life opportunities.Consequently, those who are labeled delinquent or crimi-nal find that the event is a transitional one in their lives andthat fewer legitimate opportunities will be available to themin the future because of the structural disadvantages. In thewords of Sampson and Laub:

Cumulative disadvantage is generated most explicitlyby the negative structural consequences of criminaloffending and official sanctions for life changes. The


6.4

EX

HIB

IT Vermont Statutes Annotated, Title 28, Section 2a (2007)

■ Restorative Justice(a) State policy.—It is the policy of this state that prin-ciples of restorative justice be included in shaping howthe criminal justice system responds to persons chargedwith or convicted of criminal offenses. The policy goalis a community response to a person’s wrongdoing atits earliest onset, and a type and intensity of sanctiontailored to each instance of wrongdoing. Policy objec-tives are to:

(1) Resolve conflicts and disputes by means of anonadversarial community process.

(2) Repair damage caused by criminal acts to com-munities in which they occur, and to addresswrongs inflicted on individual victims.

(3) Reduce the risk of an offender committing amore serious crime in the future, that wouldrequire a more intensive and more costly sanc-tion, such as incarceration.

(b) Implementation.—It is the intent of the generalassembly that law enforcement officials develop andemploy restorative justice approaches whenever feasi-ble and responsive to specific criminal acts, pursuant tosections [citations omitted], concerning sentencing, andthe provisions of this title, concerning persons in thecustody of the commissioner of corrections. It is thefurther intent of the general assembly that such restora-

tive justice programs be designed to encourage partici-pation by local community members, including victims,when they so choose, as well as public officials, in hold-ing offenders accountable for damage caused to com-munities and victims, and in restoring offenders to thelaw-abiding community, through activities:

(1) Which require offenders to:(A) acknowledge wrongdoing and apologize to

victims;(B) make restitution for damage to the victims,

consistent with provisions of [citationsomitted];

(C) make reparation for damage to the com-munity by fulfilling a community service;and

(D) when relevant, successfully complete treat-ment addressing the offense or otherunderlying problematic behavior, or under-take academic or vocational training orother self-improving activity.

(2) Which aid in the recovery of victims, recog-nizing that victims, particularly of violent crime,often suffer lifelong effects and, accordingly,must feel safe and involved in any programoffered to assist them.

(3) Which help in identifying the causes of crimeand ways community members and municipaland state government can reduce or preventcrime in the future.

Restorative Justice: The Vermont Approach

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 164

theory specifically suggests a “snowball” effect—thatadolescent delinquency and its negative consequences(e.g., arrest, official labeling, incarceration) increasingly“mortgage” one’s future, especially later life chancesmolded by schooling and employment. . . . The the-oretical perspective in turn points to a possible indirecteffect of delinquency and official sanctioning in gener-ating future crime.133

In an effort to assess the importance of labeling in thedevelopment of cumulative disadvantages, Jon GunnarBernburg and Marvin D. Krohn analyzed data rangingfrom adolescence to early adulthood. These researcherslooked at the relationship that official labeling might haveon future criminal activity by producing negative conse-quences that reduced conventional opportunities, specifi-cally education and job opportunities. They found thateducational attainment was affected by early labelingbecause the labeling decreased the chances the youthswould finish high school. Lack of a high school educationin turn affected future job opportunities. Bernburg andKrohn concluded as follows:

Our results are consistent with the hypothesis that offi-cial intervention during adolescence influences crimi-nal involvement as late as early adulthood when indi-viduals become fully affected by blocked life chancesshaped by education and employment. But, in additionto the indirect effects that both intervention by thepolice and the juvenile justice system have on adultcrime, official intervention also directly influences sub-sequent criminality.134

Bernburg and Krohn acknowledged that they did not mea-sure possible intervening factors that could affect the out-come and recommended that future research do so.

Evaluation of Labeling TheoryIn 1988 Daniel L. Dotter and Julian B. Robuck describedlabeling theory as “one of the most influential in the fieldof deviance over the past two decades.”135

Akers and Sellers, writing in 2004, looked at the reac-tion to labeling theory over the years. In the 1960s, thetheory “captured the imagination of social scienceresearchers, theorists, and practitioners,” but by the 1970s,the theory was on the decline in popularity, with some ofthose who had originally supported the theory moving toother perspectives. Today the theory continues to holdsome advocates, especially those, as we noted earlier in thischapter, who combine it with other theories. Akers andSellers concluded that the primary importance of labelingtheory is that it “calls attention to the unintended conse-quences of social control.” But the main weakness of label-ing theory is that “it essentially ignores primary devianceand seriously underestimates the influence that other vari-ables have on behavior in the first place and continue tohave on its future occurrence.”136

One of the most serious criticisms of labeling theory isthat it is not a theory—it is a perspective. No systematiclabeling theory has been developed. The empirical assess-ment of a theory requires that it produce testable propo-sitions. Not only is that difficult with labeling theory, butCharles R. Tittle claimed that “empirical tests of labelingtheory are both impossible and ridiculous.”137

Tittle emphasized that a systematic theory cannot becreated unless it has precisely defined terms that can bemeasured. Labeling theory’s methodological problems arecrippling, and the data are poor. Thus, the research can-not be used to support the theory, but neither can it beused to reject it. For adequate testing, it is imperative thatspecific hypotheses be derived and tested after precise,operational definitions have been articulated.138

One of the common criticisms of labeling theory is thatit avoids the question of causation and ignores the actualbehavior in question. Labeling theory assumes that whatone does is not the key to explaining behavior; the key iswho that person is and why he or she was labeled deviant.Not all would agree with that position, as the discussionof conflict and critical theory (see again Chapter 5) andevidence of discrimination against racial and ethnic groupsor decisions based solely on gender might suggest.

Another criticism of labeling theory is that it views theactor, or labelee, as too passive and that it does notacknowledge the reciprocal relationship between the actorand the reactor. Most labeling theorists overemphasize theaction of society and deemphasize the action of the sub-ject being labeled. Social interaction should receive greaterattention.

Labeling theory may also be criticized for its lack ofattention to the personality characteristics of those whoengage in deviant behavior. To the labeling theorists, char-acteristics of the individual, such as personality traits, arenot important in explaining behavior. It is the reaction tothe person that is critical. Yet labeling theory does notexplain differential law enforcement. Why do police arrestsome and not others for the same offenses?

A final category of criticisms focuses on the assump-tion that labeling produces only negative results. Criticsargue that labeling a person deviant might deter that per-son rather than plunge him or her into further deviance.Whether or not the labeling process produces negative orpositive results may depend on a number of factors thathave been overlooked by labeling theorists.

First, it appears that labeling has different effects on thedeviant at various stages in his or her career. Labelingmight thrust a male juvenile delinquent into a criminalcareer but deter an adult female shoplifter. Perhaps the keyelement is peer support, not labeling. Labeling may createa subculture, thereby establishing peer support, especiallyamong juveniles. In short, it matters who is labeled, whodoes the labeling, and how the label is applied.

A second factor concerns the confidentiality of thelabeling. If the label is confidential and given to a non-


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 165

professional deviant, that person may be more likely toabandon his or her deviant behavior than if the individ-ual is already a professional and the label is public. Third,the result is more often positive than negative when thesubject has some commitment to and is sensitive towardthe person doing the labeling. For example, former alco-holics and drug addicts may be more successful than non-peers, counselors, or psychiatrists in the rehabilitation offellow deviants. Fourth, a person is more likely to aban-don deviant behavior if the label of deviant, once given,can be removed easily.

Fifth, the reaction of friends and society is importantin whether the label results in positive or negative behav-ior. If friends and others are supportive in assisting theindividual to improve, the results are more likely to be pos-itive. Sixth, most labeling theorists have overlooked thepossibility that positive labeling can increase positive be-havior.

Because of these and other criticisms, some scholarshave argued that labeling theory is not very important.Others take the position that even if labeling does not causedelinquent and criminal behavior, it does cause problemsthat should be analyzed. For example, there is evidencethat people who have been labeled mentally ill have neg-ative feelings about both that label and the way others

respond to them even after they are released from treat-ment. The more negative these feelings, the more difficultit may be for such persons to interact positively with oth-ers. They may try to hide the fact that they have beentreated, they may avoid interaction with others, or theymay project negative attitudes that do not exist in others.Thus, labeling theory should be modified and furtherdeveloped rather than abolished.139

The major contributions to labeling theory are sum-marized in Exhibit 6.5.

Integrated TheoriesThis chapter, along with Chapters 3, 4, and 5, covers the-ories that attempt to explain delinquent and criminalbehavior from various points of view. But none of thesetheories explains all crime, as we would like for a generaltheory to do, and some of the theories compete with eachother. How do we evaluate them, or make sense out ofthem?

In their concise but informative theory book, J. MitchellMiller, Christopher J. Schreck, and Richard Tewksburydiscuss the issue and make several suggestions, some ofwhich are utilized by other scholars as well. The simplest


6.5

EX

HIB

IT Frank TannenbaumAccording to Frank Tannenbaum, persons are tagged,labeled, or named—and whether the purpose of this pro-cess is punishment or reformation, disapproval is indicated.

Edwin LemertEdwin Lemert distinguished between (1) primary devia-tion, which arises in a wide variety of contexts and hasonly marginal implication for the individual’s psychic struc-ture; and (2) secondary deviation, which consists ofdeviant behavior that is a defense to society’s reaction toone’s primary deviation.

Howard S. BeckerHoward S. Becker distinguished between rule breaking anddeviance, as some people who break rules are not consid-ered deviant. Deviance is a label created by society andattached to some but not all people who engage in thedescribed acts.

Braithwaite’s Restorative JusticeJohn Braithwaite examined labeling theory in the contextof the total social structure, which might cause some toembrace criminal behavior and others to avoid it. His keyword is shaming, which is a process of heaping social dis-approval on a person until it induces remorse. Shaming is

an important element of social control, but there are twokinds. Disintegrative shaming stigmatizes the subject,while reintegrative shaming consists of attempts to rein-tegrate that person into law-abiding society, which is donethrough a process of restorative justice.

Matsueda’s Reflective AppraisalBased on the symbolic interaction theory of George Her-bert Mead, Ross L. Matsueda’s approach focuses on thedevelopment of the individual’s self-concept. This processoccurs through social interaction during which the personplays the role of the other in an effort to determine his orher reactions. Some of those reactions are negative, whichmay lead the role player to a self-fulling prophecy: “Theysay I am bad; so I will be bad.” Thus, the initially harm-less acts of a child, as defined by his or her parents, maybe labeled bad by others. The spiraling labeling processmay cause an otherwise conforming child to live up to thelabels.

Sampson and Laub’s CumulativeDisadvantageThe accumulation of structural disadvantages, over time,leads to cumulative disadvantages in future life opportuni-ties, which may lead to delinquency and crime.

Labeling Theories: A Summary

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 166

approach is singular theory assessment, which involvesevaluating one individual theory and the tests of thattheory’s propositions. This approach is used often, andmost theories will generate some support despite the factthat theories are often difficult to test empirically. Two ormore theories may be compared on the basis of the scopeof their coverage of crime and the empirical support forthem, a process referred to as theory competition. But,they remain as individual theories.140

In an effort to develop a theory that can provide us witha broader explanation of crime and, hopefully, reach a gen-eral theory of criminal behavior, social scientists in recentyears have focused on theoretical integration, the combi-nation of two or more theories to form a new one. Theassumption is that the resulting theory will be superior toany of the individual theories standing alone. This is notan easy task. For example, some of the theories have con-tradictory underlying premises—behavior is predeter-mined, behavior is learned. How do you combine thosepropositions?

Theoretical integration can be accomplished in twoways. In the first, scientists examine the distinct but com-plimentary aspects of various theories in an effort to reacha broader explanation, a process referred to as sequentialintegration. The second approach, conceptual fusion,involves linking together propositions of various theoriesthat are somewhat different but similar enough to be com-bined, or fused.141

Despite the difficulties of theoretical integration, nu-merous scholars have attempted it (although most of theseefforts are limited to integrating only a few theories). Thefollowing section focuses on some of these integratedtheories.

Akers: Conceptual AbsorptionIn his earlier publications, Ronald L. Akers wrote about theoverlap of several theories, including social learning, socialbonding, labeling, conflict, anomie, and deterrence.142 Insubsequent work, Akers proposed theoretical integrationby means of conceptual absorption (similar to conceptualfusion, just discussed), which he defined as “subsumingconcepts from one theory as special cases of the phenom-ena defined by the concepts of another.”143 To illustrate,Akers suggests that the commitment element of socialbonding can be absorbed within his social learning theory’sdifferential reinforcement, as follows:

Differential reinforcement refers to the reward/punish-ment balance for both conforming and deviant activity.Commitment identifies only one specific item (loss ofinvestment) in a larger classification of costs. Further-more, that item enters into differential reinforcement ononly one side of the overall reinforcement balance. Thereis nothing in commitment that is not already wholly con-tained in differential reinforcement.144

Developmental and Life-CourseTheoriesOne of the most frequently analyzed theories in criminol-ogy today is the developmental and life-course (DLC) the-ory approach. In a 2002 address to the American Society ofCriminology, David P. Farrington examined the key theo-retical and empirical issues of the theory, which, he stated,attempts to explain three main issues: “the development ofoffending and antisocial behavior, risk factors at differentages, and the effects of life events on the course of develop-ment.” Farrington explored the 10 conclusions about delin-quency and criminal behavior that are widely accepted andthat must be explained by DLC (or any other) theories.

