In January 2001, a group of doctors at Stanford University released one of the most encouraging studies I’d seen in a long time. They found that when elementary school kids voluntarily cut back on the number of hours they watched TV and played video games, they subsequently behaved less aggressively. Everything from playground shoving matches to kids snapping, “I’m gonna kick your butt!” decreased when the kids started turning off the TV after about an hour a day. As the study’s lead author, Dr. Thomas N. Robinson, commented, “What this shows is that there is something you can do in a practical way, in a real-world setting, and see the effects.” That was great news to all of us who sometimes despair at our power to make the world a little saner for our kids.
The study didn’t get the news coverage it deserved (probably because it was released right before the Super Bowl), but nearly every news article I saw interpreted it in a similar way. The San Francisco Chronicle led off its front-page story with, “Aggressive tendencies fostered in children by violent television shows and video games can be tempered if they cut back their viewing and playing, a new Stanford University study shows.” What none of those news stories mentioned, however, was the fact that the study didn’t distinguish among types of media content. Not only did the kids who watched the World Wrestling Federation and played Dead on Arrival improve their behavior when they cut back on tube time, but so apparently did the kids who watched only Rugrats and Arthur, or Blind Date and Saved by the Bell.Those results make no sense if what fostered their aggression was violent imagery.
What, then, does this study show? As with most studies, there are many possibilities. It may show that if kids get off their rear ends and go do something active for a change, they’ll be less bored, less restless, and less likely to pop off out of annoyance–a finding that anyone who has spent much time with kids can endorse. In fact, in 1999,Dr. Robinson authored another study, based on the same research, suggesting that reducing TV and game time reduced kids’ obesity, a finding that presumably reflects the benefits of greater physical activity more than the content of the programming.
The 2001 study may also show that kids who spend less time in mental isolation develop more effective social skills, which, as Dr. Robinson pointed out, is supported by the anecdotal evidence surrounding the study. “We had parents who said, ‘This is the best thing that’s happened to our family–we talk to our kids at dinnertime now.’ One mother called and said her daughter used to sit at home and watch TV, and now she’s found a friend down the block and they play outside every afternoon.” The study also suggests that if kids take control of their leisure time instead of running on habit–if they watch or play only what they really choose to, whether that be Zaboomafoo or Doom– they may be able to take more control of themselves in general. Most exciting to me, it suggests that if kids are given a purpose–even one as simple as, “Let’s cut our media consumption for a while and see what happens”–they’ll feel better about themselves and use their energies more constructively. If they see adults as caring about them, trusting them to meet higher expectations, and wanting to work with them instead of just leaving them to kill the hours between school and bed and noticing them only when they act up, then they will respond.
One thing the study pointedly does not show is that the children became less aggressive because they saw less violence. Why, then, did nearly every news story interpret it as showing exactly that?
The reason, I believe, is that this misinterpretation has become such an integral part of our discussion of entertainment violence that we’ve ceased to recognize it. It’s become such a habit of thought that when we hear the words “media” and “aggression” together, we go instantly back into the same looping conversation without stopping to be sure of what’s been said. We expect to see evidence that violent imagery leads to violent behavior, and so we do see it. Not only do we expect it, we often want to see it, because it provides a reassuringly familiar answer. So we look in only one direction even when logic, personal experience, and the scientific data itself should lead us to look elsewhere.
Preconception is a powerful force. We smack up against it as individuals, professional communities, and whole societies. Once we’ve come to an opinion as to what the truth is (or what we want it to be), even the best trained and most conscientious of us find our thoughts flowing to fit the mold. We seize too quickly on what looks like supporting evidence and ignore evidence to the contrary. And if our preconceptions concern something as important as the health of our society or the welfare of our children, we may even angrily resist opposing information because we fear it will weaken our resolve to do what (we think) is best.
