Social Learning and Social Structure

A General Theory of Crime and Deviance

Ronald Akers

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781351490146
• Edition: 1
• Topic: Social Sciences
• Subtopic: Criminology
• Index: Social Sciences

The social learning theory of crime integrates Edwin H. Sutherland's diff erential association theory with behavioral learning theory. It is a widely accepted and applied approaches to criminal and deviant behavior. However, it is also widely misinterpreted, misstated, and misapplied.This is the fi rst single volume, in-depth, authoritative discussion of the background, concepts, development, modifications, and empirical tests of social learning theory. Akers begins with a personal account of Sutherland's involvement in criminology and the origins of his infl uential perspective. He then traces the intellectual history of Sutherland's theory as well as social learning theory, providing a comprehensive explanation of how each theory approaches illegal behavior. Akers reviews research on various correlates and predictors of crime and delinquency that may be used as operational measures of differential association, reinforcement, and other social learning concepts.Akers proposes a new, integrated theory of social learning and social structure that links group diff erences in crime to individual conduct. He concludes with a cogent discussion of the implications of social learning theory for criminology and public policy. Now available in paperback, with a new introduction by the author, this volume will be invaluable to professionals and for use in courses in criminology and deviance.

1
A Personal History of the Development of Social Learning Theory

In carrying out the purposes stated in the Introduction, I show the origin, development, and current status of the social learning theory of crime and deviance. This portrayal revolves around intellectual, theoretical, conceptual, and empirical issues. Such an approach risks giving the erroneous impression that the progress of the theory has been a straight line, guided only by scientific considerations and the next logical step needed for theoretical elaboration/specification or empirical testing. For a true picture, however, one must also paint in scenes of how the progress and problems of the theory have been affected by the vagaries of an academic career, opportunities, personal happenstance, and just being in the right (or wrong) place at the right time. Therefore, I begin with a brief history of my own involvement with the theory to provide some supplemental information to the reader who is interested in this dimension of its development and to flesh out its presentation.

I am unpersuaded by the extreme sociology-of-knowledge argument that all ideals, ideology, theory and knowledge are simply reflections of the locations of their authors and promoters in a particular sociocultural context. I do believe, however, that personal history being in a particular place at a particular time, does have some impact on a scholar's work. Certainly one's own ideas are strongly influenced by interaction with colleagues and students. This is one of the main reasons why books on theory in the social sciences often include biographical material on the authors. One's personal history, values, beliefs, and interests may influence one to pay attention to one problem and not another, to be intrigued or persuaded by one line of argumentation and not another. They may introduce some element of "bias," and to know "where someone is coming from" aids in assessing her or his theoretical and value judgements, The old adage that "where one stands depends on where he sits" has some truth to it. Nevertheless, the validity of our ideas cannot be judged on that basis; the truth lies in the content of what is said, not in who says it. But in recounting the intellectual heritage of the theory it is informative to relate something of what Williams and McShane (1988) refer to as its "social heritage" as well.

Background

I can trace my introduction to Sutherland's differential association theory back to my undergraduate days (1956-60) at Indiana State University [known originally as Indiana State Normal School, as Indiana State Teacher's College when I enrolled, and simply as Indiana State College by the time I was graduated). At that time the primary majors available were secondary and elementary education, and most ISU students began careers in public school teaching immediately following graduation. I entered Indiana State with this as my goal, and until two months before graduation my intention was to teach in high school. Indeed, in the final quarter of my senior year I had a contract offer from the Vigo County school superintendent to teach social studies and English and to coach the cross-country team in a Terre Haute high school. At the last minute, I turned it down and decided to enter graduate studies in sociology.

Social studies was one of the "subject matter" specializations allowed within my secondary-education major, which included history, psychology, anthropology, sociology, and economics. There was no course in sociological theory or methods at the time, but several worthwhile sociology courses were offered as electives. I was very much attracted to sociology and anthropology (there was much less distinction between the two in those days than there is now) and took almost all of my electives in them. The courses included introductory sociology, social problems, sociology of education, juvenile delinquency, family, stratification, public opinion, and cultural anthropology.

It was in one of these elective courses (juvenile delinquency) that I was introduced to Sutherland's differential association theory. There was, as I recall, little theoretical content to the course, but the text covered, and the instructor devoted time to, "causes" of delinquency. Some attention was paid to biological, psychological, and sociological causes. Sutherland's theory, which the instructor interpreted to mean "differences in associations" with bad companions, was covered in this context. This introduction did not really convey the significance of the theory in the history of criminological thought; there was, instead, more emphasis on the concentration of delinquency in the lower class, on Merton's theory and Cohen's then relatively new book on gang delinquency, and on family factors in delinquency. For a term paper on social class and delinquency I read F. Ivan Nye's Family Relationships and Delinquent Behavior (1958).

