Humans are a gregarious species, and the notion of peer influence is neither difficult to understand nor far removed from everyday experience. Few readers of this book could genuinely claim that their preferences and practices with respect to music, dress, politics, entertainment, or religion were acquired solely from their parents or developed wholly in isolation. Nevertheless, one investigator (Reiss, 1986) was entirely correct when he described the nature of peer influence as “murky.”

When it comes to understanding the role of peer influence in delinquent behavior, the principal difficulty confronting investigators lies in the sheer number and manner of ways in which peer influence may operate to encourage criminal conduct. Most criminologists who invoke the words “peer influence” probably have in mind several mechanisms of social influence, and those conceptions often vary considerably from one investigator to the next. My objective in this chapter is to sur vey the possibilities concerning the nature of peer influence, drawing where possible on existing evidence from the social sciences to inform the discussion and establish a case for each explanation.

The aim, however, is not to champion one explanation over another. Testing and adjudicating among these explanations is likely to require years of concerted effort from social scientists, and it would be premature at this point to select one theory over another. Rejecting a correct theory is no less serious than affirming a false one, and under the present state of evidence the risk of both errors is substantial.

This book is a mixture of the old and the new, of ideas that took root in me as a young man and those that have steered my research in more recent years. It is a book that brings together questions that intrigued criminologists decades ago with ones that hold center stage today in the field of criminology.

As a graduate student at the University of Arizona in the middle and late 1970s, I was privileged to work with a group of faculty members that included Maynard Erickson, Jack Gibbs, and Gary Jensen. Their joint research concentrated on the deterrent effect of legal sanctions, but unlike many others working on the issue at the time, they conceived of the topic in a broad sociological sense. How much does the public know about the criminal sanctions prescribed by law, and how can sanctions deter those who are unaware of them? How does one measure the severity of a legal punishment, and how does the objective severity of a punishment (say, years spent in prison) correspond to its perceived severity? Do public perceptions of crime (e.g., the relative incidence of different offenses) and of punishment (e.g., the conditional probability of arrest) accord with reality? Only later was I to realize how truly original and prescient much of this work was. But the mutual interest of these investigators in the social psychology of crime and punishment was to stimulate one persisting line of research in my own career.

This book has been an effort to organize and evaluate the evidence concerning peer influence and criminal behavior. That delinquent behavior is predominantly group behavior is beyond dispute. That having delinquent friends is associated with delinquent behavior is equally indisputable. That these facts have any etiological significance when it comes to delinquent behavior, however, is disputable.

Some criminologists, as we have seen, dismiss these facts with scarcely a pause, either because they appear to be inconsequential, contradict a preferred theory, or can be readily explained away. In one of the more influential books in criminology in recent years, for example, Gottfredson and Hirschi (1990) utterly rejected the notion of peer influence and instead attributed criminal behavior to an absence of self-control. Gottfredson and Hirschi's distaste for peer explanations seems odd, however, when one considers that one of the principal ways by which groups seem to affect individuals is precisely by dissolving their self-control (see Chapter 4). The authors also make much of the similarities between criminal events and accidents – both ostensible consequences of low self-control. Yet among adolescents, the probability of a fatal motor vehicle accident increases in direct proportion to the number of adolescents in the vehicle (Chen et al., 2000).

Like a gorilla in the living room, the social nature of crime seems to push some investigators to extraordinary lengths in their efforts to ignore it.

The term “peer” properly refers to a social equal (Hartup, 1983), but it is conventionally used by social scientists to refer to associates of the same age. In the course of daily living, individuals normally interact with persons from all stages of life, but it is among age-mates that the most intimate of human relationships – deep and abiding friendships and romantic relationships – are often formed.