In the late nineteenth century, the standard position of the medical establishment, to paint it with a broad brush, was that sexuality was a dangerous force best left unaroused and undiscussed. Scientific research and anecdotal material were interpreted in the light of that preconception, so that sexual dysfunctions were normally explained as an excess of sexual stimulation. If a criminal were found to be a masturbator, the masturbation would be assumed to be a cause of his criminality. As the new schools of thought led by Freud and others began to convince people that the problem wasn’t so much sexuality itself but our relationship with it, a relationship forged largely in childhood, the same body of knowledge began to look very different. The same case studies and statistics that had once seemed to point to sexual stimulation as the source of trouble and the concealment of sexuality as the best solution now began to suggest, more compellingly, that sexual repression was the greater problem and that open acceptance of sexuality was a more effective way to deal with it.
Unfortunately, many of those Freudian ideas became the bases of new preconceptions in the decades that followed. In seeking the cause of abnormal behavior in childhood maladjustment, doctors and psychiatrists fell into some dangerous misinterpretations of the evidence before them. Early research into autism looked at its possible biological origins, but then orthodoxy shifted toward blaming bad parenting. Families were torn apart, children mistreated, and years of work wasted until new generations of researchers were able to show that autism was biologically determined after all. Homosexuality was overwhelmingly considered to be a neurosis produced by skewed parent-child relationships, one that could and should be “cured” by psychoanalysis. Depression, anxiety, alcoholism, and other conditions were often approached as solely mental conditions, even when they had physiological manifestations.
Most of those preconceptions dominated the public health establishment well into the 1960s. Then, as new ideas swept through the culture at large–especially a willingness to accept and work with innate differences instead of attempting to “normalize” them–they were broadly reexamined. Not only did new research now generate revealing new data, but the old data were reinterpreted in new ways. Suddenly new views of all these aspects of human behavior became possible, and from those views new policies and treatments could be developed that have since proven to be far more helpful and humane. As Emerson wrote, “People only see what they are prepared to see.”
On July 26, 2000, officers of the American Medical Association, the American Academy of Pediatrics, the American Psychiatric Association, the American Psychological Association, the American Academy of Family Physicians, and the American Academy of Child and Adolescent Psychiatry issued a “Joint Statement on the Impact of Entertainment Violence on Children,” which was subsequently endorsed by both houses of the United States Congress.
At this time, well over 1,000 studies–including reports from the Surgeon General’s office, the National Institute of Mental Health, and numerous studies conducted by leading figures within our medical and public health organizations–our own members–point over-whelmingly to a causal connection between media violence and aggressive behavior in some children. The conclusion of the public health community, based on over thirty years of research, is that viewing entertainment violence can lead to increases in aggressive attitudes, values, and behavior, particularly in children. . . .
The effect of entertainment violence on children is complex and variable. Some children will be affected more than others. But while duration, intensity, and extent of the impact may vary, there are several measurable negative effects of children’s exposure to violent entertainment. . . .We in no way mean to imply that entertainment violence is the sole, or even necessarily the most important factor contributing to youth aggression, anti-social attitudes, and violence. . . . Nor are we advocating restrictions on creative activity. The purpose of this document is descriptive, not prescriptive: we seek to lay out a clear picture of the pathological effects of entertainment violence. But we do hope that by articulating and releasing the consensus of the public health community, we may encourage greater public and parental awareness of the harms of violent entertainment, and encourage a more honest dialogue about what can be done to enhance the health and well-being of America’s children.
Clearly this was no unreasoned statement. It stood upon extensive research, and it spoke pointedly of “some children.” But the view of children and entertainment violence it presented was broad and simple. The impact of entertainment violence varies in “duration, intensity, and extent,” but not in its essential nature or quality; its effects are “pathological” and “negative,” never beneficial. When such a broad, simple statement is made about the infinitely complex and mutable behavior of human beings, we need to look very hard at what has led up to it. We need to determine whether there is in fact a consensus of opinion among the experts, whether the research has in fact been viewed from every angle, and whether our conclusions are leading us to the most effective actions in the real world.
The answer to the first question is simple: there is no unity of expert opinion on the effects of entertainment violence.