In retrospect, I must admit that the theoretical differences and similarities between the control theory Nye presented in his book, the anomie theories of Merton and Cohen, and Sutherland's differential association theory pretty much escaped me. I thought the key sociological issue was lower-class status as a cause of delinquency. I spent the first few years of my life in a lower-class mixed-race neighborhood, and until completing high school I frequently returned to run around with my cousins and friends there. Although my upbringing included little in the way of "class consciousness," it made sense to me to hear that social class is important in understanding a whole range of social interactions, attitudes, and problems. It seemed pretty much settled that crime and delinquency was concentrated in lower-class neighborhoods. Thus, I was surprised and intrigued by the finding of little class difference in self-reported delinquency (measured by a technique developed by James Short in collaboration with Nye), as opposed to strong class differences in official delinquency.

That interest stayed with me into graduate school. After I took a seminar in criminology, it motivated me to do master's thesis research in 1961 replicating the part of the Short and Nye study dealing with social class and delinquency; my first published article was based on that thesis (Akers, 1964). By then, three other self-report studies had been published (Dentler and Monroe, 1961; Reiss and Rhodes, 1961; Clark and Wenninger, 1962). In the next decade, the self-report technique mushroomed to become the standard, most frequently used methodology in all studies of delinquency. For most of that time, virtually every published self-reported delinquency study noted my article in the string of citations to Short, Nye, Clark and Wenninger, and others. By 1972 it had become one of the most cited pieces in the criminological literature published since 1945 (Wolfgang et al., 1978).¹ In the thesis and article, I had no theoretical framework other than the hypothesis that delinquent behavior is related to social class. The only theoretical implications I drew were that the findings raised questions about theories that focused on the concentration of delinquency in the lower class.

However, I became thoroughly familiar with differential association theory and grew to understand its significance as the leading general criminological theory. My knowledge and interpretation of the theory was strongly influenced in graduate study (1960-61) with the late Oscar Ritchie at Kent State University, who directed the thesis, and in doctoral studies (1962-65) with Richard Quinney at the University of Kentucky, who directed the dissertation. At that time, there was great excitement among faculty and graduate students in the sociology of crime and deviance about the newly developing theories of the late 1950s and early 1960s. Although differential association theory received its due, most of the excitement was generated by Albert Cohen's and Richard Cloward and Lloyd Ohlin's theories of delinquent subculture, Void's conflict theory, the conflict approach that Quinney was developing, and labeling theory (earlier introduced by Edwin Lemert and made highly visible when I was in graduate school by the publications of Howard Becker). I found all of these to be highly attractive intellectually, but I became more and more persuaded that Sutherland's theory made more sense as an explanation of criminal behavior than any of the others. However, I did nothing special with the theory in graduate school.

My major doctoral specialty within sociology was criminology and sociology of law (with an out-of-department minor in social anthropology). My dissertation was a study of political power and enactment of professional and licensure laws. I had come to this problem through studies in criminology and an interest in criminal and deviant behavior. What began as a study of white-collar crime among health professionals turned into a study in the sociology of law. The theoretical orientation of the dissertation was taken from Quinney's conflict theory of law. It included nothing on criminal behavior or differential association.

Development of the Theory

Robert Burgess and I arrived at the University of Washington in Seattle in 1965. We were untenured, acting assistant professors with unfinished dissertations who were assigned adjoining offices in the Department of Sociology. Burgess was one of the first behavioral sociologists specifically trained in psychological operant conditioning theory and research under Robert Hamblin at Washington University in St. Louis. Later, another behaviorally oriented sociologist from that program, David Schmitt, joined the UW faculty. Burgess's graduate training included intensive study of behavioral learning theory At Washington University, he had access to the theories and programs of two of the top applied behavioral psychologists of the day, Nathan Azrin and Theodore Ayllon, who were developing behavior modification programs (in fact, token economies) at a nearby state mental institution in southern Illinois.

Burgess's specialty was social psychology, and his dissertation was an experimental test of an operant model of organizational communication patterns. He had also worked on a delinquency prevention project as a graduate student and had acquired some knowledge of differential association theory. I was quite familiar with the theory, but had acquired only a little knowledge of psychological behaviorism, mainly by reading George Homans while in graduate school. I was not overly impressed with Homans and did not see that his approach had much to offer for understanding crime and deviance. Burgess was already persuaded that all of sociology would benefit from the infusion of psychological behaviorism.

