Dale Kunkel & Lara Zwarun. Handbook of Children, Culture, and Violence. Editor: Nancy E Dowd, Dorothy G Singer, Robin Fretwell Wilson. Sage Publication. 2008.
Concern on the part of the public and policy makers about the harmful influence of media violence on children dates back to the 1950s and 1960s (Murray, 1996; Potter & Warren, 1996). The legitimacy of that concern is corroborated by extensive scientific research that has accumulated since that time. Indeed, in reviewing the totality of empirical evidence regarding the impact of media violence, the conclusion that exposure to violent portrayals poses a risk of harmful effects on children has been reached by the U.S. Surgeon General (Surgeon General’s Scientific Advisory Committee, 1972), the National Institute of Mental Health (Pearl, Bouthilet, & Lazar, 1982), the National Academy of Sciences (Reiss & Roth, 1993), the American Medical Association (1996), the American Psychological Association (1993), the American Academy of Pediatrics (Cook, 2000), and a host of other scientific and public health agencies and organizations.
It is well established by a compelling body of scientific evidence that television violence is harmful to children. These harmful effects include (1) children’s learning of aggressive attitudes and behaviors; (2) desensitization, or an increased callousness toward victims of violence; and (3) increased or exaggerated fear of being victimized by violence (Wilson et al., 1997). While exposure to media violence is not necessarily the most potent factor contributing to real-world violence and aggression in the United States today, it is certainly the most pervasive. The average American child spends roughly 20 hours per week watching television (Roberts & Foehr, 2004), and the cumulative exposure to violent images over time can shape young minds in unhealthy ways.
While brief conclusions such as these may be helpful and important in understanding the problem of television violence, they also risk masking or oversimplifying complex issues and relationships. Despite the compelling evidence of harmful effects, not all children are adversely affected and not all portrayals of violence pose a risk of harm. In this chapter, we seek to explicate this and other complexities of the research findings on media violence. More specifically, we will (1) review and summarize the evidence regarding the risk of harmful effects from children’s viewing of televised violence; (2) examine the nature of violence depicted on television, linking our analysis directly to the known risk factors; and (3) consider the implications of the findings about media violence for public policy makers, reviewing previous efforts and analyzing future prospects for resolving the concerns.
The Evolution of Research on Media Violence Effects
Although concern about television violence first surfaced in the 1950s and received some attention in the 1960s, it was not until the 1970s that researchers began to study the topic in depth. At that time, the U.S. Surgeon General was commissioned by Congress to examine the issue and to draw conclusions upon which policy makers could act.
The Surgeon General’s final report on the topic (Surgeon General’s Scientific Advisory Committee, 1972) was complicated and filled with caveats, reflecting its status as both a political and scientific document (Cater & Strickland, 1975). Indeed, the process by which the study was conducted afforded the television industry extraordinary influence and control, such that leading academic scholars who had published evidence of harmful effects were “blackballed” from participation on its advisory board, while researchers employed by the industry were allowed onto the committee (Cater & Strickland, 1975; Liebert & Sprafkin, 1988). The American Association for the Advancement of Science (AAAS) voiced a public complaint at this blatant politicization of science (Boffey & Walsh, 1970).
In the end, critics of the television industry found within the Surgeon General’s report strong confirmation of the harmful effects of children’s exposure to televised violence, while television industry officials argued that the conclusions were too equivocal to be definitive (Kunkel, 2003). Certainly the report triggered more controversy than it resolved, with subsequent debates pursued concurrently in the popular press and the policy arena as well as in the scientific community (Bok, 1998).
By the 1980s, much more empirical research and scientific analysis had been accomplished, and a strong consensus began to emerge in the academic community that exposure to television violence was harmful for children. The effects were not understood to be direct and powerful such that every child who watched would be adversely affected; rather, exposure to violent media was viewed as a risk factor in contributing to the likelihood that children would behave aggressively. This viewpoint is well summarized in an overview of the evidence published by the National Institute of Mental Health (Pearl, Bouthilet, & Lazar, 1982):
Ehe consensus among most of the research community is that violence on television does lead to aggressive behavior by children and teenagers. … Not all children become aggressive, of course, but the correlations between [viewing] violence and aggression are positive. In magnitude, television violence is as strongly correlated with aggressive behavior as any other behavioral variable that has been measured, (p. 6)
In the 1990s, the public health community began to play a more active and prominent role in the debate about media violence. Physicians regarded violence as an “epidemic” (American Medical Association, 1996), and approached the topic as they would any other epidemiological analysis in which causes and cures are examined. Research articles on the topic of media violence started to surface at least as often in medical and public health journals as in the social science literature, their traditional home. The most visible action by the public health community was the Joint Statement on the Impact of Entertainment Violence on Children (Joint Statement, 2000) issued at a Congressional briefing by the American Academy of Pediatrics, the American Psychological Association, the American Medical Association, and the American Academy of Child and Adolescent Psychiatry. This document declared that research points “overwhelmingly to a causal connection between media violence and aggressive behavior” (p. 1), putting the credibility of these organizations behind a push for policy makers to do something to address the problem.