Dr. Stuart Fischoff, founder of the Media Psychology lab at California State University in Los Angeles, said in his 1999 address to the American Psychological Association:“Whether we cite 100, 1,000, or 10,000 research studies which conclude that exposure to violent media produces violent behavior, 10,000 is no more persuasive or credible than 100, if the designs of the research are flawed and/or the generalizations to an external population of behaviors are patently unjustified. The current violence in society is disturbing to all of us. The current excessive, gratuitous violence in film, in video games, in music lyrics is disturbing to all of us. But because two phenomena are both disturbing and coincident in time does not make them causally connected. . . .Abhorrent as what I have to say may be, I believe that there is not a single study that is externally valid. . . . After 50 years . . . , there is, I submit, not a single research study which is even remotely predictive of [events like] the Columbine massacre.”
Dr. Helen Smith is a forensic psychologist who has evaluated thousands of violent minors for law enforcement and the courts, administered a nationwide voluntary survey of violent youth, and written The Scarred Heart: Understanding and Identifying Kids Who Kill. “Not one young person in my experience has ever been made violent by media influence,” she asserted. “Young people who are already inclined to be violent do feel that violent media speaks to them. A few do get dangerous ideas from it. But more of them find it to be a way to deal with their rage.”
“Entertainment has the power to overexcite or present a distorted worldview to some children,” reported Dr. Edwin Cook, a professor of psychiatry and pediatrics at the University of Chicago who specializes in children’s developmental issues. “But for other children, the right aggressive entertainment might be the best thing they could see.”
Dr. Lynn Ponton, psychiatrist, authority on adolescent behavior, and author of The Romance of Risk and The Sex Lives of Teenagers, said, “There are dangers to a young person in isolation trying to contain his or her anxieties through the habitual use of movies or video games, of any type or genre. But I don’t believe the content of the movies or games really matters.”
Although one would be hard-pressed to find a single expert asserting that physical abuse, neglectful parents, or poverty is not hurtful to children, there is a chorus of authoritative dissent from the usual condemnation of violent entertainment. Even the Surgeon General’s Report on Youth Violence, released seven months after the Joint Statement, disagreed on some key points. The latter claimed that “children exposed to violent programming at a young age have a higher tendency for violent and aggressive behavior later in life.” In contrast, the Surgeon General found that media violence can have a short-term effect (a teenager is excited by a scene in a movie or game and wants to go out and do it) but found no convincing evidence of a long-term effect. He also asserted that media is a minor factor in youth violence compared to family and peer-group issues and that the relatively few youngsters likely to be affected by it are those who are “already aggressive for some other reason.”
According to Dr. Jonathan Freedman of the University of Toronto, a widely respected critic of the research literature on media violence, “the further people are from the data, the more excited they are by it.” He points out, for example, that according to every meta-analysis of the research, including those conducted by supporters of the media-aggression hypothesis, there have been not 1,000 but about 200 studies, and many of those have contradicted the conclusion of the Joint Statement. It’s important to remember that the doctors and administrators who author documents like the Joint Statement are rarely better versed in the media-aggression research than those of us who read about the latest study in the morning newspaper. They are so busy with their own professional duties that they are dependent on secondhand opinion and broad summaries of the research literature. AMA spokesman Edward Hill has stated that neither he nor anyone else on the AMA board was able to read the research before authoring the statement.
It’s also important to consider the typical lag time in the progress of institutional thought. While one body of research is being compiled and synthesized into a shared opinion, a very different one may be developing unnoticed in its wake. The studies cited by the Joint Statement were mostly designed in the 1950s and early 1960s and executed from the 1960s to the 1980s. Since that time, as we’ll see, media studies have taken a significant turn.
Finally, it’s important to remember the power of preconception. Social science research is rarely launched from the position of “let’s look at this and just see what we find.” Common practice is for researchers to begin by declaring the results they expect to find, in order to address questions of “researcher bias.”Reviewing the literature, I find in study after study that the researchers expected to find evidence of negative effects. The same preconception even determines who asks the questions and what they ask. “The conviction runs so deep in the academic and political communities that their most important mission is to warn of the dangers of media violence,” said Stuart Fischoff, “that it’s virtually impossible to get a large-scale study funded in this country unless it’s designed to look for harmful effects.”