■ “First, the prevalence of offending peaks in the lateteenage years—between ages 15 and 19.

■ Second, the peak age of onset of offending isbetween 8 and 14, and the peak age of desistance[stopping, ceasing the behavior] from offending isbetween 20 and 29.

■ Third, an early age of onset predicts a relativelylong criminal career duration and the commissionof relatively many offenses.

■ Fourth, there is marked continuity in offending andantisocial behavior from childhood to the teenageyears and to adulthood. . . .

■ Fifth, a small fraction of the population . . . com-mit a large fraction of all crimes. . . .

■ Sixth, offending is versatile rather thanspecialized. . . .

■ Seventh, the types of acts defined as offenses areelements of a larger syndrome of antisocialbehavior. . . .

■ Eighth, most offenses up to the late teenage yearsare committed with others, whereas most offensesfrom age 20 onwards are committed alone. . . .

■ Ninth, the reasons given for offending up to the lateteenage years are quite variable, including utilitarianones . . . for excitement or enjoyment . . ., orbecause people get angry. In contrast, from age 20onwards, utilitarian motives become increasinglydominant.

■ Tenth, different types of offenses tend to be firstcommitted at distinctively different ages.”145

In general, DLC theories may be distinguished from theposition taken by Hirschi and Gottfredson (in their gen-eral theory, discussed earlier in this chapter) on the rela-tionship between age and crime. The DLC theorists dis-agree with the position of Hirschi and Gottfredson thatonce the child has developed low self-control or a tendencyto misbehave, those characteristics remain essentiallythroughout life. Since it is the lack of self-control that, intheir view, causes delinquency and crime, the causes of


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 167

such deviancy are the same at all ages. In contrast, DLCtheories take the position that age-related variables, not ageper se, explain the changes in delinquent and criminalbehavior. And DLC theories deemphasize the importanceof biology and early childhood experiences.146

Consider, for example, this statement by Michael L.Benson:

Theoretically the assumption that adult behavior canbe predicted in early childhood flows against the cen-tral premises of the life course approach. The life courseapproach is based on the premises that development isan ongoing process that unfolds over the entire lifespan, and development involves interactions betweenthe individual and the environment. . . . Predictingthe life course of an individual based only on factorspresent at an early age ignores all of the causal factorsthat come into play later.147

Numerous researchers have worked on DLC theories.The works of several are summarized here.

Sampson and Laub: Age-Graded TheorySeveral references have already been made to Robert J.Sampson and John H. Laub. Their theory is called age-graded theory, referring to age-graded, informal socialcontrols such as the bonding that occurs within the familyand schools, with peers, and in later life, with marriage andjobs. Sampson and Laub attempt to explain why people donot engage in delinquent and criminal offenses (assumingthat offending is natural). They believe that people areinhibited from offending by their social bonds to the age-graded institutions mentioned here, and they make a deci-sion to offend on a cost-benefit basis, which is related totheir fund of social capital, defined as life-sustaining, pos-itive relations with other people and institutions, whichincrease their bonding with society.

In his 2003 presidential address before the annual meet-ing of the American Society of Criminology, Laub focusedon the life course of the discipline of criminology but madethe analogy to the 17 years he and Sampson had spentdeveloping their theory of the life course of criminalbehavior. By life course, Sampson and Laub mean thepaths through which a person travels during a life span. Itexamines change and continuity, internal and externalforces, age, and other important variables.148

Some terms critical to life-course theory need to bedefined. Trajectories are long-term patterns of behavioror lines of development—such as marriage, career, or par-enthood—that consist of transitions, which are short-term events, such as a first job or even a first marriage. Athird concept that is important is that of turning points,which refers to a “change in the life course” generated bythe “interlocking nature of trajectories and transitions.”Thus, the life-course theory examines the long-term viewof a person’s life, implying a strong connection betweenchildhood and adulthood. But by also looking at behavior

in terms of the shorter view, the theory acknowledges thatlife trajectories can be modified or redirected by transi-tions or turning points. According to Sampson and Laub,“Social institutions and triggering life events that maymodify trajectories include school, work, the military,marriage, and parenthood.”149 These events give a personan opportunity to change the course of his or her life, ineffect, to start over or to make new friends and associatesand, at least to some extent, to discard the past life.

By gaining access to the data cards from the research ofSheldon Glueck and Eleanor Glueck (see discussion inChapter 4), who sampled 500 delinquents and 500 non-delinquents who were followed for almost two decades,Sampson and Laub were able to analyze criminal and non-criminal behavior over much of the critical life span oftheir sample of boys/men.150

Sampson and Laub emphasize a fact we have knownfor a long time: Crime by teenagers is disproportionate tothat of other age groups. Most researchers have concen-trated on this age group in their formulations of theoriesabout crime. Sampson and Laub believe that it is a mis-take to ignore the implications that childhood behaviorhas for the total life span of criminal activity. Further, theycriticize most researchers for largely ignoring “the linkbetween social structural context and the mediating pro-cesses of informal social control.”151 Thus, it is not onlythe social structures within which one is raised that areimportant but also the social processes that occur withinthose structures that must be considered.

Like Gottfredson and Hirschi in their general theory, dis-cussed earlier in this chapter, Sampson and Laub acknowl-edge that early childhood experiences are important in thedevelopment of self-control. Similar to the social bondingtheory of Travis Hirschi, also discussed earlier in this chap-ter, Sampson and Laub’s theory maintains that when a per-son’s social bonds to society become weak or nonexistent,delinquency and crime will result. But unlike Gottfredsonand Hirschi and others, Sampson and Laub reject the impli-cation that adulthood experiences are of little relevance. “Inother words, we contend that social interaction with bothjuvenile and adult institutions of informal social control hasimportant effects on crime and deviance.”152

Sampson and Laub developed three propositions for theirapproach. In their words, those propositions are as follows:

1. “structural context mediated by informal family andschool social controls explains delinquency in child-hood and adolescence;2. in turn, there is continuity in antisocial behaviorsfrom childhood through adulthood in a variety of lifedomains; and3. informal social bonds in adulthood to family andemployment explain changes in criminality over the lifespan despite early childhood propensities.”153

Sampson and Laub are thus invoking both social con-trol and learning theory, as they emphasize the learning


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 168

processes that occur in the family. They contend that bothabrupt turning points and gradual changes that occur asone matures, such as getting married or entering a career,will increase that individual’s bonding to society. Thus, itis understandable that even those who are delinquent dur-ing youth will decrease their illegal activity as they growolder. But those who have negative experiences, or do notgo through such changes as marriage and a career, maypersist in illegal activities. Those who are strongly bondedwith their spouses or attached to their jobs will be less likelyto risk losing these and disrupting their lives by engagingin illegal behavior. Mark Warr referred to this approachas a “straightforward summary of control theory—strongties to conventional institutions or persons create stakesin conformity and thereby inhibit deviance.”154

But Warr noted limitations of the approach of Samp-son and Laub, arguing that it may not be marriage per sethat inhibits crime but the fact that marriage leads spousesto change their social contacts significantly, and thosechanges are associated with desistance from crime. Thus,it is the peer influences that are at work. In his own work,Warr found strong support for this position.155

The issue of peer influences within marriage was alsoinvestigated by Ronald L. Simons and his collaborators,who thought that Warr’s additions to Sampson and Laub’stheory were advised but suggested that Warr needed toembellish his own approach. “He [Warr] emphasizes thefact that adult friends can either encourage or discouragecriminal behavior, but fails to recognize that the same istrue for romantic partners.” Further, the assumption thatromantic partners encourage men to desist from crimeshould be refined. According to DLC theory, this wouldbe true only if the partner were conventional, not also anti-social. And research shows that people do tend to fall inlove and marry people who are similar to themselves. So,if an antisocial, criminal male marries the same type ofwoman, the marriage might not result in influencing himto change his social network and desist from crime. Inshort, it is important to look at the nature and quality ofthe relationship with romantic partners to assess theimpact they will have on future criminal behavior. Simonsand his colleagues found that for the women in their sam-ple, “a conventional romantic partner, strong job attach-ment, and conventional adult friends all served to moder-ate the chances that a woman with a delinquent historywould graduate to adult crime.” For men, however, theonly one of these factors that was influential was that ofconventional adult friends.156

Sampson and Laub have continued their examinationof the life course, however, and in a 2003 publication, theyreported the results of what they believe is the “longestlongitudinal study of crime to date.” The project exam-ined “trajectories of offending over the life course of delin-quent boys followed from ages 7 to 70.” The researcherslooked at the childhood characteristics, family back-ground, and individual differences of the sample and

reported that these factors alone were not predictive of thefuture offending of persistent criminals.157

In a 2006 publication, Sampson and Laub, along withChristopher Wimer, reported research findings supportingtheir position that marriage is one of life’s transitions thatis associated with desisting from criminal behavior, reduc-ing it by approximately 35 percent.158 Other researchersreport that the effect of marriage on desisting from crim-inal behavior exists but is more pronounced for men thanfor women.159

Another of life’s activities that Sampson and Laubfound to be associated with reductions in criminal activ-ity was stable employment, which, they argued, buildssocial capital and thus improves bonding with society.John Paul Wright and Francis T. Cullen studied the rela-tionship of associating with prosocial coworkers to thechanging of peer networks, criminal behavior, and druguse. They found some support for the reduction of mis-behavior by stable employment. They emphasized that itis not only the quality of the employment that is important(as recognized also by Sampson and Laub) but the qual-ity of the peer associations as well. “In some work contexts,employment may be positively associated with criminalconduct because it builds criminal capital, while in otherwork contexts, it may be associated with declines in ille-gal behavior that are due to the attainment of prosocialcapital.”160

The influence of the work of Sampson and Laub onother criminologists has been enormous. Wright andCullen described the classic work by Sampson and Laub,Crime in the Making, as

perhaps the most influential life-course study of crim-inal behavior. . . . [It is] an ambitious effort . . . toput forth a theory of crime that attempts to explainmisbehavior across the life course. . . . [The book is]now required reading for serious criminologists.161

Moffitt: Developmental TheoryIn her psychological development theory, Terri Moffittconsiders the fact that not all teen offenders continue tocommit criminal acts as adults but that most chronic adultoffenders committed illegal acts as teens. Moffitt examinestwo categories of antisocial adolescents. The adolescentlimiteds are those whose antisocial behavior peaks duringadolescence and then declines. They constitute the major-ity of antisocial adolescents. They are strongly influencedby their peers. In contrast are those few antisocial adoles-cents who had social problems as children, began delin-quent behavior early, and continued it. These juveniles arethe life-course persisters, those young people who begincrime early in life, persist, and commit even more seriouscrimes as they mature. They are thought to have a neu-ropsychological deficit; they differ qualitatively from oth-ers. They have low self-control, exhibited difficult tem-peraments during childhood, may be impulsive, and may


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 169

demonstrate hyperactivity. They may be expected to con-tinue offending into adulthood, and they are not stronglyinfluenced by their peers.162

In their analysis of Moffitt’s approach, Miller, Schreck,and Tewksbury concluded that the empirical tests are gen-erally supportive of her integrative theory. They refer toher developmental theory as “one of the most influentialpsychological theories in contemporary criminologicaltheory.”163

Cullen and Colvin: Social Support or CoercionFrancis T. Cullen suggested that all criminological theorycan be integrated around the concept of social support,which refers to social integration within group relation-ships in which individuals provide social, emotional, andmaterial support for each other. The higher the level ofsocial support, the less likely crimes will result. Cullenadvances 13 propositions concerning social support, con-cluding with the proposition that in many cases, a precon-dition for social control is the existence of social support.164

Cullen and Robert Agnew maintain that this approachexplains much crime and that it adds a dimension ofhumanitarianism to efforts to control crime.165

In reacting to Cullen’s approach, Mark Colvin arguedthat it is coercion (defined as “a force that compels orintimidates an individual to act because of the fear or anx-iety it creates” and that “can also involve the actual orthreatened removal of social supports”166) that unifiestheories that explain crime.167 Coercion may come frominterpersonal sources, such as individuals and smallgroups, or from impersonal sources, such as governments,the economy, or criminal justice agencies. Together Cullenand Colvin propose that crime is caused by coercion andprevented by social support. After a discussion of the pol-icy implications of their proposal, Cullen and Colvin,along with Thomas Vander Ven, concluded as follows:

To live up to our value of freedom while reducingcrime, it is imperative that we offer social support toindividuals and families that free them from coerciveforces. Social support offers an avenue to freedom thatfosters responsibility, mutual help, and trust. It is in thiscontext of social support that freedom can coexist withlow levels of crime.168