The frequency of new studies of television violence effects has diminished in the past decade, largely as a function of the strong consensus about the conclusion that viewing violence is harmful to children. Yet even in this context, several significant new empirical findings from longitudinal studies have emerged in recent years. One of these, a panel study that followed 6- and 10-year-old subjects over a 15-year period into adulthood, found that childhood exposure to media violence predicted adult aggressive behavior (Huesmann, Moise-Titus, Podolski, & Eron, 2003). Another similar project, published in the AAAS journal Science (arguably the nation’s most respected scientific venue), demonstrated that the level of violence viewed at an average age of 14 is significantly related to assaults and fights resulting in injury at an average age of 22 (Johnson, Cohen, Smalles, Kasen, & Brook, 2002). In both of these studies, the influence of exposure to media violence remained significant even when controlling statistically for a wide range of other factors known to be associated with aggression, such as childhood neglect, family income, parental education, and neighborhood violence.
It is abundantly clear that exposure to television violence poses a risk of harmful effects for children. That conclusion is strongly affirmed by both the social scientific and public health communities, based upon consistent research evidence produced over the past 40 years.
Summary of Key Findings about Media Violence Effects
Over the years, studies of the impact of media violence have encompassed the full range of research methodologies, including experiments, surveys, field research, and longitudinal studies, among others. It is the convergence of findings across the totality of empirical evidence that yields the key conclusions offered here.
Children Learn Aggressive Attitudes and Behaviors from Viewing Televised Violence
Employing both social learning theory and its more recent derivative, social cognitive theory, Albert Bandura (1977, 1978, 2002) has established that observational learning is a critical component of human social behavior. Learning that occurs by observing others is particularly important for young children, who are early in the process of developing an understanding of normative behavioral patterns. While Bandura was one of the first to demonstrate that children may learn and imitate aggressive acts observed in the media (Bandura, Ross, & Ross, 1961, 1963a, 1963b), he has been followed by many other experimenters who have produced comparable results (Comstock & Paik, 1991; Potter, 1999).
Complementing these laboratory studies, an impressive collection of field research has documented a strong association between children’s viewing of televised violence and subsequent aggressive behavior (Huesmann & Eron, 1986; Huesmann, Eron, Lefkowitz, & Walder, 1984). Exposure to television portrayals where problems are resolved by violence may contribute to scripts that children learn and then apply in their own world under trying circumstances. Berkowitz (1984) and his colleagues (Berkowitz & Rogers, 1986; Jo & Berkowitz, 1994) refer to this as “media priming.”
Literally hundreds of experimental and longitudinal studies support the conclusion that viewing televised violence leads to increases in subsequent aggression, and that such aggression can become part of a lasting behavioral pattern. As Huesmann (1986) notes,
Aggressive habits seem to be learned early in life, and once established, are resistant to change and predictive of serious adult antisocial behavior. If a child’s observation of media violence promotes the learning of aggressive habits, it can have harmful lifelong consequences. Consistent with this theory, early television habits are in fact correlated with adult criminality, (pp. 129-130)
Clearly, children’s exposure to media violence, particularly at young ages, can lead to serious adverse consequences. Some who are affected directly may suffer by committing their own acts of aggression; while others may be impacted indirectly as they become the victims of violent actions committed against them. The evidence is overwhelming that viewing televised violence contributes significantly to increases in real-world violence and aggression. An occasional critic has challenged this conclusion (Fowles, 1999; Freedman, 1984), but their arguments or objections have never been sustained (Friedrich-Cofer & Huston, 1986; Huesmann & Taylor, 2003).
Heavy Viewing of Televised Violence Desensitizes Children to Violence and Its Victims in the Real World
Over the years, research has examined the relationship between exposure to media violence and feelings of concern, empathy, or sympathy that viewers may have regarding victims of actual violence. Numerous studies indicate that heavier viewers of media violence demonstrate less physiological reactivity to violent film clips compared to lighter viewers; that general physiological arousal decreases as viewers watch more violent media; and that children as well as adults are susceptible to this effect (Cline, Croft, & Courrier, 1973; Drabman & Thomas, 1974; Thomas, Horton, Lippincott, & Drabman, 1977).
While desensitization effects from exposure to typical action-adventure program portrayals of violence have been well documented, there is also evidence that viewing more explicit or graphic depictions of violence enhances the effect (Linz, Donnerstein, & Penrod, 1984, 1988; Mullin & Linz, 1995).
Viewing of Media Violence Increases Children’s Fear of Being Victimized by Others
George Gerbner and his colleagues (Gerbner & Gross, 1976; Gerbner, Gross, Morgan, & Signorielli, 1986) have repeatedly demonstrated the cultivation effect, by which viewers exposed to heavy doses of television violence come to believe that the world is a violent and scary place. This effect leads to an exaggerated fear of crime or victimization that persists over time, in both children and adults (Wilson et al., 1997). While most of this evidence is correlational, there are also experimental studies that demonstrate the same effect (e.g., Bryant, Carveth, & Brown, 1981).