Because of our spatial, intellectual, organizational, and career proximity, it did not take long for the two of us to become friends and begin carrying on lengthy conversations about a wide range of mutual concerns, The most intensive of our talks had to do with the relevance of psychological behaviorism to sociology and criminology, especially the connections between Sutherland's learning theory formulated in the 1940s and operant psychology which had begun at about that time but did not come into prominence until the 1950s. Were it not for this fortuitous juxtaposition, I believe, neither of us would have become involved in a behavioristic reformulation of Sutherland's theory. Moreover (as will become apparent below), there was a unique combination of faculty and graduate students at the University of Washington who also played a role in my early focus on theoretical issues in criminology.

Although I was drawn to Sutherland, during the early years at the University of Washington I was involved in the same kind of spirited discourse over the new theories of the late 1950s and early 1960s that I had encountered in graduate school. I became steeped in the subcultural and gang theories and studies of Cohen, Cloward and Ohlin, Short, and their colleagues; I knew the control theories of Nye, Reckless, and others. My mentor, Quinney, had work on differential association theory under way with Melvin DeFleur, an effort that they had begun while I was in graduate school. There were other exciting theoretical developments brewing at the time. I knew directly of Quinney's and Turk's style of conflict theory and incorporated it into my dissertation (which I completed during my first year on the UW faculty). I had studied Becker's widely heralded work on labeling theory. William Chambliss and I became good friends through team-teaching the social problems course and through our mutual interest in criminology. Chambliss was developing his conflict approach and doing some extremely interesting research, ä la Sutherland, with Harry King, a professional thief, whom I came to know and with whom I continued to work after Chambliss left UW. Conflict theory was challenging functionalism in sociology. The whole intellectual atmosphere of the time was to question or revise existing theoretical perspectives and to develop new approaches. The political conflicts over the Vietnam War were still nascent, but there was already a distinct antiestablishment, underdog orientation in sociology.

One major contention, promoted by both conflict and labeling theorists but also widely endorsed by others, involved a shift in emphasis. They argued that the field should move from a focus on explaining the behavior of "ordinary" criminals and the lower class toward a focus on the establishment. Such a change would involve not only accounting for social control, law, and the criminal justice system but also studying white-collar and corporate crime (taking up a theme developed earlier by Sutherland). Chambliss persuasively advocated (and still advocates) this position. This was my inclination as well, and that was reflected in the dissertation project, in my articles on the political power of professions, and in teaching courses in criminology and sociology of law.

I had no intention of developing etiological theory or working specifically on differential association theory; my goals were to complete the dissertation, teach classes, and get enough published from the dissertation and other research to pass muster at tenure and promotion time. It was not until the interaction with Robert Burgess that I began to think seriously about refocusing on the question of criminal etiology and a general theory of criminal behavior. Burgess was very interested in promoting the behavioral perspective in sociology, and we had common career concerns—a desire to make an impact in sociology and to make tenure. Richard Emerson, who also was on the UW sociology faculty at the time, had made a major contribution to the field with his power-dependence theory (Emerson, 1962) and was later active in the group of social psychologists developing social exchange theory in sociology (Emerson, 1972a; 1972b). Burgess and I spent a good deal of time in discussion with him about behavioral sociology and how operant theory fits with sociological theory.

My friend had no previous inclination to develop his behaviorist interest with differential association theory and had little interest in criminology as a field of study. I was not inclined toward studying operant psychology. Once we perceived our common interests, however, we began collaborating. Burgess guided my reading and study of operant psychology and behavior modification, and I helped him with differential association theory and criminology. Although he was interested in social interaction and was a well-trained sociologist, Burgess tended to take a fairly strict Skinnerian approach and to agree with Homans that there is virtually no sociological principle or generalization that cannot be reduced to the basic behavioral principles of operant conditioning theory. I agreed that the operant principles were a very powerful explanation of behavior that, by attributing primary causes to the environment, fit sociological theory better than explanations of behavior that gave primacy to the internal emotional or biological states of the individual. But I leaned toward retaining cognitive concepts from the symbolic interactionist tradition and Sutherland's emphasis on interaction in primary groups. It seemed to me that the concept of "environment" in operant conditioning theory was fairly artificial, modeled on Skinner's apparatus for experiments with pigeons and other animals. It would mesh with sociology only if it were conceptualized so as to fit the naturally occurring social environments (studied with observational and survey techniques) with which sociologists were mainly concerned. In addition, I thought the "successive approximation" process in Skinnerian theory—involving gradual, incremental "shaping" in which all "behavior is constructed by a continual process of differential reinforcement from undifferentiated behavior" (Skinner, 1953:92) without assuming any cognitive processes— was too cumbersome and undirected to apply to most social behavior. This leaning toward relying on cognitive processes pointed me in the same theoretical direction in which Bandura and other behavioral psychologists were moving.