Thornberry: Interactional TheorySocial learning, social bonding, and social-structural theo-ries are combined by Terence P. Thornberry into what hecalled interactional theory. According to Thornberry, theinteraction of adolescents with other people and institu-tions results in behavioral outcomes. Adolescents are notpropelled toward certain forms of behavior, but behavioris learned through the process of interaction within the fam-

ily, within the school, within other social institutions, andwith peers. The chances of delinquency are reduced by astrong attachment to parents, commitment to school, andacceptance of society’s conventional values and increasedwhen these conditions do not exist. But it is the interactionover time, not just the child’s perception and level of attach-ment and commitment, that is the key to explaining behav-ior. Thornberry emphasizes that these interactions changeover time; they are not static. Thus, he looks at three stages:early adolescence (ages 11 to 13), middle adolescence (ages15 to 16), and late adolescence (ages 18 to 20). Thornberryalso advocates that although weak social controls may bethe fundamental cause of delinquency, they do not leaddirectly to delinquency (as classic control theory suggests).Rather, the weakening of controls permits a wider array ofbehavior, including conventional as well as delinquent andcriminal behavior. Also, the interactions are reciprocal; sojust as delinquent peer groups may increase the chances ofan adolescent becoming delinquent, delinquent behaviormay increase the chances of the person’s movement towarda delinquent peer group. Finally, Thornberry also looks atthe structural variables of class, race, sex, and communityof residence, which, he claims, set the stage for the socialinteractions discussed here.169

In later works Thornberry, along with Marvin D. Krohn,developed the interactional theory perspective, suggestingthat social bonding and offending affect each other and thatboth change over time. Life events, such as joining a gang,are influential in this reciprocal interaction. Those juvenileswho begin their delinquent activities later in their develop-ment differ from the early starters in degree but not in kind;but in both categories, the key is interaction.170

The importance of the life event of gang membership onthe development and duration of delinquency was investi-gated by numerous researchers, who built on the works ofThornberry and his collaborators as well as other researcherswho found higher delinquency among gang members thanamong nongang members. The important question iswhether delinquency increases as the result of gang mem-bership. Most recently, researchers, such as Rachel A. Gor-don and her collaborators, have reported that “boys whojoin gangs are more delinquent before entering the gangthan those who do not join” but that “drug selling, druguse, violent behaviors and vandalism of property increasesignificantly when a youth joins a gang. The delinquency ofpeers appears to be one mechanism of socialization.” Butthe research also shows that when boys leave the gangs, theyless frequently engage in delinquent behavior; thus, theinfluence of the gang is temporary.171

In their review of interactional theory, Cullen andAgnew acknowledge that the research is limited and theconclusions tentative, but they do believe that the researchsupports the importance of both developmental issues andreciprocal effects in explaining behavior. Cullen andAgnew also note that recently, Thornberry and Krohn haveproposed a revision of their theory to include individual


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 170

traits and biological variables as well as the considerationof other life stages.172

Krohn: Network AnalysisThe social learning theory of differential association, alongwith social bonding theories, form the basis of a theory ofnetwork analysis proposed by Marvin D. Krohn. Networksare social or personal. Personal network refers to a person’sties to others, including persons within the family, friends,church, and school. Social network includes sets of indi-viduals or groups who are linked by friendship or otherrelationships. Individual behavior is constrained by socialnetworks; that constraint could be toward conforming ortoward delinquent behavior, with the direction dependingon the number and the nature of those networks. For exam-ple, a youngster who lives in a small village where every-one knows everyone else will be more constrained than onewho lives in a more anonymous setting.173

After looking at the nature of friendship networksamong adolescents, Dana L. Haynie concluded as follows:

[C]ontrary to the common-sense idea that adolescentsbecome involved in crime/delinquency because of a lackof social and/or human capital (e.g., social control andsocial disorganization theory) or due to an impulsivepersonality trait (e.g., self-control theory) adolescentsbecome delinquent if they are located in friendship net-works that support and facilitate delinquency.174

Kaplan: Self-Derogation TheoryThe self-derogation theory of Howard B. Kaplan proposesthat young people become involved in drug use, delinquentor criminal behavior, or other deviant behaviors becausethey have low self-esteem or self-derogation. Kaplan pro-posed that the following factors, explained by the mentionedtheories, can be integrated into self-derogation theory:

■ Stress caused by the inability or failure to live up tosociety’s expectations (strain theory)

■ School and family (control theory)

■ The influence of deviant friends (social learningtheory)

■ Self-concept (labeling and symbolic interactiontheories)

According to Kaplan, individuals take actions that max-imize their positive self-image and minimize their nega-tive self-image, a process he refers to as a self-esteemmotive. Many people are able to achieve positive self-images through this process, but those young people whoare not able to do so may turn to deviant acts and deviantpeers to enhance their own self-esteem and self-images.Kaplan and others tested the self-derogation theory pri-marily on adolescent drug abusers and found some sup-port for it.175

Tittle: Control Balance TheoryOne of the most complex of the integrated theories is con-trol balance theory, which attempts to combine deter-rence, rational choice, conflict theory, social learning,anomie, and social bonding theories and proposes toexplain the type of criminal behavior in which people willengage as well as why people engage in criminal behaviorat all. The major proponent of this approach, Charles Tit-tle, begins his description of the theory with the followinghypothetical situation.

You are a student in a class in which the teacher humil-iates you, for example, by calling you stupid. What canyou do? You could work very hard to prove to the teacherthat you are not stupid, but that could take a very longtime and might not be worth the effort. Furthermore, theother students who heard the humiliating put-down mightbe gone by then. You could hit the teacher, but, althoughthat might bring personal satisfaction, it also might resultin arrest, being charged with a crime, and being processedthrough the criminal justice system. Another option is tocriticize the teacher to others and to engage in question-able behavior in class. But that might result in a low gradeand limit your future opportunities. Wealthy studentsmight get someone to act on their behalf or, although rare,buy the school, dismiss all the teachers, and start a newschool. A final possibility Tittle suggests is that you giveup, leave school, abandon your dreams, and become deca-dent. So, how do you react?

According to Tittle, under control theory, you wouldweigh the options for reducing or eliminating your humil-iation with the cost of such action. In his words:

The central premise of the theory is that the amount ofcontrol to which people are subject relative to theamount of control they can exercise affects their gen-eral probability that they will commit specific types ofdeviance. Deviant behavior is interpreted as a device,or maneuver, that helps people escape deficits andextend surpluses of control. . . . [A person is predis-posed to engage in delinquent, criminal, or otherdeviant acts as the result of] an unbalanced controlratio, in combination with a desire for autonomy andfundamental bodily and psychic needs.176

This view assumes that control theory is central inexplaining why people conform, but it goes further andassumes that control is also crucial in explaining deviance.According to Tittle, those who are controlled by othersmay engage in deviant behavior in order to avoid or escapethat control, while those who control others may engagein deviant behavior in order to continue that control. It isnot control per se but, rather, control balance, meaning thatpeople have about as much control over others as othershave over them, that explains behavior. Those with themost control and those with the least control tend to bethe most deviant. The more balanced the control ratio, the


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 171

less likely it is that a person will engage in deviant behav-ior regardless of whether the direction of that imbalanceis negative or positive.

Originally, Tittle divided control balance into severalcategories, but recently he has revised his theory, aban-doned those categories, and classified deviant acts by con-trol balance desirability. In 2004, Tittle explained thisconcept as involving (1) the extent to which the actor willgain long-term control and (2) the extent to which the actcan be done indirectly by the actor. An example would bethe attempt of a company to drive another company outof business by massive price lowering. If it works, in thelong run, the company will gain more control. Further-more, the act is impersonal. Thus, the act has significantcontrol balance desirability. In his revision, Tittle alsoplaces greater emphasis on self-control, stating that per-sons who engage in deviance are generally rational, andthey select actions that will maximize their control balancedesirability. But those who have low social control arelikely to act more irrationally in their selections of behav-ior. For example, actors low in social control are morelikely to hit the teacher in the situation described at thebeginning of this section; those higher in social controlwould be more likely to make efforts to get the teacherfired.177

Although Tittle does relate his theory to traditionalvariables, such as social class, race, gender, age, marriage,parenting, and so on, he recognizes that his theory has notbeen tested directly. His theory is young, but some theo-rists have concluded, “Tittle’s theory does a nice job ofexplaining deviance committed by several segments ofsociety, including both skid-row bums and political andcorporate criminals.”178

Elliott, Ageton, and Canter:Integrated Strain/Control PerspectivePerhaps the most recognized of the theories that integratetraditional strain, social control, and social learning the-ories is that of Delbert S. Elliott, Suzanne S. Ageton, andRachelle J. Canter. These researchers proposed to developa theory that would explain how delinquency is generatedand maintained. They developed two paths, or a combi-nation of conditions, through which adolescents moveinto delinquent behavior. Probably the most frequentpath involves “weak bonds to conventional society andexposure and commitment to groups involved in delin-quent activity.” The second path incorporates “(1) strongbonds to conventional society, (2) conditions and expe-riences which attenuate those bonds, and, in mostinstances, (3) exposure and commitment to groups in-volved in delinquency.”179

Bonds or controls may be internal, which Elliott, Age-ton, and Canter referred to as integration, or they may beexternal, which they call commitment. Integration includesoccupying conventional social roles and participating in

activities, organizations, and institutions that are conven-tional. Integration also involves effective networks forsanctioning behavior. Commitment, or external bonds,include “perceived legitimacy of conventional norms,normlessness, social isolation, societal estrangement, pow-erlessness, attachment to parents and peers, belief in con-ventional goals and values, and tolerance for deviance.”180

Bonds may be reinforced or weakened by experiencesor conditions, such as social disorganization within thefamily, school, or community; failures and successes; andlabeling (both positive and negative). Even young peoplewith strong bonds to society may become involved indeviant behavior when faced with one or more of thesestressors. For example, although the most importantsocialization occurs within the family, if that process is notadequate, young people may become more bonded to theirpeers. If those bonds are stronger with delinquent thanwith conventional peers, the youth are more likely tobecome involved in deviant, delinquent, and even crimi-nal behavior.

According to Cullen and Agnew, Elliott, Ageton, andCanter found support for their integrated theory, con-cluding that the factor that is most predictive of delin-quency (other than prior delinquent acts) is associationwith peers who engage in delinquent behavior. The impactof those associations was greater on individuals who hadlow social control, which was influenced by strain.181

Recent support for the model was reported by CarloMorselli and associates, who studied offenders and theirmentors.182

Robinson: Integrated Systems Theoryof Antisocial BehaviorIn his recent book, Why Crime? An Integrated Systems The-ory of Antisocial Behavior,183 criminologist Matthew Robin-son explains his position concerning theory integration,which is based on integrating academic disciplines, not justconcepts or propositions. Robinson calls his approach theintegrated systems theory (IST). It begins with a list offacts that we know about crime, such as the fact that mostpeople who commit crimes begin as juveniles. Robinsonclaims that IST is consistent with all of the facts aboutcrime, and it is aimed at explaining the likelihood of engag-ing in antisocial behavior in general. Robinson lists sixphilosophical assumptions of his approach:

■ “People choose whether or not to commit crime.

■ People’s choices are influenced by factors beyondtheir control.

■ The facts that influence people’s choices (and hencetheir behaviors) are risk factors and protective factors.

■ These risk and protective factors exist among sixlevels of analysis, including cells, organisms, groups,communities/organizations, and society.


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 172

■ Exposure to risk factors generally increases the riskof antisocial and criminal behavior, especially whenexposure is frequent, regular, intense, and occursearly in life.

■ Exposure to protective factors generally decreasesthe risk of antisocial and criminal behavior, espe-cially when exposure is frequent, regular, intense,and occurs early in life.”184

Risk factors are found at six levels of analysis (cell,organ, organism, group, community, and society) andinclude anything that increases the tendency to engage inantisocial behavior, especially when the exposure is fre-quent, regular, and occurs early in life. Risk factors havebeen identified in numerous academic disciplines, frombiology to sociology. Robinson develops a timeline involv-ing risk factors, showing which factors are most likely tobe influential at which time in a person’s life. For exam-ple, genetics is most influential in early childhood,employment in early adulthood. Robinson concludes that“the more risk factors a person is exposed to during anystage of life, the more likely antisocial behavior willoccur.” These risk factors can be counteracted by protec-tive factors.185

Robinson’s developments are recent and have not yetbeen subjected to rigorous testing. However, criminolo-gists have reacted favorably to the theory, with StuartHenry and Mark M. Lanier describing Robinson’s work as“perhaps the most ambitious, comprehensive interdisci-plinary attempt so far to move integration of criminolog-ical theory to new heights.”186 Criminologist AnthonyWalsh, in reviewing Robinson’s book, states that his inte-grative theory makes “major contributions to criminol-ogy” and that the book “is a tour de force for the crimi-nologist who wants to learn something about the biosocialperspective.”187 Finally, the Danish National Televisionmade a documentary about Robinson’s book on integratedtheory.188

Agnew: General TheoryRobert Agnew, whose strain theory was discussed in Chap-ter 5, has recently contributed an integrated theory, whichhe refers to as a general theory that incorporates a broaderrange of variables than other integrated theories. For exam-ple, Agnew’s general theory includes personality traits(such as irritability), measures of social support, and peerabuse, to mention a few. Agnew states that his general the-ory incorporates all of the variables associated with othertheories as well as the mechanisms associated with thosetheories. Finally, it describes the relationships among thevariables.189

Agnew classifies the variables that cause crime intothese five clusters associated with life domains:

■ Self (irritability/low self-control)

■ Family (poor parenting practices, no/bad marriages)

■ School (negative school experiences, limited education)

■ Peers (peer delinquency)

■ Work (unemployment/bad jobs)190

According to Agnew, others organize the variables bytheories, but there are too many variables for individualexamination, and separating them may be misleading. Forexample, measuring the impact of family conflict is toonarrow. When family is a risk factor, there are probablyother variables associated with the family at work. It makessense to combine all the family variables together to mea-sure the effect of the family on behavior. The same goesfor other variables.191

Agnew continues with these propositions:

■ “The life domains have reciprocal effects on oneanother which vary over the life course. . . .