In addition, more transitory reactions such as extreme emotional fright and sleep disturbances may result from children’s exposure to graphic violence or intense scenes of danger, particularly associated with concrete threats such as from monsters (Cantor, 1994; Cantor & Wilson, 1988; Wilson & Smith, 1998).
These three effects—increases in children’s aggressive attitudes and behaviors, desensitization to real-world victims of violence, and heightened fear of victimization—represent the most significant impacts of children’s exposure to media violence. They occur largely as the product of a slow, cumulative process of watching countless acts of violence across years of viewing. Particularly because this influence process is so gradual, it is important to consider the broad patterns regarding how violence is presented on television over time. Fortunately, as we discuss below, there is an abundance of content-analysis evidence about the depiction of violence on television from which to draw.
Patterns of Violence on Television
Studies of the nature and extent of televised violence have been conducted since the 1960s and remain common today (Comstock & Sharrer, 1999; Gunter, Harrison, & Wykes, 2003; Potter, 1999). Most tend to employ somewhat different definitions of violence for purposes of the tallies employed, and hence they yield somewhat different findings. For example, if one study defines violence to include comic or slapstick actions, its count of violence will likely yield a higher figure than would a comparable study that excluded such actions in its definition. Despite these differences, virtually every study of the topic finds that violent behavior is widespread across the television landscape, with violence typically observed in the majority of programs examined.
Gerbner defined violence as “the overt expression of physical force (with or without a weapon) against self or other, compelling action against one’s will on pain of being hurt or killed, or actually hurting or killing” (Gerbner, Gross, Morgan, & Signorielli, 1980). Using this definition, Gerbner and his colleagues reported findings of approximately 5 to 6 violent acts per hour of television over many years in the 1970s and 1980s. Other scholars have used broader definitions of violence, including actions that cause psychological or emotional harm rather than solely physical damage, and have reported averages ranging from 18 to 38 violent acts per hour (Greenberg, Edison, Korzenny, Fernandez-Collado, & Atkin, 1980; Potter et al., 1995; Williams, Zabrak, & Joy, 1982). Still other researchers count differently, measuring the number of scenes with violence, rather than the number of violent acts. For example, Lichter and Amundson (1994) reported that an average of 10.7 scenes per hour of prime-time programming in 1992 contained violence. Note that a scene could contain 1, 5, 10, or any number of violent actions and it would still be coded identically as simply “contains violence” under this approach.
It is apparent that the strategy for measuring violence on television plays an important role in shaping the results. Most studies employ a highly reductionistic approach in framing their central findings, by which we mean that their summary statistics count all acts of violence as equivalent to one another in adding the totals. This common tactic poses a significant validity problem if one is trying to draw implications for effects in the three critical areas we have presented above— learning aggressive attitudes and behaviors, becoming desensitized, and developing an exaggerated fear of violent attack. That is because there is strong evidence that not all portrayals of violence pose the same degree of risk for contributing to each of these various outcomes, as we explicate below.
The Importance of Context
In the debate about television violence, a frequent oversimplification—and a critical focus here—is to assume that all violence is essentially the same in terms of its implications for effects. Such an assumption is implicit in any analysis that aggregates all observations of violent behavior. Upon closer examination, however, such an assumption is clearly unwarranted, though it is easy to understand why this perspective has become so widespread over time. From the 1960s to the 1980s, the most visible researcher who studied media violence was George Gerbner, and his violence index tracked the overall levels of violent behavior from year to year, indicating whether violence on television was “up or down” (Rowland, 1983). The news media embraced this angle and reported the annual findings dutifully (Hoffner, 1998). Policy makers typically gave these reports careful scrutiny (Cooper, 1996; Murray, 1996). Given that Gerbner and his colleagues were interested primarily in predicting television’s influence on beliefs and perceptions about social reality, such a tactic may well have been reasonable. It is not, however, a prudent approach for examining the impact of television violence on subsequent aggressive behavior.
There are many ways in which to depict violence on television. For example, the violence may occur on-screen and be shown graphically or it may occur offscreen but be clearly implied. Violent acts may be shown close-up or at a distance. There are differences in the types of characters who commit violence and their reasons for doing so. And there are differences in the outcomes of violence—some depictions focus on the pain and suffering of victims, whereas others avoid showing the negative consequences of physical aggression. Simply put, not all portrayals of violence are the same. Their context can vary on many important dimensions. These differences matter because there is substantial evidence that such differences in the message characteristics hold important implications for the impact of particular violent scenes on viewers (Kunkel et al., 1995; Wilson et al, 1997).