Sociologists and developmental psychologists are virtually unanimous in agreeing that peer relations play a critical role during the transition from childhood to adulthood, when individuals must cast off their childhood identity and enter the larger world of adults:

Because peers take on heightened importance during the teenage years, and because criminal behavior peaks at these ages as well, it is natural to wonder whether the two phenomena are somehow linked. Before we can address that question, however, it is important to look more closely at the role of peers in human development.

On a warm, clear Sunday morning, I walked out onto the deck at the rear of my house to read the morning newspaper. The house sits atop a high, steep hill that looks down upon a nearby street lined with homes on either side. As I sat engrossed in the events of the day, the air was suddenly punctuated by the sound of loud and garrulous voices wending up from below. Glancing down, I saw four young males walking together up the steep slope of the street. Judging from their size and dress, they appeared to be about fourteen or fifteen years of age, though it was difficult to estimate their age with any certainty from the distance at which I was observing.

As they ambled up the street, these young men turned abruptly to one side of the road and climbed up onto the foundation of an expensive new home under construction near the edge of a narrow, precipitous gorge. Trees surrounding the construction site allowed only fleeting glimpses of the boys, but the nature of their activity was evident from the sounds emanating from the site. The crack of splintering wood, the explosive impacts of bricks and other materials being hurled down the hill, the sound of concrete bags tearing open, all testified to the damage these boys were inflicting on the home, damage that probably amounted to many thousands of dollars.

The preceding chapter described a variety of possible mechanisms of peer influence and offered some grounds for treating peers as an important element in the etiology of delinquency. This chapter extends the case for peer influence further by demonstrating how peer variables can help to explain some of the most fundamental features of delinquent behavior. It also contrasts the strength of peer influence with another very powerful source of influence in the lives of adolescents – the family.

AGE AND CRIME

One of the most indisputable features of criminal behavior is its age distribution. Though a seemingly mundane phenomenon, the age distribution of criminal behavior is intriguing enough to have caught the attention of Quetelet, the Gluecks, Sutherland, and, in more recent times, Hirschi and Gottfredson, Sampson, Blumstein, Farrington, and others.

What so arouses the attention of criminologists is the lawlike relation between age and criminal conduct. According to all major methodologies for measuring crime (self-reports, official data, and victimization data), age-specific rates of offending in the general population peak in middle to late adolescence for most offenses, and drop sharply and permanently thereafter (Hirschi and Gottfredson, 1983; Wilson and Herrnstein, 1985; Farrington, 1986; Blumstein and Cohen, 1987; Wolfgang, Thornberry, and Figlio, 1987; Steffensmeier et al., 1989; Moffitt, 1993). Drug offenses are one of the few exceptions to this rule, reaching a peak at later – but still early – ages (Bachman et al., 1984; Kandel and Yamaguchi, 1987; Akers, 1992).

The preceding chapter demonstrated the importance of peers in the lives of most adolescents, and the next chapter will examine how peer influence can exert itself when it comes to delinquent conduct. Before considering the nature of peer influence, however, it is important to understand the nature of delinquent behavior itself. Are most delinquent acts the work of solitary offenders who operate without the cooperation or support of others, or are they acts committed by groups of adolescents? If the latter, are these groups large or small, stable or unstable, homogeneous or heterogeneous, organized or disorganized? Without some understanding of the nature of delinquent events, it is premature to speculate on their causes.

DELINQUENCY AS GROUP BEHAVIOR

One of the most consistently reported features of delinquent behavior is its group nature. Throughout the last century, investigators have noted a tendency for offenders to commit delinquent acts in the company of others. The “gregarious and companionate character” of delinquency, as Empey (1982: 121) has described it, was documented more than seventy years ago by Shaw and McKay (1931), who discovered that more than 80 percent of juveniles appearing before the Chicago Juvenile Court had accomplices. Similar findings drawn from official data were regularly reported by scholars from the 1920s through the 1960s (see Klein, 1969; Erickson, 1971; Reiss, 1986).

As self-report methodology came to be widely adopted by criminologists, evidence for the group nature of delinquency mounted.