■ Crime affects the life domains and prior crimeaffects subsequent crime. . . .

■ Life domains interact in affecting crime and oneanother. . . .

■ Life domains have nonlinear [i.e., ‘As the level of alife domain increases it has an increasingly largereffect on crime and the other domains’]and largelycontemporaneous effects on crime and oneanother. . . .

■ Biological factors and the larger social environmentaffect the life domains.”192

Agnew concludes that his theory reflects the basic con-cepts and propositions of the major theories of crime andthat it is more than a patchwork result. He believes thathis theory provides an answer to the questions of “whythey do it?”—”although you, the reader, are the finaljudge.”193

Farrington: Integrated CognitiveAntisocial PotentialWe began our discussion of DLC theories with a referenceto a 2002 address by David P. Farrington, in which hereviewed the life course of the discipline of criminology. Inthat presentation, Farrington discussed the “ten widelyaccepted conclusions about the development of offendingthat any DLC theory must be able to explain.” He thenoffered his own contribution to integrated theory, whichhe referred to as the integrated cognitive antisocial poten-tial (ICAP) theory. This theory, according to Farrington,integrates concepts from strain, control, learning, labeling,and rational choice theories. The key concept of ICAP the-ory is antisocial potential (AP), which refers to a person’spotential to engage in antisocial acts. AP “assumes that thetranslation from antisocial potential to antisocial behaviordepends on cognitive (thinking and decision-making) pro-cesses that take account of opportunities and victims.” Far-rington distinguishes long-term from short-term AP. The


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 173

former “depends on impulsiveness, strain, modeling,socialization processes, and life events.” Short-term APdepends on “motivating and situational factors.”194

Individuals can be placed on a continuum from lowto high with regard to their AP (and this ordering per-sists over time), with few people having relatively highlevels, but those are the persons most likely to commitdeviant acts. AP levels vary with age and reach a peakduring the teen years “because of changes within indi-viduals in the factors that influence long-term AP (i.e.,from childhood to adolescence, the increasing impor-tance of peers, and decreasing importance of parents).”The primary factors influencing high AP are “desires formaterial goods, status among intimates, excitement, andsexual satisfaction. However, these motivations only leadto high AP if antisocial methods of satisfying them arehabitually chosen.” Those who have difficulty satisfyingtheir needs through legitimate means—such as personswith low incomes, those who are unemployed, and thosewho are not successful in school—are most likely toengage in antisocial methods. They will also be influencedby physical capabilities and behavioral skills. As Farring-ton points out, a small child would have difficulty steal-ing a car.195

Long-term AP is influenced by the processes of attach-ment (children who are not attached to prosocial parentswill have high AP) and socialization (children who areconsistently punished for bad behavior and rewarded forgood behavior will have low AP). Exposure to antisocialrole models increases long-term AP as does impulsiveness,as those who are impulsive tend to act quickly without giv-ing significant thought to their behavior. Life events affectAP; for example, AP decreases with marriage or movingfrom a high crime area; it increases after separation ordivorce. Interaction among these factors is also important.Thus, persons who are exposed to antisocial role modelsmay have higher AP if they also experience poor social-ization or strain. Biological factors (such as low intelli-gence) may be important.196

With regard to his theory, Farrington concludes thefollowing:

According to the ICAP theory, the commission ofoffenses and other types of antisocial acts depends onthe interaction between the individual (with hisimmediate level of AP) and the social environment(especially criminal opportunities and victims). Short-term AP varies within individuals according to short-term energizing factors such as being bored, angry,drunk, frustrated, or being encouraged by male peers.Criminal opportunities and the availability of victimsdepend on routine activities. Encountering a tempt-ing opportunity or victim may cause a short-termincrease in AP, just as a short-term increase in AP maymotivate a person to seek criminal opportunities andvictims.197

The factors that determine whether a person with a par-ticular level of AP actually commits a crime include thecognitive consideration of the costs and benefits of suchbehavior—for example, the material goods that might besecured; the chances of being apprehended by law enforce-ment officials; and the likely reaction of peers, parents, andpartners—as well as the person’s past experiences. Mostpeople, says Farrington, act in ways that they think arerational, “but those with low levels of AP will not commitoffenses even when (on the basis of subjective expectedutilities) it appears rational to do so.” And persons withhigh short-term AP levels—for example, levels caused byheavy drinking or anger—may commit offenses that arenot rational. Further, as a result of the social-learning pro-cess, the consequences of offending may create changes inlong-term AP as well as in the cognitive level of futuredecisions.198

In concluding his address to the 2002 American Soci-ety of Criminology, upon receiving the prestigious EdwinH. Sutherland award, David P. Farrington acknowledgedthe need for significant future research on his and otherintegrated theories, but, he said, “I am happy to end onan optimistic note. Given the current state of DLC theo-ries and DLC knowledge, there seems great scope forcumulative advancement of knowledge and for wide-spread agreement about key facts that need to be ex-plained and key assumptions that need to be included inDLC theories.”199

Conclusion to Integrated TheoriesThe theories discussed in this section give only a partial pic-ture of the extensive efforts by social scientists to integratetheories about crime. Some of the theories have been tested;others have not. All have met with mixed success. One ofthe strongest reactions has been from Travis Hirschi, dis-cussed earlier in this chapter in terms of his social bond-ing theory and general theory (with Gottfredson). Hirschimaintains that theories about crime should remain sepa-rate because they are not compatible; they have differentperspectives and should be examined on their own merits.Those who propose integration theories are not integrat-ing but, rather, developing new theories.200

Akers and Sellers conclude that reactions to integratedtheories are mixed, with some accepted and some evenignored. There appears to be a willingness to accept thesetheories, “but there continues to be a considerable indif-ference and a healthy skepticism toward integration as atheory-building strategy.”201

Miller, Schreck, and Tewksbury are more optimistic.They acknowledge that integrated theories have notreceived as much support as one might have predicted;criminologists remain excited about this approach, how-ever, and theoretical integration “is likely to continue as aprimary means of theory building for years to come.”202

Integrated theories are summarized in Exhibit 6.6.


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 174

Conclusion to CriminalBehavior TheoriesWe have looked at the various theories of criminal behav-ior: the economic, biological, physiological, psychological,psychiatric, and sociological approaches. All have method-ological problems: Crime is difficult to define in opera-tional terms, samples are limited, and follow-up studies areexpensive and time-consuming. All the research today is

plagued with an increasing lack of public interest in whatcaused the behavior; rather, the hue and cry is “Let the pun-ishment fit the crime,” although there is some evidence ofa return to treatment rather than punishment issues. Butstill, gaining an understanding of the reasons for the behav-ior does not always command a high priority in researchfunding.

What have we come to know about crime by analyzingofficial data? We know that most crimes are committed bymen but that some crimes have been increasing among


6.6

EX

HIB

IT Conceptual AbsorptionRonald L. Akers and Christine S. Sellers advocate subsum-ing or combining into one theory the concepts of another.They believe their approach integrates social learning, socialbonding, labeling, conflict, anomie, and deterrence theories.

Developmental and Life-Course (DLC) TheoriesThe DLC theories attempt to explain how offending andantisocial behavior develop, the risk factors persons face atdifferent ages, and the effect that various life events haveon development.

Age-Graded TheoryRobert J. Sampson and John H. Laub referred to age-gradedinformal social controls such as the bonding that occurswithin the family and schools, with peers, and, in later life,with marriage and jobs. A person’s bonding with these age-graded controls inhibits delinquent and criminal behavior.

Developmental TheoryTerri Moffitt, a psychologist, looks beyond a single dimen-sion and distinguishes two categories of antisocial adoles-cents: the adolescent limiteds (whose antisocial behaviorpeaks during adolescence and then declines) and the life-course persisters (who have social problems as children andcontinue their antisocial behavior even into adulthood).

Social Support or Coercion and CrimeFrancis T. Cullen and Mark Colvin believe that all crimi-nological theory can be integrated around the concept ofsocial support, which refers to social integration withingroup relationships in which individuals provide social,emotional, and material support for each other. The higherthe level of social support, the less likely crimes will result.

An Interactional Theory of DelinquencyTerence P. Thornberry combined social learning, socialbonding, and social structural theories, claiming that delin-quency is the result of the fact that children interact withtheir parents and others over time. All change is a resultof these interactions, and all must be considered in explain-ing delinquent and criminal behavior.

Network AnalysisMarvin D. Krohn uses social and personal networks to inte-grate social learning and social bonding theories.

Self-Derogation TheoryHoward B. Kaplan believes that delinquent behavior isexplained by low self-esteem or self-derogation. Personshave a self-esteem motive, meaning they take actions thatmaximize their positive self-image and minimize their neg-ative self-image.

Control Balance TheoryCharles Tittle’s theory of the ratio between how much a per-son needs to control and how much that person is able tocontrol. The more balanced the control ratio the less likelyit is that the individual will engage in deviant behavior.

An Integrated Strain/Control Perspective on DelinquencyA theory developed by Delbert S. Elliott, Suzanne S. Age-ton, and Rachelle J. Cantor to integrate traditional strain,social control, and social learning theories.

Integrated Systems Theory of AntisocialBehaviorMatthew Robinson’s approach integrates academic disci-plines as well as concepts and propositions. He begins witha list of facts we know about crime, looks at risk factorsat six levels of analysis, and advocates that his approachis aimed at explaining antisocial behavior in general.

General TheoryRobert Agnew contends that his general theory incorpo-rates a broader range of variables (such as personality traits)than other integrated theories and describes the relation-ships among those variables.

The Integrated Cognitive Antisocial Potential(ICAP) TheoryDavid P. Farrington’s effort to integrate strain, control,learning, labeling, and rational choice theories is based onthe concept of long-term or short-term antisocial potential(AP), which refers to a person’s potential to engage in anti-social acts.

Integrated Theories: A Summary

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 175

women. We know that most crimes are committed by per-sons who are mobile and who live in large cities. We knowthat among juveniles, those who are closely bonded to theirfamilies and schools are less likely to commit delinquentacts or crimes than those who are not so closely bonded.We know that those who are unemployed and hovering atthe bottom of society’s class structure are more likely thanthose who are at the top of the social structure to be appre-hended for serious property crimes. But the official datamust be considered along with data about white-collarcrimes (discussed in Chapter 9), which involve greaterfinancial losses to society than the serious property crimesof arson, larceny-theft, burglary, and motor vehicle theft.

We know that according to official data, a dispropor-tionate amount of crime is committed by African Ameri-cans, and we know that most crimes are committed by per-sons who are not married. So, how do we make sense ofthese and other crime data? That is what the theories, andthe tests of those theories, are about. If you become dis-couraged because we do not know the cause of crime, per-haps you can compare this venture to health problems forwhich we have not found the cause, problems such as can-cer and the common cold. We know about the virus thatcauses AIDS, but we have not found a vaccination or acure. Physical scientists continue forming and testing theirtheories in an effort to find answers to these and otherhealth problems. Likewise, social scientists continue towork toward understanding the causes of crime.

Some social scientists look for a general theory to explainall crime (or perhaps all behavior); we have seen some ofthose approaches that look at the social-structural or social-process theories. In 1990, after reviewing 20 years of testingtheories of criminal behavior, Thomas J. Bernard concludedthat we had made tremendous progress in empiricalresearch on crime but little progress on theoretical devel-opments.203 In 1991, Roland Chilton expressed his view thata general theory to explain all crime cannot be discovered:“Crime is too variable, too influenced by moral and politi-cal entrepreneurs to be seen as a unitary phenomenon.”204

More recent research has, perhaps, been more promis-ing, but criminologists still have a long way to go. AsGeorge B. Vold, Thomas J. Bernard, and Jeffrey B. Snipesstate in their analysis of the state of criminological theory,“criminology researchers simply have not fully sorted outthe issues.”205

But, they conclude, taken together, the research doesprovide us with the following characteristics associatedwith criminal behavior:

1. “A history of early childhood problem behaviors andof being subjected to poor parental childrearing tech-niques, such as harsh and inconsistent discipline; schoolfailure and the failure to learn higher cognitive skills suchas moral reasoning, empathy, and problem solving.

2. Certain neurotransmitter imbalances such as low sero-tonin, certain hormone imbalances such as high testos-

terone, central nervous system deficiencies such as frontalor temporal lobe dysfunction, and autonomic nervous sys-tem variations such as unusual reactions to anxiety.