In the mid-1990s, a large-scale project called the National Television Violence Study (NTVS) reviewed all existing effects of research that had previously examined the influence of various contextual features, and devised a content-analysis framework that measured attributes known to either increase or diminish the likelihood of one of the three basic types of harmful influence (Wilson et al., 1997). For example, violence that is rewarded or shown without punishment is known to increase the likelihood of subsequent viewer aggression, whereas violence that is punished tends to diminish the risk of such effects. Similarly, violent actions that are more realistic increase the probability of subsequent aggression, whereas portrayals that are less realistic reduce that prospect. A summary table of the key contextual features associated with violence on television, and their relationship to the three basic types of harmful effects.
Employing this analytical framework, the NTVS project examined approximately 10,000 television programs across all times of day on the most frequently viewed channels over a 3-year period, from 1994 to 1997. Its definition of violence was similar to Gerbner’s and focused on physical harm:
Violence is defined as any overt depiction of a credible threat of physical force or the actual use of such force intended to physically harm an animate being or group of beings. Violence also includes certain depictions of physically harmful consequences against an animate being or group that occur as a result of unseen violent means. (Wilson et al, 1997, p. 53)
Based upon this project, which is certainly among the largest scientific studies of television violence, the following three key conclusions were drawn.
Violence is Widespread across the Television Landscape
When viewers turn on a television set and pick a channel at random, the odds are better than 50-50 that the program they encounter will contain violent material. More specifically, 60% of all shows sampled across the entire 3-year project contained some form of violence. An average of 6,000 violent interactions were observed in a single week of programming across the 23 channels studied each year, including both broadcast and cable networks. More than half of the violent shows (53%) contained lethal acts, and 1 in 4 of the programs with violence (25%) depicted the use of a gun (Smith et al., 1998).
Most Violence on Television is Presented in a Manner that Increases Its Risk of Harmful Effects on Child Viewers
Most violence on television follows a highly formulaic pattern that is both sanitized and glamorized. A sanitized depiction of violence means that the portrayal fails to show realistic harm to victims, both from a short- and long-term perspective. Immediate pain and suffering by victims of violence is included in less than half (46%) of all scenes of violence. More than a third of violent interactions (36%) depict unrealistically mild harm to victims, grossly understating the severity of injury that would occur from such actions in the real world. Most depictions sanitize violence by making it appear to be much less painful and less harmful than it really is.
A glamorized depiction refers to violence that is performed by attractive role models who are often justified for acting aggressively and who suffer no remorse, criticism, or penalty for their violent behavior. More than a third of all violence (35%) is committed by attractive characters, and more than two-thirds of the violence they engage in (73%) occurs without any signs of punishment. All of these patterns are important because violence that is sanitized or glamorized increases the risk of harmful effects on child viewers.
There is Remarkable Stability in the Presentation of Violence on Television
A summary of findings comparing the 1994-1995, 1995-1996, and 1996-1997 television seasons illustrates the tremendous degree of consistency that is found in both the nature and extent of violence on television. On each of the critical contextual attributes associated with the presentation of violence, the frequency statistics varied no more than a few percentages from year to year over the 3 years studied and across roughly 10,000 programs. That consistency demonstrates that the portrayal of violence is highly stable and formulaic—and unfortunately, this formula of presenting violence as glamorized and sanitized is one that enhances the risk of harmful effects for the child audience.
The NTVS study establishes clearly that the level of violence on television poses substantial cause for concern. It demonstrates that violence is a central aspect of television programming that enjoys remarkable consistency and stability over time. The emergence of these data in the 1990s, coupled with the growing consensus from both the social scientific and public health communities about the harmful effects of media violence, created an impetus for policy makers to consider the issue more closely. In the next section, we examine the history of how policy makers have sought to address the problem of television violence.
Media Violence and Public Policy
Public concern about violent content in motion pictures (Jowett, Jarvie, & Fuller, 1996) and comic books (Wertham, 1954) long preceded the advent of television. Thus, it is not entirely surprising that policy makers seemed eager to engage with the issue of television violence, even while the medium was still in its infancy. Congressional hearings on the topic were conducted beginning in the mid-1950s, before a single study of the effects of televised violence had yet been completed (Cooper, 1996; Kunkel & Wilcox, 2001).
In the early years of television, all programming viewed in the United States was delivered by broadcast stations; cable television did not emerge as a viable consumer option until the 1980s (LeDuc, 1987). Interestingly, the regulatory framework applied to broadcast television is more stringent than that applied to other electronic media such as cable television. That is because broadcasters transmit their signals over the publicly owned airwaves.
To use the public airwaves, broadcasters must be licensed by the government, and by accepting a license they agree to serve the “public interest, convenience, and necessity.” The power both to license broadcasters and to establish the public-interest policies they must follow was granted by Congress to the Federal Communications Commission (FCC) in the Communications Act of 1934. This power has been used at times to place content-specific regulations on the broadcast industry, such as in the area of indecency restrictions (Hilliard & Keith, 2003). The Supreme Court has consistently upheld the constitutionality of the FCC’s authority to regulate content on broadcast channels even in the face of the First Amendment, so long as the regulations are narrowly drawn and further compelling government interests (Action for Children’s Television et al. v. FCC, 1995; FCC v. Pacifica, 1978; Red Lion Broadcasting Co. v. FCC, 1969). In contrast, other media that do not use the public airwaves and thus are not licensed by the government (e.g., motion pictures, cable television networks) generally receive greater First Amendment protections from direct content regulation, which in theory makes them harder to regulate (Carter, Franklin, & Wright, 2003). Thus, governmental interest in violence on broadcast television is uniquely justified from a legal perspective.