What does the conventional research mean, then? It’s a mountain of material, but like most mountains it’s only intimidating when we’re standing in its shadow. Once we’ve climbed to the top of it, it rewards us with an exhilarating new view. We can see how our preconceptions have led us into a few recurring habits of misinterpretation, and we discover that although the research may not mean what we thought, what it does mean may be more useful.
The conventional research falls into three broad categories: correlative studies that compare the viewing habits of children who behave violently with those who don’t; laboratory studies of the ways children behave after being exposed to violent imagery; and field studies that try to apply lab-like tests to children in the real world.
It’s commonly accepted that the correlative studies have established a 10 percent correlation between media violence and heightened aggression. In fact, they vary wildly in their findings, from some that find considerably higher correlations to some that find none at all to others that find correlations between media violence and lower aggression. But the most famous of the early studies found approximately a 10 percent correlation, and that number has become truism in the years since.
The trouble with using such studies as evidence of media’s effect on children is that a correlation is not a cause. The researchers know this well, which is why all their studies refer to a “link” or “relationship” between violence-viewing and violence-doing, but never of cause and effect. The “link” may mean only that aggressive kids are more inclined to like violent entertainment. We all know that starry-eyed romantics like love stories, but few would argue that early and intense exposure to sappy melodrama causes a romantic temperament. Young people who are more aggressive, more restless, more angry at the world, or simply more inclined to enjoy roughhousing are also more likely to want to watch action shows, listen to angry songs, or play combative video games.
These correlations may demonstrate just the opposite of the media- as-cause hypothesis. Aggressive kids may be more drawn to violent entertainment precisely because they need a compelling alternative to acting out or because they want help making sense of their own aggressive feelings. Several studies support this possibility, although conventional interpretations of the data tend to ignore it. Dr. William Belson’s landmark study of violent teenage criminals is frequently cited for its discovery that those who watched violent entertainment committed a higher number of serious crimes than those who watched very little. Much less frequently mentioned is the fact that the ones who watched the most violence committed far fewer serious crimes than those who watched more moderate amounts (which may reflect a cathartic effect of violent entertainment or simply the fact that the more time a teenager spends in front of the tube the less time he has to get into trouble).
The one correlative study that has been cited again and again as support for the causal hypothesis is the famous longitudinal study that Leonard Eron conducted among a few hundred children in rural New York from 1960 to 1970. Dr. Eron found that among third-graders in 1960, the boys who liked violent TV (mainly The Three Stooges, 77 Sunset Strip, and assorted Westerns) accounted for about 20 percent of those considered “aggressive” by their classmates. Among those same boys in 1970, who were now out of high school, he found that the ones who had liked violent TV ten years earlier had come to account for about 30 percent of those now described by their peers as “aggressive” (hence, the famous 10 percent correlation). He concluded that watching violent television fostered an aggressiveness in children that came out in their late teens. As one reviews the literature, this particular study keeps turning up at the center of virtually every discussion of the purported long-term effect of media violence on children; it is practically the foundation of the argument that, in Dr. Eron’s words, “there is a probably causative effect of watching violent television programs in early formative years on later aggression.”
However, in the thirty years since the report’s release, social scientists have pointed out several serious problems with it. Subsequent studies along similar lines have failed to find any such correlation. The correlation holds for the boys in the study but, inexplicably, not the girls. The nine-year-olds’ viewing preferences were based not on observation or the subjects’ reports but on the opinions of their mothers. Most tellingly, Dr. Eron’s correlation showed up only when the boys were labeled “aggressive” or “nonaggressive” by peer nomination (asking classmates who they felt were the most aggressive kids in the class); when the study determined the boys’ aggressiveness by two other methods, objective personality tests and self-description, no correlation was found. Peer nomination is a demonstrably unreliable means of drawing conclusions about a child’s behavior. It is especially suspect in this case, in which the first peer group consisted of nine-year-olds in the macho milieu of 1960 and the second consisted of college-age adults immersed in the anti-violence ethos of 1970.