3. Ingesting alcohol, a variety of illegal drugs, and sometoxins such as lead; head injuries; and complications dur-ing subject’s pregnancy or both.

4. Personality characteristics such as impulsivity; insen-sitivity, a physical and nonverbal orientation, and a ten-dency to take risks.

5. Thinking patterns that focus on trouble, toughness,smartness, excitement, fate, and autonomy, and a ten-dency to think in terms of short-term rather than long-term consequences.

6. Association with others who engage in and approveof criminal behavior.

7. Weak attachments to other people, less involvementin conventional activities, less to lose from committingcrime, and weak beliefs in the moral validity of the law.

8. A perception that there is less risk of punishment forengaging in criminal behavior.

9. Masculinity as a gender role.”206

In addition, the researchers list these factors as increasingthe probability of victimiztion: “Frequently being awayfrom home, especially at night; engaging in public activi-ties while away from home; and associating with peoplewho are likely to commit crime.”207

The fact that we have differences and problems withcriminological research does not mean that we shouldabandon our efforts to develop and test theories thatexplain the causes of crime. It does mean that we must becareful in our interpretations of the existing theories,research, and data on crime.

It is important that we do not conclude that researchon criminal behavior is hopeless. As sociologist George A.Lundberg stated in 1961, “Many of the fruits of sci-ence . . . can be used to advantage while still in the pro-cess of development. Science is at best a growth, not a sud-den revelation. . . . We do not abandon cancer researchbecause the patients of today may not be saved by it.”Research in the social and physical sciences can be used“imperfectly and in part while it is developing.”208

In reemphasizing the importance of theory, a 2001statement by the editor of the leading professional journalin the field noted why the annual meetings of the Ameri-can Society of Criminology require that research papers begrounded in theory. Robert J. Bursik Jr. quoted a passagefrom a 1963 book by noted sociologist Peter Berger enti-tled Invitation to Sociology: “Statistical data by themselvesdo not make sociology. They become sociology only whenthey are sociologically interpreted, put within a theoreti-cal frame of reference.” Bursik emphasized that althoughanyone can take a set of data and explain what is there—for example, more men than women commit crime—“atruly criminological contribution to the field not only


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 176

documents the ‘what,’ but also offers insightful argumentsabout the ‘why.’” Bursik continued by noting that this ispossible only “if the selection of variables upon which tofocus and the interpretation of the patterns that are iden-tified are grounded in theoretical considerations.”209

But there is concern that criminologists are not beingheard by policy makers (public policy is discussed next).In 2003, James Austin stated as follows:

Despite the annual publication of hundreds of peer-reviewed articles and textbooks proudly displayed atour annual conventions, policy makers are paying lit-tle if any attention to us. When Congress or state leg-islatures debate new crime bill legislation or the effec-tiveness of past actions, their first question is not,“What do the criminologists think?” I would venturethat one would be hard pressed to cite another disci-pline that has been so ignored for such a long time.[Another scholar’s] lament that criminology has beenunable to evaluate the relative effects of various lawenforcement strategies on illegal drug consumption useis another example of how impotent criminology is inshaping and improving criminal justice policy. . . .[With regard to policy changes] over the years Ilearned, . . . such efforts amounted to nothing morethan spitting in the wind.210

It is critical that criminologists, sociologists, and othersocial scientists communicate their theoretical explana-tions and empirical findings in language that can be under-stood (and, we hope, accepted!) by those who are in a posi-tion to make policy decisions. The chapter closes withsome examples of such decisions that are based on socialscience research.

Policy Implications ofCriminal Behavior TheoriesPublic policy decisions regarding how to deal with crimeare made at all levels of government as well as within otherinstitutions, such as families, churches, and schools. Thesedecisions may be made after a careful analysis of theresearch data; they may be made on the basis of intuition;they may be the result of political decisions; or they madeinvolve all these approaches. In most cases, we really do notknow the reasons policy decisions are made. We may begiven reasons, but those are not always accurate. Part of theproblem lies with social scientists, who do not always com-municate their findings and theoretical developments in amanner that is readily accessible and fully understood bythose in policy-making positions. At times, however, thelatter simply refuse to listen. Still, it is important to con-sider the policy implications of what we know (or do notknow) about crime.

Throughout this text you will read about administra-tive rules and regulations, local ordinances, state and fed-

eral laws, court decisions, and even some constitutionalchallenges that will have an impact on the way we react todeviant, juvenile, and criminal behavior. We have alreadynoted that even the definitions of the word crime are notalways uniform. And even when we agree on what is acrime, we do not always count the acts or react to them inthe same ways in various jurisdictions. It is suggested thatas you progress through the text, you consider the theo-ries you have learned in Chapters 3, 4, 5, and 6 and ques-tion their relevance to the policy decisions that have beenmade.

Let us consider a few examples. In his presidentialaddress before the American Society of Criminology inNovember 2000, Roland Chilton spoke about public pol-icy. He referred to a meeting in which a group of scien-tists were discussing murders that appeared to be drugrelated. Chilton suggested that it might be time to reex-amine the government’s policies regarding drug enforce-ment. The implication was that we might consider legal-izing drugs. The response was that such a suggestion wouldnot be “viable policy.” In other words, it would not bepolitically acceptable. Chilton saw this response as a seri-ous problem.211

In Chapter 10, we will look at some aspects of the crim-inalization of drugs in greater detail, but the purpose isnot to advocate that drugs should be legalized. Rather, itis to examine the implications of the direction we havetaken in the so-called war on drugs. Chilton refers to thiswar as “the folly of our time.” By folly he means “a costlyundertaking that has an absurd or ruinous outcome” or“conduct that lacks good sense, understanding, or fore-sight.” According to Chilton, the U.S. drug policies are“one of the best examples of criminal justice folly in ourlifetime. Our drug laws are indeed tragically foolish andexcessively costly.” And, he noted, they consume enor-mous resources within our criminal justice systems.

If research shows that the long sentences that we havebeen assessing for minor drug offenses have had little orno impact on drug sales (i.e., no impact on the big drugdealers), have had no impact on rehabilitation of drugoffenders, and have crowded many prisons almost beyondour capacity to process more serious offenders, should werefuse to look for new ways to deal with the issue? Orshould we look more closely at the position taken in thefall of 2000 by California voters (and discussed later in thistext) and provide treatment rather than punishment forminor drug offenses? Also, since Chilton made this speech,some of the harsh drug laws have been revised, as we willnote later.

Chilton ended his speech by stating: “Political viabilityis of little value if it produces foolish and unfair policiesand practices.”212

In connection with Chilton’s thoughts, we might con-sider whether marijuana should be permitted for medici-nal reasons and whether nonviolent drug offenders shouldbe processed through special courts (discussed in Chapter


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 177

13). In a 2003 research publication, Denise Gottfredson,Stacy S. Najaka, and Brook Kearley reported their analy-sis of prior research on the effectiveness of drug treatmentcourts. They concluded: “Drug use and criminal activityare substantially reduced for drug court participants whilethey are in the program” and that drug courts have “gen-erated savings in jail costs, probation supervision, policeovertime, and other criminal justice costs, and they havefostered savings outside of the criminal justice system aswell.” In their own study of the Baltimore city drug treat-ment court, which was established in 1990, these researchersfound a reduction in criminal offending. “Drug subjectswho participated in ten or more consecutive days of cer-tified drug treatment were much less likely to recidivatethan were both untreated drug court subjects and controlsubjects.”213

John S. Goldkamp emphasized that the impact of thedrug court movement “is hard to ignore. . . . [The] con-tribution of the drug court innovation leaves a mark onjustice practices far beyond what consistently positive find-ings from the most rigorous evaluation of reoffending ratescould probably ever produce.”214

Not all policies have shown positive results. Considerthose requiring the mandatory arrests of alleged domesticbatterers. Robert C. Davis, Barbara E. Smith, and Bruce Tay-lor published the results of their analysis of problems thatresulted when such policies were implemented in Milwau-kee, Wisconsin. These researchers found that increasedprosecutions of those alleged to have engaged in domesticviolence may result in court delays, increased recidivismprior to trial, and fewer convictions. They stated:

Arresting more batterers does not necessarily result inmore prosecutions. . . . [W]ell-intended policiesregarding domestic violence can backfire. . . . Goodintentions of policy makers need to be balanced with arealistic view of how the criminal justice system func-tions and what the actual effects of new policies arelikely to be.215

Research by Laura Dugan produced similar results. Sheconcluded that although police should continue to beaggressive in domestic violence cases, “local officials should

recognize that mandatory arrest laws could reduce thenumber of cases that enter the system” and that when thestate awards domestic violence victims the custody oftheir children, it should provide those children with ade-quate protection and needed services.216

Another policy issue is what to do about repeat offend-ers. Later in this text we consider special laws recentlyenacted in all states and the federal government. Theselaws provide for very long sentences—such as 25 years tolife without parole—for repeat offenders even when theoffenses are not serious (although most do require a seri-ous third offense to invoke the statutes). This situation hasincreased prison populations and contributed little to pub-lic safety.

But does it make sense that putting people in prison forlonger periods of time will deter them from committingcrimes as well as provide a warning to others? Yes, it mightseem logical, but what did we learn from the classical the-orists about letting the punishment fit the crime? Did itwork then? Will it work now?

The importance of considering the policy implicationsof research and theory in criminology is dramatically il-lustrated by the most recent edition of the criminologytheory book by Ronald L. Akers and Christine S. Sellers:Criminological Theories: Introduction, Evaluation, andApplication. These scholars proclaimed the following:

Every criminological theory implies a therapy or a pol-icy. . . . All major criminological theories have impli-cations for, and have indeed been utilized in, criminaljustice policy and practice. Every therapy, treatmentprogram, prison regimen, police policy, or criminal jus-tice practice is based, either explicitly or implicitly, onsome explanation of human nature in general or crim-inal behavior in particular.217

Akers and Sellers discuss the policy implications of allcriminological theories, giving detailed information onspecific programs that have been instituted as a result ofan acceptance of each theory. Their examples range frompolicies of involuntary sterilization to fundamentalchanges within institutions based on the theory that crim-inal justice policies have been sexist.


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 178


Summary

This chapter examined theories that emphasize theprocess by which a person becomes a criminal. Manytheories were discussed, and several exhibits were

presented to summarize them for you.The chapter began with a brief overview of learning the-

ory before looking more closely at the contributions ofGabriel Tarde, who used his imitation theory to explainhow people learn acceptable as well as deviant, delinquent,or criminal behavior. That set the backdrop for consider-ation of the modern media’s influence on behavior, espe-cially criminal behavior.

The chapter then turned to a more extensive analysisof Edwin H. Sutherland’s theory of differential association.Sutherland viewed his theory as one that explains allbehavior, not just deviance, delinquency, or crime. Indi-viduals learn those behaviors in the same ways they learnacceptable behaviors. Sutherland’s work influenced othersocial scientists, many of whom attempted to test thepropositions of differential association.

A second type of learning theory is that of differentialassociation-reinforcement theory, in which Burgess andAkers attempted to integrate Sutherland’s theory with thatof the psychologist B. F. Skinner, and, eventually forAkers, the development of his own version of social learn-ing theory.

The influence of sociology may be seen in the applica-tion of social control theories as a supplement to (orreplacement for) psychological control theories (discussedin Chapter 4) in explanations of criminal and delinquentbehavior. Control theories explain deviance in terms of abreakdown in social controls. Reiss and Nye’s control the-ory approach looked at the impact of personal inner con-trols and external social structures on behavior. Reckless’scontainment approach viewed social controls as providinginsulators or buffers against improper behavior. Hirschideveloped control theory in terms of his position that it isconforming behavior that must be explained. He did so interms of four components of social bonds: attachment toconventional persons, commitment to conventional behav-ior, involvement with conventional people, and belief inconventional norms. In testing his approach by means of self-report studies, Hirschi found a lack of support forthe traditionally reported high association between lowsocioeconomic class and crime. Rather, he found that weakattachments to parents and to school are more predictiveof delinquent and criminal behavior than class is and thatpositive attitudes toward one’s own accomplishments aremore characteristic of law-abiding than of law-violatingyouth.

Gottfredson and Hirschi developed a general self-control theory. They argued that low self-control, which

results primarily from poor parenting, is the best predic-tor of criminal behavior. They attempted to measure thecharacteristics of persons with low self-control and relatethem to criminal behavior.

The chapter next focused on labeling theory, whichconsiders why, of all the people who engage in deviant,delinquent, or criminal behavior, some are processedthrough criminal justice systems while others are ignored.The emphasis is on the external labeling process ratherthan on the reason for the behavior in question. Labelingtheory has been applied to criminal behavior since the1930s but has received more attention in recent years. Thechapter discussed the earlier works of Tannenbaum,Becker, and others before focusing on more recentresearch, much of which tests the approach utilizing juve-nile subjects. The contributions of Braithwaite’s restora-tive justice, Matsueda’s reflective appraisal, and Sampsonand Laub’s cumulative disadvantages were examined.

A major section of this chapter’s discussion of social-process theories was devoted to integrated theories. Theelements of integration were explained, including the con-tributions of Akers’s conceptual absorption approach.