The 1950s to 1980s: Diminishing Threats of Regulation
From the 1950s to the 1980s, the principal tactic employed by the Congress and FCC to address concern about televised violence was known as the “raised eyebrow” approach (Cole & Oettinger, 1978; Krasnow, Longley, & Terry, 1982). With this strategy, policy makers assume threatening postures in hearings and public statements, voicing their strong concern and vaguely warning of regulation if the television industry does not voluntarily address the concern, or euphemistically speaking, “clean up its act.” When such threats about television violence were sustained over time and viewed as serious by industry officials, as often occurred in the 1970s, they indeed have resulted in significant reductions in the levels of violence on television (Murray, 1980). But because violent action is an important staple for attracting large audiences to television (Bartholow, Dill, Anderson, & Lindsay, 2003; Hamilton, 1998), such reductions tend to diminish quickly once the intensity of the regulatory threat seems to abate.
The “raised eyebrow” tactic was well tailored to fit the equivocal status of the research evidence that was still accumulating during this period. Policy makers never had to have an “air-tight” case to use their bully pulpit, since no specific regulatory action was ever formally proposed. During this time, the television industry used its political muscle to focus the debate about television violence squarely on the question of whether any harms could actually be established, rather than on whether violence ought to be regulated (Baumgartner & Jones, 1993). This strategy effectively defined the debate in a way that inherently weakened its opponents’ position. For example, the ABC network published its own report (American Broadcasting Companies, 1983) in response to the National Institute of Mental Health (1982) review of research that had drawn the first strong causal conclusions about harmful effects. Acknowledging that “the body of literature on television and violence continues to expand” (p. 1), the ABC report criticized the NIMH’s emphasis on convergence of the evidence, stating that “in social science, convergence—the analysis of many different studies which point in the same direction—is sometimes used when no definitive evidence can be found to clearly support a position” (p. 6). In the face of such challenges, and in the absence of overwhelming research evidence, the broadcast industry successfully parried all the raised eyebrow threats and concerns from policy makers for several decades.
The policy debate about televised violence quieted for a time by the late 1980s. From a regulatory perspective, this was a period that emphasized deregulation throughout the government (Derthick & Quirk, 1985), and in particular at the FCC (Fowler & Brenner, 1982). In this context, threats of regulation lacked much of their punch as compared to the past. Moreover, after decades of ritual warnings without any formal sanctions ever being adopted, the industry seemed to have grasped the hollow nature of the threats and appeared less concerned than ever. Starting in the early 1990s, however, the efforts of a single U.S senator, Paul Simon of Illinois, reinvigorated the debate.
The 1990s: Finally Some Firm Policy Action
Convinced by his own reading of the evidence that violent portrayals were harmful to children, yet uncomfortable with the prospect of any governmental censorship of program content, Senator Paul Simon sought an alternative path to resolve the issue. He sponsored legislation enacted by Congress in 1990 that encouraged the television industry to adopt industry wide guidelines regarding its treatment of violence (MacCarthy, 1995). Technically speaking, the legislation merely exempted the networks from federal antitrust law that might preclude them from reaching industrywide agreements affecting their business. In other words, no formal action by the industry was required by Senator Simon’s legislation; rather, it simply provided a clear opportunity for a socially responsible initiative. In speeches and public appearances, Senator Simon implored television officials to both reduce the amount of violence on television, and to avoid glamorized and sanitized violence, which research has shown to increase the risk of harmful effects.
Perhaps braced by too many years of successfully dodging policy makers’ concerns, the television industry turned a blind eye to Senator Simon’s legislation and proposal. This time, however, the industry faced a much different political context, which ultimately led to a much different policy outcome. Public concern about the issue of violence in society was at an all-time high, fueled in part by sensationalistic media coverage of school shootings and other violent attacks by youth (Chyi & McCombs, 2004; Pew Research Center for People and the Press, 1999). At the same time, the growing body of research evidence about the harmful effects of television violence had generated an unprecedented consensus from the scientific and public health communities indicting the media as a causal factor in real-world violence and aggression. Similarly, public opinion about television violence reflected more impressive levels of concern. Surveys at the time reported that strong majorities agreed that television violence is harmful to children and that there is too much violence on television (Duston, 1993; Guttman, 1994; Times Mirror Center for People and the Press, 1993).
In the eyes of policy makers and the public, the industry appeared callous to the growing concern about televised violence. Its long-standing tactic of contesting the research evidence was no longer tenable, and its failure to respond more forcefully to Senator Simon’s initiatives left it more vulnerable than ever before to some sort of government regulation. In this context, policy makers were poised to act. The focus of debate had clearly shifted from whether television violence should be regulated to how such regulation might be best accomplished. The challenge, as legal scholar Kevin Saunders (1996) put it, “lies in finding a way to work with and within the First Amendment” (p. 58).