The study may support Dr. Eron’s conclusion. On the other hand, it may indicate that the second peer group–those young people in 1970 who were probably already steeped in media coverage of the dangers of TV violence–were predisposed to think that the kids who loved make-believe violence must be more aggressive, regardless of whatever behavior they actually displayed. Research, like a computer or a mythological oracle, answers only the questions we ask of it. It doesn’t tell us what we haven’t thought to ask. In this case, the only question asked by Dr. Eron and most of the researchers after him was, “Do the data show that watching violence leads to later aggression?”When we ask a different question–such as, “What do the data say about the ways young people label their peers?”–we get a very different answer.
This is the kind of ambiguity that runs through the conventional research but is almost never mentioned in the popular discussion of it. The social scientists who conduct the studies know that well; they admit that they’ve found no explanation for the causes of violence. In 1994, amid a flurry of academic summits and government hearings at which he spoke as a leading critic of media violence, Leonard Eron himself acknowledged that his research “cannot definitely settle the issue of causal direction. . . . It’s in the laboratory where you can actually show cause and effect.”
Jib Fowles, a professor and researcher at the University of Houston, described the typical behavioral lab experiment from a child’s point of view:
Selected as a subject, a child would have to be brought to a strange universe. . . .The setting is institutional, with hard surfaces and angles. There are none of the textures of a home nor the school’s familiar display of handwork. Other youngsters are also arriving, few of whom the child is likely to know, but all of whom are to comprise a novel social group to which the child must be aware and attuned. Around and above the children are adult strangers with clipboards who are in charge.
Now in a room with other unmet children, the child may be unexpectedly frustrated or angered by the experimenters–shown toys but not allowed to touch them, perhaps, or spoken to brusquely. The child is then instructed to look at a video monitor. It would be highly unlikely that the young child would sense that this in any way resembled television viewing done at home. At home, everything is known; here, everything is unknown, demanding attentiveness. At home, the child may be prone and comfortable, and viewing is nonchalant; here, the room is overlighted, the child is seated upright, and viewing is concentrated. Most signally, at home television viewing is an entirely
voluntary activity. . . . In the behavioral laboratory, the child is compelled to watch and, worse, compelled to watch material not of the child’s choosing and probably not to the child’s liking. The essential element of the domestic television-viewing experience, that of pleasure, has been methodically stripped away.
The youngster then views footage quite unlike what he’s seen at home: violent scenes snipped out of a show without a story to make sense of it, humor to relieve it, or the dramatic closure that ends nearly every TV show. Then he is told to play with the other kids, all strangers, all under stress, while the researchers watch. As for what to play:
There are typically only a limited number of options, all behavioral, for the young subjects. Certainly, no researcher is asking them about the meanings they may have taken from the screened violence.
In summary, laboratory experiments . . . are concocted schemes that violate all the essential stipulations of viewing in the real world and in doing so have nothing to teach about the television experience (although they may say much about the experimenters). . . . Laboratory research has taken the viewing experience and turned it inside out so that the viewer is no longer in charge. In this manner, experimenters have made a mockery out of the everyday act of television viewing. Distorted to this extent, laboratory viewing can be said to simulate home viewing only if one is determined to believe so.
The crux of the media-as-cause idea, the bolt that holds the argument together, is that because dozens of correlative studies show that aggressive kids watch violent entertainment, and because dozens of laboratory studies show that watching violent entertainment inspires kids to behave aggressively, we can assume that real-life aggression is being caused by real-life entertainment. However, one problem with experimentation on human behavior, as critics of lab research have been pointing out for decades, is that researchers are creating an abnormal situation and thus inducing abnormal reactions. A child choosing to watch Dragon Ball Z because he knows it will make him happy is having a fundamentally different experience from a child who doesn’t even like Dragon Ball Z being told, “You have to watch this now”–and his reactions will be just as different.