Extensive coverage was given to developmental and life-course theories, with emphasis on the age-graded approachof Sampson and Laub. This approach considers the age-graded, informal social controls such as the bonding thatoccurs within the family and schools, with peers, and, inlater life, with marriage and jobs. A person’s bonding withthese age-graded controls inhibits delinquent and criminalbehavior. The section closed with an examination of Mof-fitt’s developmental theory.

Cullen and Colvin attempted to integrate criminologi-cal theories around the concept of social support or coer-cion and proposed that crime is caused by coercion andprevented by social support. For Thornberry, social learn-ing, social bonding, and social-structural theories can becombined into interactional theory, whereas Krohn usedsocial and personal network analysis to integrate sociallearning and social bonding theories. Under Kaplan’s self-derogation theory delinquent behavior is explained by lowself-esteem or self-derogation.

Tittle’s control balance theory refers to the ratiobetween how much a person needs to control and howmuch that individual is able to control. The more balancedthe control ratio, the less likely it is that a person willengage in deviant behavior. Elliott, Ageton, and Canter uti-lize an integrated strain/control perspective to combinetraditional strain, social control, and social learning theo-ries. This approach is probably the most popular andaccepted of the integrated theories.

Robinson’s integrated systems theory of antisocial be-havior is directed as integrating academic disciplines aswell as theoretical concepts and propositions. Although

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 179


recently developed, the theory has been well received inthe profession and by the media.

Agnew has contributed a general theory, which hebelieves integrates more than other theories, includingsuch variables as personality traits, generally excluded byother integrated theories. He classifies the variables he con-siders into five categories: self, family, school, peers, andwork.

Finally, the integrated cognitive antisocial potential(ICAP) theory of Farrington was noted. Farrington claimsthat this theory integrates strain, control, learning, label-ing, and rational choice theories.

After a conclusion on integrated theories, the chaptercontinued with a brief analysis of criminal behavior theo-ries. Social-process theories in particular have been instru-mental in achieving a better understanding of criminalbehavior. As a group, social-process theories may be moreacceptable than other current approaches of sociologistsand criminologists. The theories attempt to explain dif-ferential reaction to the social structure; they are based on the sociological proposition that criminal behavior isacquired through social interaction just as noncriminalbehavior is acquired.

The chapter closed with a discussion of some of the pol-icy implications of sociological theories.

This chapter concluded the text’s formal analysis of the-ories, although references will be made to many of themthroughout the remainder of the text. Chapter 7 begins thestudy of particular crimes.

Study Questions1. Define learning theory.2. What was Tarde’s contribution to criminology, and

what is its relevance today? What is the impact of themedia on crime?

3. Discuss the impact of Edwin H. Sutherland oncriminology, and summarize the reactions to his majortheoretical developments.

4. State the elements of differential association.5. Discuss and evaluate Akers’s social learning theory.6. Explain the control theory of Reiss and Nye.7. Explain and critique Hirschi’s bonding theory.8. Why is the social control theory of Gottfredson and

Hirschi considered a general theory?9. Explain the primary difference between labeling theory

and other explanations of deviant behavior.10. Are the effects of labeling always negative? Discuss.11. Explore Braithwaite’s concept of restorative justice and

relate it to modern criminal justice systems.12. Define reflective appraisal, and discuss Matsueda’s

development of this approach.

13. Define cumulative disadvantages, and discuss Sampsonand Laub’s development of this concept.

14. Discuss briefly the use of labeling theory to explainjuvenile delinquency. What are the racial implications ofthe theory?

15. What is meant by integrated theories? Define anddiscuss conceptual absorption.

16. Name and briefly explain the integrated theoriesdiscussed in the chapter. Which do you think have thegreatest promise for explaining delinquency and crime?

Brief Essay Assignments1. Define and contrast social learning theory and social

control theory, and compare the contributions of themajor proponents of each.

2. Define, explain, evaluate, and discuss the impact of self-control theory as developed by Gottfredson and Hirschi.

3. Explain how labeling theory differs from social controltheories.

4. Select the four most important integrative theories andexplain your choices.

5. Discuss the policy implications of criminological theories.

Internet Activities1. Learn about efforts to control violence in the media and

the legal implications of those efforts by logging on tothe Media Coalition site, http://www.mediacoalition.org/reports/index.htm., accessed 20 February 2008. What arethe major legal arguments concerning efforts to censorthe media? How should they be resolved?

2. Search newspapers, magazines, or journals, and theInternet for information on a recent well-publicized crime.Find out as much information as you can about thedefendant(s), victim(s), and offense(s). In your opinion,do any of the theories discussed in this and the previouschapters explain or contribute to explaining the actionsthat occurred? If so, which one(s)?

Notes1. See Gabriel Tarde, “Penal Philosophy,” in The Her-

itage of Modern Criminology, ed. Sawyer F. SylvesterJr. (Cambridge, MA: Schenkman, 1972), p. 84.

2. For further details on Tarde’s theories, see MargaretS. Wilson Vine, “Gabriel Tarde,” in Pioneers inCriminology, 2d ed., ed. Herman Mannheim(Montclair, NJ: Patterson Smith, 1973), p. 292.Tarde’s book is Penal Philosophy, trans. R. Howell(Boston: Little, Brown, 1912).

3. Tarde, “Penal Philosophy,” p. 90.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 180


4. Vine, “Gabriel Tarde,” p. 295.

5. Tarde, “Penal Philosophy,” p. 94.

6. Vine, “Gabriel Tarde,” p. 295.

7. Stephen Schafer, Theories in Criminology (NewYork: Random House, 1969), p. 239.

8. See Robert T. Bower, The Changing Television Audi-ence in America (New York: Columbia UniversityPress, 1985); and Mediamark Research, MultimediaAudiences (New York: Mediamark Research, 1987).

9. “Television Gets Closer Look as a Factor in Real Vio-lence,” New York Times (14 December 1994), p. 1.

10. Albert Bandura, Social Learning Theory (EnglewoodCliffs, NJ: Prentice Hall, 1977), p. vii.

11. Albert Bandura, “The Social Learning Perspective:Mechanisms of Aggression,” in Psychology of Crimeand Criminal Justice, ed. Hans Toch (New York:Holt, Rinehart & Winston, 1979), pp. 204–205.

12. J. Ronald Milavsky, TV and Violence, NationalInstitute of Justice (Washington, DC: U.S. Depart-ment of Justice, 1988), p. 3.

13. James Q. Wilson and Richard J. Herrnstein, Crimeand Human Nature (New York: Simon & Schuster,1985).

14. See David Pearl et al., Television and Behavior: TenYears of Scientific Progress and Implications for theEighties (Washington, D.C.: U.S. GovernmentPrinting Office, 1982). For the full report, see J.Ronald Milavsky et al., Television and Aggression: APanel Study (New York: Academic Press, 1983).

15. Tannis MacBeth Williams, “Summary, Conclusions,and Implications,” in The Impact of Television: ANatural Experiment in Three Communities, ed.Tannis MacBeth Williams (New York: AcademicPress, 1986), p. 411.

16. Walter C. Bailey, “Murder, Capital Punishment,and Television: Execution Publicity and HomicideRates,” American Sociological Review 55 (October1990): 628–633; quotation is on p. 633.

17. Ray Surette, Media, Crime, and Criminal Justice:Images and Realities (Pacific Grove, CA:Brooks/Cole, 1992), pp. 108, 118.

18. “Media Violence Plays Part in Shootings: Ex-ArmyRanger Argues TV, Films Affect Children,” Wash-ington Times (10 November 1998), p. 2.

19. Jana Bufkin, “Book Review Essay: Exploring the Mediaand Crime,” Criminal Justice Review, 26, no. 1 (Spring2001): 62–68; quotations are on pp. 63–64.

20. Albert K. Cohen et al., The Sutherland Papers (Bloom-ington: Indiana University Press, 1956), p. 19.

21. Edwin H. Sutherland and Donald R. Cressey, Crim-inology, 10th ed. (Philadelphia: Lippincott, 1978),

pp. 80–82, emphasis omitted; Edwin H. Sutherland,Donald R. Cressey, and David F. Luckenbill, Crimi-nology, 11th ed. (Dix Hills, NY: General Hall, 1992).

22. Edwin H. Sutherland and Donald R. Cressey, Crim-inology, 5th ed. (Philadelphia: Lippincott, 1960), pp. 78–79.

23. George B. Vold and Thomas J. Bernard, TheoreticalCriminology, 3d ed. (New York: Oxford UniversityPress, 1986).

24. James F. Short Jr., “Differential Association and Delin-quency,” Social Problems 4 (January 1957): 233.

25. James F. Short Jr., “Differential Association as aHypothesis: Problems of Empirical Testing,” SocialProblems 8 (Summer 1960): 14–25.

26. Albert J. Reiss Jr. and Albert L. Rhodes, “An Empir-ical Test of Differential Association Theory,” Jour-nal of Research in Crime and Delinquency 1 (Febru-ary 1964): 5–18; quotation is on p. 12.

27. Cohen et al., The Sutherland Papers, p. 37.

28. Donald R. Cressey, “The Theory of DifferentialAssociation: An Introduction,” Social Problems 8(Summer 1960): 3.

29. Edwin H. Sutherland and Donald R. Cressey, Crim-inology, 9th ed. (Philadelphia: Lippincott, 1974), p. 78. This position was retained in the 10th edi-tion, published in 1978.

30. Sutherland and Cressey, Criminology, 10th ed., p. 84.

31. See, for example, Ross L. Matsueda and KarenHeimer, “Race, Family Structure, and Delinquency:A Test of Differential Association and Social Con-trol Theories,” American Sociological Review 52(1987): 826–840.

32. Mark Warr, “Age, Peers, and Delinquency,” Crimi-nology 31 (February 1993): 17–40, emphasis in theoriginal; quotation is on p. 19.

33. Mark Warr and Mark Stafford, “The Influence ofDelinquent Peers: What They Think or What TheyDo,” Criminology 29 (November 1991): 851–866;quotations are on pp. 851 and 864.

34. Ross L. Matsueda, “The Current State of Differen-tial Association Theory,” Crime & Delinquency 34(July 1988): 277–306; quotations are on p. 277.

35. See, for example, Robert L. Burgess and Ronald L.Akers, “A Differential Association-ReinforcementTheory of Criminal Behavior,” Social Problems 14(Fall 1966): 128–147.

36. Christine S. Sellers and Ronald L. Akers, “SocialLearning Theory: Correcting Misconceptions,” inThe Essential Criminology Reader, ed. Stuart Henryand Mark M. Lanier (Boulder, CO: Westview Press,2006), pp. 89–99; quotations are on p. 90. See also

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 181


Ronald L. Akers et al., “Social Learning and DeviantBehavior: A Specific Test of a General Theory,”American Sociological Review 44 (August 1979):637–638.

37. Ronald L. Akers and Christine S. Sellers, Crimino-logical Theories: Introduction, Evaluation, and Appli-cation, 4th ed. (Los Angeles: Roxbury, 2004), p. 87.

38. Sellers and Akers, “Social Learning Theory,” p. 91.

39. Akers and Sellers, Criminological Theories, 4th ed.,pp. 96–97.

40. Sellers and Akers, “Social Learning Theory,” p. 96.

41. Akers and Sellers, Criminological Theories, 4th ed.,p. 91.

42. Ibid., pp. 93, 94, citations omitted.

43. Ibid., pp. 94, 95, 96; quotation is on p. 94.

44. Ibid., p. 99.

45. Mark Warr, Companions in Crime: The SocialAspects of Criminal Conduct (Cambridge, UK: Cam-bridge University Press, 2002), p. 42.

46. Ibid., pp. 78–79, as quoted in Akers and Sellers,Criminological Theories, 4th ed., p. 101.


48. Travis Hirschi, Causes of Delinquency (Berkeley:University of California Press, 1969), p. 34.

49. See Albert J. Reiss Jr., “Delinquency as the Failureof Personal and Social Controls,” American Socio-logical Review 16 (1951): 196–207.

50. F. Ivan Nye, Family Relationships and DelinquentBehavior (New York: Wiley, 1958), as quoted inAkers and Sellers, Criminological Theories, 4th ed., p. 113.

51. Ronald L. Akers, Criminological Theories: Intro-duction, Evaluation, and Application, 3d ed. (LosAngeles: Roxbury, 2000), p. 102.

52. Walter C. Reckless, “Containment Theory,” in TheSociology of Crime and Delinquency, 2d ed., ed.Marvin E. Wolfgang et al. (New York: Wiley, 1970), pp. 401–402.

53. Ibid.

54. Walter C. Reckless, The Crime Problem, 5th ed.(Englewood Cliffs, NJ: Prentice-Hall, 1973), pp.55–57; reprint of article from Federal Probation 25(December 1961): 42–46.

55. Harwin L. Voss, “Differential Association and Con-tainment Theory: A Theoretical Convergence,” inSociety, Delinquency and Delinquent Behavior, ed.Harwin L. Voss (Boston: Little, Brown, 1970), pp. 198, 206.