If the government were to adopt any limits on the depiction of violence on television, such restrictions would face numerous hurdles in order to be considered constitutional. According to legal precedent, the research evidence must establish a “compelling governmental interest” in reducing children’s exposure to such content; the policy must effectively accomplish the goal of significantly reducing children’s exposure to violence known to be harmful; and the policy must be the “least restrictive means” of accomplishing the goal, avoiding restraints on the rights of adults to see violent media content if they wish (Edwards & Berman, 1995; Prettyman & Hook, 1987). These challenges have no doubt contributed to policy makers’ long-standing reluctance to formally regulate television violence. Indeed, legislators have long searched for “First Amendment-friendly” solutions such as industry self-regulation, but had not yet succeeded in finding any that seem promising.
The ó-Chip: A New Technology Offers a New Policy Solution
A technological development in the early 1990s offered the prospect of an innovative solution that had the potential to skirt most of the First Amendment concerns. That development was the V-chip, an electronic device that could serve as a filter to block programming that posed any risk of harmful effects on children. If television programming could be categorized with ratings that identified violent content, then parents could implement the electronic blocking technology to shield their children from exposure to inappropriate content. By shifting the policy intervention from a sender-based to a receiver-based restriction, the government was removed “from the constitutionally disfavored position of having to decide what its citizens are allowed to see and hear” (Samoriski, Huffman, & Trauth, 1997, p. 146).
Legislation to adopt the V-Chip had been pending in Congress when President Clinton emphatically praised the device in his 1996 State of the Union address. Congress responded to this exhortation by including a section on the V-chip in the Telecommunications Act of 1996, which was passed shortly following the president’s speech. The law requires that all televisions sold in the United States include a V-chip device that will facilitate program-blocking capabilities, and establishes uniform technical standards for its use. In order to avoid First Amendment controversies, the legislation does not formally require the television industry to devise a rating system by which its programming must be clearly identified. Of course, without such a rating system the V-chip would be worthless, but the industry quickly acquiesced and agreed to implement a ratings framework (Spitzer, 1998). Under the law, the FCC must review the industry’s system for rating programs and deem it acceptable in order to link it to the electronic blocking capability. This review was accomplished in 1997. Even if it was not approved, however, no sanctions could be applied. Thus, the ratings system is considered “voluntary” from a legal perspective.
As currently implemented, there are two key dimensions to the V-chip ratings: age-based advisory categories, and content-based descriptors (Federman, 2002). The age-based advisory categories fall into two groups: shows directed to adults and shows for children. Adult programming includes four rating categories: TV-G (suitable for all ages); TV-PG (parental guidance may be necessary); TV-14 (unsuitable for children under 14); and TV-MA (specifically designed for adults). Children’s programs may be rated in either of two rating categories: TV-Y (appropriate for all children) or TV-Y7 (designed for children aged 7 and above).
In addition, the V-chip rating system also includes content descriptors to help clarify why a program received the age-based rating it was assigned. In adopting the V-chip legislation, Congress called upon the television industry to rate its programs not just for violence, but also for sex and other material that parents might find objectionable for their children. Thus, programs for adults may include the content descriptors V for violence, S for sexual behavior, D for sexual dialogue, or L for adult language. A “V” rating indicates moderate violence when attached to a TV-PG program, intense violence when attached to a TV-14 program, and graphic violence when attached to a TV-MA program. V ratings are not applied to TV-G programs. In addition, an FV rating may be attached to children’s programs designated TV-Y7 to identify intense or combative fantasy violence.
By the end of 1997, V-chip ratings were being applied to most programs on television. A small billboard in the upper corner of the screen displays the rating during the first 15 seconds of each show. By 2000, the V-chip device was included in all new television sets sold in the United States with 13-inch or larger screens. At that point, use of the V-chip’s electronic blocking capability became feasible for those families who had obtained a television set featuring this new technology.
Evaluating the V-Chip
In the early years of the V-chip’s availability, studies showed that the electronic blocking capability was used by only modest numbers of parents (Foehr, Rideout, & Miller, 2001; Stanger & Gridina 1999). This is hardly surprising due to the limited diffusion of the technology, which essentially required the purchase of a new television set. In contrast, more than half of parents typically reported that they had used information about a program’s V-chip rating (as opposed to the electronic blocking capability) to help them decide whether or not their child should view a particular program (Foehr et al., 2001; Kaiser Family Foundation, 2001; Krcmar, Pulaski, & Curtis, 2001).
Several years later, additional data indicate that use of ratings information by parents has remained stable over time at about 50%, while use of the V-chip’s blocking technology has increased significantly (Kaiser Family Foundation, 2004). Of parents who have a V-chip-equipped television set and know it, 42% report using the electronic blocking capability; and of that group, nearly two-thirds (61%) say they found it “very useful.” In sum, it appears that the utility of the V-chip for parents is increasing over time, although substantial numbers remain unaware or uninterested.