Burgess and I agreed that Sutherland's theory was the most fruitful place to explore the ramifications of behavioral psychology for sociology beyond what Homans had done with regard to small groups. Both Sutherland's view and operant conditioning were self-consciously "learning" theories, involving behavioral acquisition, maintenance, performance, and change. We were fully aware that the disciplinary, theoretical, and methodological traditions of the two differed dramatically. We knew from our discussions (sometimes heated) with colleagues in the department that there was enormous skepticism among sociologists about the value and relevance of behaviorism to sociology

Behavioral psychology was (and to a great extent still is) viewed in sociology as a mechanistic theory that treats humans as "hollow" automatons whose actions are the result of passive and mindless conditioning by environmental stimuli in exactly the way animals are conditioned. Ulis was in contrast to symbolic interactionism theory which stresses the thinking and reflexive individual interacting in society. Many of our colleagues were of the opinion that the two approaches were so incompatible that no integration of any kind was possible; indeed, it was fruitless to try. Burgess and I so violated the conventional wisdom that our efforts to blend Sutherland and behaviorism were later branded as "theoretically illiterate" (Taylor et al., 1973). Behaviorism was also viewed as hopelessly tautological. Moreover, some of our colleagues and students saw it as somehow bent to serving the establishment and maintained that greater knowledge of the behavioral principles would only lead to more effective, coldly efficient techniques of behavior and mind control by the power elite.

Burgess and I disagreed with this. We argued that knowledge of behavior can be put to use as readily in the cause of dissent and change as in the cause of maintaining the status quo. We pointed to Skinner's arguments in Walden Two that behavioral principles could be used to model a more humane and free society. It seemed unequivocal to us that operant principles contained the precise mechanisms of learning that Cressey had said were needed for differential association theory. While these principles emphasized the functional relationship between behavior and observable environmental consequences, they allowed for human awareness and placed the person at the center as acting upon the environment. It was not simply a theory of reflex conditioning. We argued that, despite being directed toward individuals and small groups rather than institutions or whole societies, it had the same formal characteristics as functionalist theory in sociology. That is, both explained the form of the thing to be explained (behavior in the case of operant theory, institutional structure in the case of functionalism) by its consequences.

We also argued that Sutherland's fundamental idea of learning through association or interaction with others in a social environment was entirely compatible with the basic notion in operant theory that the individual's behavior is shaped by interaction with the environment, whether social or nonsocial. Sutherland's conceptions of differential association and the balance of definitions favorable and unfavorable to crime is analogous to the Skinnerian concept of differential reinforcement as the balance of reward and punishment. Therefore, we began to explore the extent to which the symbolic interactionist concepts in the sociological approach to learning and socialization incorporated into Sutherland's theory could be compatible with the radical behaviorism of operant conditioning theory. In the beginning we leaned more heavily toward the strict behavioral conditioning components of orthodox operant theory and downplayed the symbolic interactionist components from Sutherland's theory. As I have mentioned, the more I developed the theory, the greater the role I gave to direct cognitive processes more in line with symbolic interactionism in sociology and the social behaviorism of Bandura and others in psychology.

As Burgess and I were working on this task during the fall of 1965, C. Ray Jeffery's (1965) article "Criminal Behavior and Learning Theory" was published in the Journal of Criminal Law, Criminology, and Police Science. I brought it to my friend's attention as an example of an effort by a criminologist who was moving in the same direction we were. The more closely we examined the piece, however, the more it became apparent that Jeffery did not have the same goal t...

Table of contents

• Cover
• Half Title
• Title
• Copyright
• Dedication
• Contents
• List of Tables
• List of Figures
• Acknowledgments
• Introduction
• Introduction to the Transaction Paperback Edition
• 1 A Personal History of the Development of Social Learning Theory
• 2 Development and Revisions of Differential Association Theory
• 3 The Social Learning Theory of Criminal and Deviant Behavior
• 4 Social Learning and Cultural Deviance Theory
• 5 Research on Social Learning Variables in Crime and Delinquency
• 6 Social Learning Variables in Delinquency Prediction
• 7 Adolescent Drug and Drinking Behavior: The Boys Town Study
• 8 Adolescent Smoking Behavior: The Iowa Study
• 9 Elderly Drinking Behavior: The Florida Study
• 10 Rape and Sexual Aggression
• 11 Other Studies on Social Learning
• 12 Social Structure and Social Learning in Crime and Deviance
• Bibliography
• Index