56. Akers and Sellers, Criminological Theories, 4th ed., p. 116.

57. Travis Hirschi, Causes of Delinquency (Berkeley andLos Angeles: University of California Press, 1969),pp. 16–34. This classic work was republished in2002 (New Brunswick, NJ: Transaction, 2002).

58. Ibid., p. 21.

59. Ibid.

60. Ibid., p. 26.

61. Ibid., pp. 54, 56.

62. Ibid., pp. 72, 108, 132, 134.

63. See Michael J. Hindelang, “Causes of Delinquency: A Partial Replication and Extension,” Social Prob-lems 20 (Spring 1973): 471–487.

64. Rand D. Conger, “Social Control and Social Learn-ing Models of Delinquent Behavior: A Synthesis,”Criminology 14 (May 1976): 19, 35.

65. See Delbert S. Elliott and Harwin L. Voss, Delin-quency and Dropout (Lexington, MA: LexingtonBooks, 1974).

66. Eric Linden and James C. Hackler, “Affective Tiesand Delinquency,” Pacific Sociological Review 16(January 1973): 27–46.

67. Bill McCarthy, Diane Felmlee, and John Hagan,“Girl Friends Are Better: Gender, Friends, andCrime Among School and Street Youth,” Criminol-ogy 42, no. 4 (November 2004): 805–835.

68. Hirschi, Causes of Delinquency, pp. 230–231.

69. Robert Agnew, “Social Control Theory and Delin-quency: A Longitudinal Test,” Criminology 23 (February 1985): 58, 59.

70. Randy L. LaGrange and Helene Raskin White, “AgeDifferences in Delinquency: A Test of Theory,”Criminology 23 (February 1985): 36.

71. Thomas J. Bernard, “Structure and Control: Recon-sidering Hirschi’s Concept of Commitment,” JusticeQuarterly 4 (September 1987): 409.

72. Ibid., p. 422.

73. Ibid., p. 423.

74. Stephen A. Cernkovich and Peggy C. Giordano,“School Bonding, Race, and Delinquency,” Crimi-nology 30 (May 1992): 261–291; quotation is on p.286.

75. Allison Ann Payne, Denise C. Gottfredson, andGary D. Gottfredson, “Schools as Communities:The Relationships Among Communal School Orga-nization, Student Bonding, and School Disorder,”Criminology 41, no. 3 (August 2003): 749–777;quotation is on p. 751.

76. John H. Laub, “The Life Course of Criminology inthe United States: The American Society of Criminol-ogy 2003 Presidential Address,” Criminology 42, no. 1(2004): 1–26; quotations are on pp. 11 and 12.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 182


77. Michael R. Gottfredson and Travis Hirschi, A Gen-eral Theory of Crime (Palo Alto, CA: Stanford Uni-versity Press, 1990), p. 87.

78. Travis Hirschi and Michael R. Gottfredson, eds.,The Generality of Deviance (New Brunswick, NJ:Transaction), 1994, p. 3.

79. See Alex R. Piquero and Jeff A. Bouffard, “Some-thing Old, Something New: A Preliminary Investi-gation of Hirschi’s Redefined Self-Control,” JusticeQuarterly 24, no. 1 (March 2007): 1–26, with thespecific reference on p. 16.

80. Gottfredson and Hirschi, A General Theory ofCrime, p. xiv.

81. Ibid., p. 15.

82. Ibid., pp. 89–90.

83. Ibid., p. 97.

84. Ibid., pp. 89, 95–97.

85. Ibid., p. 117.

86. Christine S. Sellers, “Self-Control and Intimate Vio-lence: An Examination of the Scope and Specifica-tion of the General Theory of Crime,” Criminology37 (May 1999): 375–404.

87. Carter Hay, “Parenting, Self-Control, and Delin-quency: A Test of Self-Control Theory,” Criminol-ogy 39, no. 3 (August 2001): 707–736.

88. James D. Unnever, Francis T. Cullen, and Travis C.Pratt, “Parental Management, ADHD, and Delin-quent Involvement: Reassessing Gottfredson andHirschi’s General Theory,” Justice Quarterly 20, no. 3(September 2003): 471–500; quotation is on p. 471.

89. Carter Hay and Walter Forrest, “The Developmentof Self-Control: Examining Self-Control Theory’sStability Thesis,” Criminology 44, no. 4 (November2006): 739–774; quotation is on p. 757.

90. Eric A. Stewart et al., “Integrating the General The-ory of Crime into an Explanation of Violent Vic-timization Among Female Offenders,” Justice Quar-terly 21, no. 1 (March 2004); 159–181, quotation ison p. 175.

91. Raymond Paternoster and Robert Brame, “On theAssociation Among Self-Control, Crime, and Anal-ogous Behaviors,” Criminology 38, no. 3 (August2000): 971–982; quotation is on p. 980.

92. Richard P. Wiebe, “Reconciling Psychopathy andLow Social-Control,” Justice Quarterly 20, no. 2(June 2003): 297–336; quotation is on p. 301.Wiebe was referring to the earlier work of HerveyCleckley, The Mask of Sanity (St. Louis, MO:Mosby, originally published in 1941; 4th ed. pub-lished in 1964).

93. Wiebe, “Reconciling Psychopathy and Low SocialControl,” p. 299.

94. Ibid., p. 303.

95. Ibid., pp. 322, 323.

96. Piquero and Bouffard, “Something Old, SomethingNew.” The quotation from Hirschi is on p. 3 of thisarticle and was taken from Travis Hirschi, “Self-Control and Crime,” in Handbook of Self-Regulation:Research, Theory, and Applications, ed. R. F.Baumeister and K. D. Vohs (New York: GuilfordPress, 2004), p. 543.


98. See, for example, the works of Mark Colvin, Crimeand Coercion: An Integrated Theory of Chronic Crimi-nality (New York: St. Martin’s Press, 2000); andRobert Agnew, “Foundation for a General StrainTheory of Crime,” Criminology 30, no. 1 (February1992): 47–88.

99. Callie Harbin Burt et al., “A Longitudinal Test ofthe Effects of Parenting and the Stability of Self-Control: Negative Evidence for the General Theoryof Crime,” Criminology 44, no. 2 (May 2006):353–396; quotation is on p. 381.

100. John Paul Wright and Kevin M. Beaver, “Do Par-ents Matter in Creating Self-Control in Their Chil-dren? A Genetically Informed Test of Gottfredsonand Hirschi’s Theory of Low Self-Control,” Crimi-nology 43, no. 4 (November 2005): 1169–1202;quotation is on pp. 1190–1191.

101. See, for example, Terri E. Moffitt, “AdolescenceLimited and Life Course Persistent AntisocialBehavioral: A Developmental Taxonomy,” Psycho-logical Review 100 (1993): 674–701.

102. Elizabeth Cauffman et al., “Psychological, Neu-ropsychological and Psysiological Correlates of Seri-ous Antisocial Behavior in Adolescence: The Role ofSelf-Control,” Criminology 43, no. 1 (February2005): 133–176; quotation is on p. 159.

103. Ronald L. Simons et al., “Identifying the Psycholog-ical Factors That Mediate the Association BetweenParenting Practices and Delinquency,” Criminology45, no. 3 (August 2007): 481–518.

104. Kai T. Erikson, “Notes on the Sociology ofDeviance,” Social Problems 9 (Spring 1962): 308.

105. Ted Chiricos, Kelle Barrick, and William Bales,“The Labeling of Convicted Felons and Its Conse-quences for Recidivism,” Criminology 45, no. 3(August 2007): 547–581.

106. Frank Tannenbaum, Crime and the Community (NewYork: Columbia University Press, 1938), pp. 19–20.

107. Edwin M. Lemert, Human Deviance, Social Prob-lems, and Social Control (Englewood Cliffs, NJ:Prentice Hall, 1967), p. 17.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 183


108. Howard S. Becker, Outsiders: Studies in the Sociologyof Deviance (New York: Free Press, 1963).

109. For more details on the development of labelingtheory, see Walter R. Gove, ed., The Labeling ofDeviance: Evaluating a Perspective, 2d ed. (BeverlyHills, CA: Sage, 1980).

110. See Edwin M. Schur, Radical Nonintervention:Rethinking the Delinquency Problem (EnglewoodCliffs, NJ: Prentice Hall, 1973).

111. Stuart L. Hills, Crime, Power and Morality (Scran-ton, PA: Chandler, 1971), pp. 19–21.

112. Richard D. Schwartz and Jerome H. Skolnick, “TwoStudies of Legal Stigma,” Social Problems 10 (Fall1962): 136–141.

113. D. L. Rosenhan, “On Being Sane in Insane Places,”Science 179 (January 1973): 250.

114. Craig Haney et al., “Interpersonal Dynamics in aSimulated Prison,” International Journal of Crimi-nology and Penology 1 (1973): 69–97; reprinted inExamining Deviance Experimentally, ed. Darrell J.Steffensmeier and Robert M. Terry (Port Washing-ton, NY: Alfred, 1975), p. 223.

115. Melvin C. Ray and William R. Down, “An Empiri-cal Test of Labeling Theory Using LongitudinalData,” Journal of Research in Crime and Delinquency23 (May 1986): 169–194.

116. Malcolm W. Klein, “Labeling Theory and Delin-quency Policy,” Criminal Justice and Behavior 13(March 1986): 47–79.

117. Gordon Bazemore, “Delinquent Reform and theLabeling Perspective,” Criminal Justice and Behavior12 (June 1985): 131–169.

118. John Braithwaite, Crime, Shame, and Reintegration(Cambridge, UK: Cambridge University Press,1989), p. 100.

119. Ibid., pp. 100–101.

120. John Braithwaite, “Reintegrative Shaming,” inExplaining Criminals and Crime: Essays in Contem-porary Criminological Theory, ed. Raymond Pater-noster and Ronet Bachman (Los Angeles: Roxbury,2001), pp. 242–251; reference is on p. 244.

121. Braithwaite, Crime, Shame, and Reintegration, asreprinted in Criminological Theory, Past to Present,ed. Cullen and Agnew, pp. 277–285; quotation is onp. 278.

122. See Akers and Sellers, Criminological Theories, 4thed., p. 146, referring to the following: Toni Makkaiand John Braithwaite, “Reintegrative Shaming andCompliance with Regulatory Standards,” Criminol-ogy 32 (1994): 361–386; and Carter Hay, “AnExploratory Test of Braithwaite’s Reintegrative

Shaming Theory,” Journal of Research in Crime andDelinquency 38 (2001): 132–153.

123. Makkai and Braithwaite, “Reintegrative Shamingand Regulatory Compliance,” cited in John Braith-waite et al., “Reintegrative Shaming,” in The Essen-tial Criminology Reader, ed. Henry and Lanier, pp.286–295; quotation is on p. 292.

124. See, for example, John Braithwaite, Restorative Jus-tice and Responsive Regulation (New York: OxfordUniversity Press, 2002); Daniel Van Ness and KarenHeetderks Strong, Restoring Justice (Cincinnati, OH:Anderson, 1997); and Gordon Bazemore and MarkSchiff, eds., Restorative Community Justice: RepairingHarm and Transforming Communities (Cincinnati,OH: Anderson, 2001).

125. Frances T. Cullen, Jody L. Sundt, and John F. Woz-niak, “The Virtuous Prison: Toward a RestorativeRehabilitation,” in Contemporary Issues in Crimeand Criminal Justice: Essays in Honor of Gilbert Geis,ed. Henry N. Pontell and David Shichor (SaddleRiver, NJ: Prentice-Hall, 2001), pp. 265–286.


127. Clifford K. Dorne, Restorative Justice in the UnitedStates (Upper Saddle River, NJ: Prentice Hall,2008).

128. Ross L. Matsueda, “Reflected Appraisals, ParentalLabeling, and Delinquency,” reprinted in Crimino-logical Theory: Past to Present, ed. Francis T. Cullenand Robert Agnew (Los Angeles: Roxbury, 1999),pp. 279–285.

129. George Herbert Mead, Mind, Self, and Society(Chicago: University of Chicago Press, 1934).

130. Matsueda, “Reflected Appraisals,” p. 284.

131. Dawn J. Bartusch and Ross L. Matsueda, “Gender,Reflected Appraisals, and Labeling: A Cross-GroupTest of an Interactionist Theory of Delinquency,”Social Forces 75 (1996): 145–177.

132. Robert J. Sampson and John H. Laub, “A LifeCourse Theory of Cumulative Disadvantage and theStability of Delinquency,” in Advances in Crimino-logical Theory: Developmental Theories of Crime andDelinquency, Vol. 7, ed. Terence P. Thornberry(New Brunswick, NJ: Transaction, 1997), pp.133–161; quotation is on p. 138.

133. Ibid., pp. 147–148.

134. Jon Gunnar Bernburg and Marvin D. Krohn,“Labeling, Life Chances, and Adult Crime: TheDirect and Indirect Effects of Official Interventionin Adolescence on Crime in Early Adulthood,”Criminology 41, no. 4 (November 2003):1287–1318; quotation is on p. 1313.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 184


135. Daniel L. Dotter and Julian B. Roebuck, “TheLabeling Approach Re-Examined: Interactionismand the Components of Deviance,” Deviant Behav-ior 9, no. 1 (1988): 19.