The nature of the program content being blocked is also a critically important factor to consider in evaluating the impact of the V-chip. Content-analysis research indicates that many shows containing substantial amounts of potentially harmful violence do not receive a V rating (Kunkel et al., 2002). Perhaps more importantly, the V-chip rating system is not devised to rate programs according to their risk of harmful effects on children, but rather relies upon vague and subjective standards to differentiate “moderate,” intense, and “graphic” portrayals of violence. Furthermore, the rating judgments are obtained not by child development or media effects experts, but rather by television industry officials who are responsible for rating their own programming. To date, no network has publicly revealed any policies, rules, or criteria that are used for classifying its programs.
Given this context, it is not surprising that parents often disagree with rating judgments. One study found that 39% of parents say that most shows on television are not rated accurately (Kaiser Family Foundation, 2004), while another that asked parents to apply their own ratings to videotaped programs revealed that most judge programs more restrictively than did the television industry (Walsh & Gentile, 2001). A further complication is that many parents are confused by the rating system, potentially limiting its utility (Bushman & Cantor, 2003). For example, many parents believe that the rating FV stands for family values rather than fantasy violence, while others mistakenly assume TV-Y7 identifies programming that is safe for children 7 and under, rather than the opposite (Kaiser Family Foundation, 2004). Clearly, the V-chip cannot be effective in addressing the issue of televised violence unless the ratings are applied accurately and consistently, and parents understand the meaning of the various categories.
Notwithstanding the challenges of comprehending the V-chip ratings, parents are also faced with mastering unique media rating systems for such different platforms as motion pictures, videogames, and Internet sites, among others. According to a national survey, a strong majority of parents (78%) support the creation of a single, universal rating system rather than the mix of media ratings currently used; and a similar proportion (70%) would prefer that an independent group of parents, educators, and child development experts oversee the ratings (Common Sense Media, 2003). This prospect has received support from academic scholars (Walsh & Gentile, 2001) and has begun to draw the attention of public policy makers.
An important lesson gained from research could certainly be applied to development of a universal media ratings system: emphasize descriptive information about the content rather than age-based advisory categories. Parents clearly prefer descriptive ratings (e.g., five levels to describe the intensity of violence) over judgments about the age of children for which the material is appropriate (Bushman & Cantor, 2003). Furthermore, age-based ratings pose the risk of a “forbidden fruit” effect by which children’s interest is increased once they learn they are too young to be allowed access to the content; whereas this effect is not observed with the use of content-descriptive ratings that merely indicate levels or types of violence (Bushman & Cantor, 2003; Cantor, 1998).
It remains to be seen whether the V-chip policy will play an important role in resolving the concern about television violence. Many issues persist about the proper design and functioning of the system. But even under optimal conditions, the V-chip system can only protect youth in families where the parents play an active role in supervising their children’s media use. It would seem that latchkey children and youngsters who live in at-risk households with less attentive parents have much greater need for protection from media violence; while ironically, the V-chip is most likely to be used for children in the lower-risk families. The greatest asset of the V-chip is that it allows individual choice for each family. The greatest limitation is that it does nothing to protect society against increased violence and aggression that results from violence viewing by children whose families ignore the V-chip.
Other TV Violence Policy Alternatives
Despite the existence of the V-chip, there still seems to be interest in considering other policy alternatives for combating television violence. A recent national poll reported that 63% of parents favor new regulations to limit sex and violence during the early evening hours, when children are most likely to be watching television (Kaiser Family Foundation, 2004). The industry implemented its own self-regulatory policy in the 1970s known as the “Family Hour,” with broadcast networks voluntarily committing to air family-friendly material (i.e., no sex or violence) in the first hour of prime time. Adoption of the Family Hour policy was prompted by strong pressure from the FCC, and because of this the courts ultimately judged that the industry’s action was not voluntary, and therefore overturned it on First Amendment grounds (Cowan, 1979). It is unlikely that such a policy could legally be implemented by the government, although a related proposal has received substantial attention in recent years.
Beginning in 1993, Senator Ernest Hollings of South Carolina (now retired) introduced legislation in every session of Congress that would regulate television violence in essentially the same manner as broadcast indecency (Kunkel, 2003). Under the FCC’s indecency policy, stations may air indecent material late at night, but cannot do so between the hours of 6:00 a.m. and 10:00 p.m. when children are likely to be in the audience. This approach is termed a “safe harbor” policy, though ironically the safe harbor refers to the hours late at night when the broadcast of sensitive material is permitted. This safe harbor period makes the policy legally safe or compliant with the First Amendment by helping the government to meet the “least restrictive means” test applied in the courts, as opposed to the more restrictive alternative of banning such material entirely.
Senator Hollings argued that the social science evidence documenting the risk of harmful effects from children’s viewing of televised violence constitutes a “compelling governmental interest,” which therefore legitimizes intrusions on otherwise protected speech. His proposal is controversial for several reasons: It attracts strong opposition from free-speech advocates, who are troubled by its strong degree of government censorship. It faces a difficult practical challenge of identifying the violence that would be subject to the policy, while allowing adults access to important political information such as news depictions of violence. And finally, to be effective, the policy would have to be applied to cable as well as broadcast television, which would pose even greater challenges to the law’s constitutionality. It is well established that non-broadcast media such as cable qualify for greater First Amendment protection than do broadcast media (Carter, Franklin, & Wright, 2003).
Finally, another policy option that has long been discussed involves the prospect of tort liability, whereby the media would be held accountable for damages when exposure to their products contributes to real-world violence and aggression. This tactic faces particular difficulty given the nature of the process by which media violence influences the audience. It is rarely the case that a direct and powerful response occurs after viewing a single program. Rather, the effects of media violence are typically slow, gradual, and cumulative, in much the same way that cigarette smoking contributes to adverse physical health outcomes. This makes causality more difficult to demonstrate, though not any less real (Gentile & Sesma, 2003).
In the case of Zamora v. Columbia Broadcasting System et al. (1979), a 15-year-old plaintiff sued three major television networks, claiming that his cumulative exposure to televised violence since the age of 5 led him to murder an elderly neighbor. The case failed on legal grounds having to do with the lack of evidence about the degree of causality and the foresee ability of the outcome by the broadcasters, among other factors (Prettyman & Hook, 1987).
Another case that involved more direct imitative violence (Olivia N. vs. National Broadcasting Company, 1981) also met the same outcome. In this case, 9-year-old Olivia N. was attacked and raped by a group of youths using a plumber’s helper handle 4 days following the broadcast of the movie “Born Innocent,” in which an identical event was depicted. Again the court said that the plaintiffs had failed to establish that the criminal conduct had been facilitated or made particularly tempting by the broadcast (Prettyman & Hook, 1987).
The duty for media industries to guard against negligent or criminal actions by others after viewing their products increases if the harmful outcome is foreseeable. This factor is particularly relevant for a program such as “Jackass,” a Music Television Network (MTV) show that features a host who performs dangerous stunts for comic effect. Although MTV claims that the show targets 18- to 24-year-olds, one-third of its audience is 17 or younger, and several incidents have occurred in which younger viewers harmed themselves or others in efforts to imitate such actions. For example, a 13-year-old poured gasoline on his arms and legs and lit himself on fire in an attempt to mimic a human barbeque stunt presented on the show (Minow & Minow, 2003). At the outset of each episode, “Jackass” delivers a warning to viewers not to imitate the stunts shown on the program, but the disclaimer was initially conveyed in a humorous, mocking tone that seemed to trivialize its importance. A network spokesperson observed in the press that it is “incredibly upsetting” when young people hurt themselves, but that MTV is not responsible (“MTV Shuns Responsibility,” 2001). Following the copycat incidents, the warnings were changed to a much more serious format.
The use of legal torts to hold the media industries accountable when their violent products contribute to physical harm in the real world remains theoretically possible, but rather impotent in practice given existing legal precedent. It is worth noting that most of the key legal decisions were accomplished at a point in time historically when the scientific evidence regarding media violence effects was much less developed than is the case today. Although the hurdles remain high for demonstrating legal culpability, many view this avenue as one of the most promising for exerting an impact on the industry’s depictions of violence, should a legal breakthrough in this area occur.
In this chapter, we have demonstrated the clarity of the evidence that media violence is a significant factor contributing to real-world violence and aggression. Various observers have estimated the strength of the correlation between exposure to media violence and subsequent aggression at between +.11 (Hogben, 1998) and +.31 (Comstock & Scharrer, 2003), which on average would be greater than the strength of the relationship between homework and academic achievement or between calcium intake and bone mass (Bushman & Anderson, 2001). As the science has grown in this realm, we have come to understand that not all violence poses the same degree of risk of harmful effects. Indeed, it is theoretically possible that media violence could be presented in a manner that appears so ineffective or repugnant that exposure to such depictions could actually have beneficial outcomes, reducing the likelihood that viewers would subsequently behave aggressively. Unfortunately, content-analysis research establishes clearly that such portrayals are extraordinarily uncommon, and that the most frequent patterns associated with the violence shown on television serve to increase the risk of harmful effects, particularly for child viewers.
It is this scenario that leads policy makers to invest significant efforts to address the issue of media violence. Resolving the issue is no simple matter, as important First Amendment protections must be balanced against the risks of harm from viewing violent material. Policy makers have focused their efforts over the years on broadcast television, which affords the easiest avenue for any direct regulation of content, but this medium is diminishing in importance in an era of increasing media competition and technological innovation. The V-chip stands as the most noteworthy policy to date to address the concerns about media violence. It seems unlikely it will be the last such effort at policy making in this area. The importance of the issue warrants continued careful scrutiny from all interested parties, including parents, policy makers, media practitioners, and health care providers.