136. Akers and Sellers, Criminological Theories, 4th ed.,pp. 155, 156, 157.

137. Charles R. Tittle, “Labelling and Crime: An Empiri-cal Evaluation,” in The Labelling of Deviance, ed.Walter R. Gove (New York: Halsted Press, 1975),Chap. 6, pp. 157–179, quotation is on p. 158, cita-tions omitted.

138. Ibid., p. 176.

139. See Bruce G. Link et al., “A Modified Labeling The-ory Approach to Mental Disorders: An EmpiricalAssessment,” American Sociological Review 54 (June1989): 400–423; and Bruce G. Link et al., “TheSocial Rejection of Former Mental Patients: Under-standing Why Labels Matter,” American Journal ofSociology 6 (May 1987): 1461–1500.

140. J. Mitchell Miller, Christopher J. Schreck, andRichard Tewksbury, Criminological Theory: A BriefIntroduction, 2nd ed. (New York: Pearson Educa-tion, 2008), pp. 194–202.

141. Ibid., pp. 202–204.

142. See Ronald L. Akers, Deviant Behavior: A SocialLearning Approach (Belmont, CA: Wadsworth,1973; 2nd ed., 1977).

143. Akers and Sellers, Criminological Theories, 4th ed.,p. 270. See also Akers and John K. Cochran, “Ado-lescent Marijuana Use: A Test of Three Theories ofDeviant Behavior,” Deviant Behavior 6 (1985):323–346.

144. Aker and Sellers, ibid., p. 271.

145. David P. Farrington, “Developmental and Life-Course Criminology: Key Theoretical and EmpiricalIssues—The 2002 Sutherland Award Address,”Criminology 41, no. 2 (May 2003): 221–255, quota-tions are on pp. 223–224.

146. See Akers and Sellers, Criminological Theories, 4thed., p. 284.

147. Michael L. Benson, Crime and the Life Course: AnIntroduction (Los Angeles: Roxbury, 2002); quota-tion is on p. 14.

148. Laub, “The Life Course of Criminology in theUnited States,” pp. 3–4.

149. Robert J. Sampson and John H. Laub, “An Age-Graded Theory of Informal Social Control,” inCriminological Theory, 3rd ed., ed. Cullen andAgnew, pp. 241–253; quotation is on p. 244. Seealso Sampson and Laub, Crime in the Making: Path-ways and Turning Points Through Life (Cambridge,MA: Harvard University Press, 1993).

150. See the discussion by Cullen and Agnew, eds., inCriminological Theory, 3rd ed., pp. 241–242.

151. Sampson and Laub, “An Age-Graded Theory ofInformal Social Control,” p. 243.

152. Ibid.

153. Ibid.

154. Mark Warr, “Life Course Transitions and Desis-tance from Crime,” Criminology 36, no. 1 (February1998): 183–216; quotation is on p. 184.

155. Ibid.

156. Ronald L. Simons et al., “A Test of Life-CourseExplanations for Stability and Change in AntisocialBehavior from Adolescence to Young Adulthood,”Criminology 40, no. 2 (May 2002): 401–434; quota-tion is on p. 429.

157. Robert J. Sampson and John H. Laub, “Life-CourseDesisters? Trajectories of Crime Among DelinquentBoys Followed to Age 70,” Criminology 41, no. 3(August 2003): 555–592; quotations are on p. 555.

158. Robert J. Sampson, John H. Laub, and ChristopherWimer, “Does Marriage Reduce Crime? A Counter-factual Approach to Within-Individual CausalEffects,” Criminology 44, no. 3 (August 2006):465–508.

159. Ryan D. King et al., “The Context of Marriage andCrime: Gender, the Propensity to Marry, andOffending in Early Adulthood,” Criminology 45, no. 1 (February 2007): 33–66.

160. John Paul Wright and Francis T. Cullen, “Employ-ment, Peers, and Life-Course Transitions,” JusticeQuarterly 21, no. 1 (March 2004): 183–201; refer-ence is on p. 200.

161. Ibid., pp. 184, 197.

162. Terri E. Moffitt, “Adolescence Limited and LifeCourse Persistent Antisocial Behavioral: A Develop-mental Taxonomy,” Psychological Review 100(1993): 674–701. See also Moffitt et al., Sex Differ-ences in Antisocial Behaviour: Conduct Disorder,Delinquency, and Violence in the Dunedin Longitudi-nal Study (Cambridge: Cambridge University Press,2001). For a test of Moffitt’s theory, see Alex R.Piquero and Timothy Brezina, “Testing Moffitt’sAccount of Adolescence-Limited Delinquency,”Criminology 39, no. 2 (May 2001): 353–370.

163. Miller, Schreck, and Tewksbury, CriminologicalTheory: A Brief Introduction, p. 79.

164. Francis T. Cullen, “Social Support as an OrganizingConcept for Criminology: Presidential Address tothe Academy of Criminal Justice Sciences,” JusticeQuarterly 11 (1994): 528–559.

165. Cullen and Agnew, ed., Criminological Theory, p. 582.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 185


166. Mark Colvin, Francis T. Cullen, and Thomas Van-der Ven, “Coercion, Social Support, and Crime: AnEmerging Theoretical Consensus,” Criminology 40,no. 1 (February 2002): 19–42; quotation is on pp.19 and 20.

167. Mark Colvin, Crime and Coercion: An IntegratedTheory of Chronic Criminality (New York: St.Martin’s Press, 2000).

168. Colvin, Cullen, and Vander Ven, “Coercion, SocialSupport, and Crime,” p. 37.

169. Terence P. Thornberry, “Toward an InteractionalTheory of Delinquency,” Criminology 25 (1987):863–891.

170. Terence P. Thornberry and Marvin D. Krohn, “TheDevelopment of Delinquency: An Interactional Per-spective,” in Handbook of Youth and Justice, ed.Susan O. White (New York: Plenum, 2001). Seealso Thornberry and Krohn, Taking Stock of Delin-quency: An Overview of Findings from ContemporaryLongitudinal Studies (New York: Kluwer/Plenum,2003).

171. Rachel A. Gordon et al., “Antisocial Behavior andYouth Gang Membership: Selection and Socializa-tion,” Criminology 42, no. 1 (February 2004):55–87; quotations are on p. 56.

172. Cullen and Agnew, Criminological Theory, 3rd ed.,p. 551. See also Terence P. Thornberry and MarvinD. Krohn, “Applying Interactional Theory to theExplanation of Continuity and Change in AntisocialBehavior,” in Integrated Developmental & Life-Course Theories of Offending, ed. David P. Farring-ton (New Brunwsick, NJ: Transaction, 2005), pp.183–210.

173. Marvin D. Krohn, “The Web of Conformity: ANetwork Approach to the Explanation of Delin-quent Behavior,” Social Problems 33 (1986):S81–S93, as cited in Akers and Sellers, Criminologi-cal Theories, 4th ed., pp. 276–277.

174. Dana L. Haynie, “Friendship Networks and Delin-quency: The Relative Nature of Peer Delinquency,”Journal of Quantitative Criminology 18 (2002):99–134; quotation is on p. 104.

175. Howard B. Kaplan, Self Attitudes and DeviantBehavior (Pacific Palisades, CA: Goodyear, 1975);Kaplan, “Drugs, Crime, and Other Deviant Adapta-tions,” in Drugs, Crime and Other Deviant Adapta-tions, ed. Kaplan (New York: Plenum, 1995), pp.3–46; and Kaplan, “Empirical Validation of theApplicability of an Integrative Theory of DeviantBehavior to the Study of Drug Use,” Journal ofDrug Issues 26 (1996): 345–377.

176. Charles R. Tittle, “Control Balance Theory,” inCriminological Theory, 3rd ed., ed. Cullen andAgnew, pp. 563–565. The quotation is from Tittle,Control Balance: Toward a General Theory ofDeviance (Boulder, CO: Westview Press, 1995), p.142.

177. Charles R. Tittle, “Redefining Control Balance The-ory,” Theoretical Criminology 8 (2004): 395–428,with references to pp. 405, 406. For a discussion,see Cullen and Agnew, Criminological Theory, 3rded., pp. 563, 564.

178. Vold et al., Theoretical Criminology, p. 311.

179. Delbert S. Elliott, Suzanne S. Ageton, and RachelleJ. Canter, “An Integrated Theoretical Perspective onDelinquent Behavior,” Journal of Research in Crimeand Delinquency (1979): 3–27, as reprinted inCullen and Agnew, eds., Criminological Theory,pp. 537–550. See p. 548.

180. Ibid.

181. Cullen and Agnew, eds., Criminological Theory,p. 537.

182. Carlo Morselli et al., “Mentors and CriminalAchievement,” Criminology 44, no. 1 (February2006): 17–43.

183. Matthew Robinson, Why Crime? An Integrated Sys-tems Theory of Antisocial Behavior (Upper SaddleRiver, NJ: Prentice-Hall, 2004).

184. Matthew Robinson, “The Integrated Systems The-ory of Antisocial Behavior,” in The Essential Crimi-nology Reader, ed. Henry and Lanier, pp. 319–335;quotation is on p. 326.

185. Ibid., p. 333.

186. Stuart Henry and Mark M. Lanier, Essential Crimi-nology (Boulder, CO: Westview Press, 2004), p. 351.

187. Anthony Walsh, book review, Human NatureReview 5 (31 December 2005): 97–94; quotation ison p. 93, http://www.human-nature.com/nibbs/05/awalsh.html, accessed 5 January 2008.

188. The film can be viewed online at http://www.dr.dk/DR2/VidenOm/Videnskab%20der%20udfordrer/Programmerne/20070615142459.html, accessed 5 January 2008.

189. Robert Agnew, Why Do Criminals Offend? A Gen-eral Theory of Crime and Delinquency (Los Angeles:Roxbury, 2005), p. 208.

190. Ibid., p. 209.

191. Ibid., p. 210.

192. Ibid., pp. 210–213.

193. Ibid., p.214.

0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 186


194. Farrington, “Developmental and Life-Course Crim-inology,” pp. 223, 231.

195. Ibid., pp. 231, 232.

196. Ibid., p. 233.

197. Ibid.

198. Ibid.

199. Ibid., pp. 247–248.

200. Travis Hirschi, “Separate but Unequal Is Better,”Journal of Research in Crime and Delinquency 16(1979): 34–38; and Hirschi, “Exploring Alternativesto Integrated Theory,” in Theoretical Integration inthe Study of Deviance and Crime, by Steven F.Messner, Marvin D. Krohn, and Allen E. Liska(Albany: SUNY Press, 1989), pp. 37–49, both cited in Vold et al., Theoretical Criminology,5th ed., pp. 304, 305.


202. Miller, Schreck, and Tewksbury, CriminologicalTheory, p. 205.

203. See Thomas J. Bernard, “Twenty Years of TestingTheories: What Have We Learned and Why?”Journal of Research in Crime and Delinquency27 (November 1990): 330–347.

204. Roland Chilton, “Urban Crime Trends and Crimi-nological Theory,” Criminal Justice Research Bulletin6, no. 3 (Huntsville, TX: Sam Houston State Uni-versity Criminal Justice Center, 1991), p. 2.

205. Vold et al., Criminological Theories, 5th ed., p. 320.

206. Ibid., pp. 325–326.

207. Ibid., p. 326.

208. George A. Lundberg, Can Science Save Us? (NewYork: Longman, 1961), pp. 143–144.

209. Robert B. Bursik Jr., Editorial Statement, Criminol-ogy 39, no. 1 (February 2001): n.p.

210. James Austin, “Why Criminology Is Irrelevant,”Criminology & Public Policy 2, no. 3 (July 2003):557–564; quotation is on p. 557.

211. Roland Chilton, “Viable Policy: The Impact ofFederal Funding and the Need for IndependentResearch Agendas—The American Society of Crim-inology 2000 Presidential Address,” Criminology 39(February 2001): 1.

212. Ibid., p. 7.

213. Denise C. Gottfredson, Stacy S. Najaka, and BrookKearley, “Effectiveness of Drug Treatment Courts:Evidence from a Randomized Trial,” Criminology &Public Policy 2, no. 2 (March 2003): 171–196; quo-tations are on pp. 173 and 190.

214. John S. Goldkamp, Reaction Essay: “The Impact ofDrug Courts,” Criminology & Public Policy 2, no. 2(March 2003): 197–206; quotation is on p. 203.

215. Robert C. Davis, Barbara E. Smith, and Bruce Tay-lor, “Increasing the Proportion of Domestic Vio-lence Arrests That Are Prosecuted: A NaturalExperiment in Milwaukee,” Criminology & PublicPolicy 2, no. 2 (March 2003): 263–282; quotation ison p. 280.

216. Laura Dugan, “Domestic Violence Legislation:Exploring Its Impact on the Likelihood of DomesticViolence, Police Involvement, and Arrest,” Crimi-nology & Public Policy 2, no. 2 (March 2003):283–312; quotation is on p. 283.


0195370447_0142-0187_CH_06.qxp 6/18/08 10:20 AM Page 187

Sociological Theories

Documents

Transcript of Sociological Theories