That may explain the results of the famous Coates-Pusser-Goodman study, which found that preschoolers were three times more aggressive after watching a video than before–even though the video was Mister Rogers’ Neighborhood. This led some analysts to conclude that television viewing itself, regardless of content, inspires violent behavior. It more likely means that being made by a strange adult to watch television makes a child anxious or angry. I love Fred Rogers, but I suspect if I were forced to sit in a hard plastic chair in a strange room and stare at him when I’d rather be out playing, I’d act aggressively too.
Social scientists are aware of such limitations, of course, and have tried to adjust for them in more recent tests. But, as most reviews of the literature have shown, attempts to gather evidence through more sophisticated tests have only led to more ambiguous results. As Drs. Haejung Paik and George Comstock (a pair of researchers generally sympathetic to the media-as-cause hypothesis) found in their metaanalysis of the literature in 1994, the more naturalistic a study is, the smaller the findings tend to be. For example, one research team in 1983 tried to compensate for the unpleasantness of lab conditions by treating their young subjects with extra generosity and pleasantry before showing one group a violent film, another group a nonviolent film, and a third group no film at all. Afterward, the children who watched the violent film behaved more altruistically and cooperatively than the children of the other groups.
When experimental methods have been taken out of the laboratory for field studies, the results have been similarly complex. Some have supported the media-as-cause argument, but there have been many like the Feshbach-Singer study, which followed boys watching TV in their own environs and found that whereas affluent boys were apparently unaffected by video violence, poor boys with records of delinquency seemed to be made less aggressive by it. Are such studies evidence for the cathartic power of entertainment, as some expert analyses have maintained? Or are they, when combined with all the other research, only more evidence that something as complex as children’s relationship with fantasy cannot be reduced to numbers in an experiment?
All of this research is based on the idea of studying children’s “exposure to violence,” in the same way that we might study their “exposure” to lead or cigarette smoke. But as appealing as that medical model may be, with its promise of clear, scientific answers, it just can’t tell us much about children’s feelings and behavior. Entertainment isn’t a chemical substance. A child’s imagination doesn’t behave like the cells of the body, with a predictable, somatic response. Playing a game or choosing a TV show is a conscious choice, an individual action, and part of a complex exchange between what a child needs and what the entertainment provides. If we studied the emotions of children as they were forced to kiss their loud, cigar-stinking Uncle Walters, we’d find plenty of hard, statistical data to show that kissing leads to fear, anxiety, and resentment. If we studied children as they willingly kissed their mommies good night, we’d come up with drastically different results. If we believed that kissing was harmful to children, we could surely come up with evidence to support that belief. Some psychologists, most famously John Watson, were in fact biased toward that belief in the 1920s and 1930s, and they did indeed find the evidence they were looking for. Fortunately for most of us, more compassionate views prevailed.
The correlative and laboratory data are valuable. They show us that there is a relationship between our children’s entertainment and their behavior, a powerful relationship with many implications. It is essential, however, that we understand how to use that data to help children in the real world. This is where our determination to find a simplistic cause-and-effect relationship keeps doing us in. For, not only do we misinterpret the research by trying to find that too-simple cause, but we also neglect even to define what it is that we think is being caused.
During the U.S. Senate’s hearings on entertainment violence, Senator John McCain brought the media-as-cause argument down to its most vivid essentials. He showed footage of preschoolers playing after watching Barney and again after watching Power Rangers. After the singing and dancing of the big purple dinosaur and his grinning juvenile sidekicks, the kids played fairly quietly and cooperatively–every tired parent’s dream. After the five hyped-up superheroes had finished their blasting and karate-chopping of city-smashing rubber monsters, the kids jumped up and ran around and yelled and wrestled and kicked at the air. That made it easy for Senator McCain to say, in effect, “Look how aggressive TV violence makes them!”That in turn made it easy for us to imagine those same kids beating each other up on the playground, then falling into gangs of delinquents, then opening fire on their schoolmates or mugging us on a dark street.
Except, of course, that they weren’t mugging or shooting or really even fighting at all. They were playing.
The benefits of rough-and-tumble play are well documented. It can be annoying for parents, it can get out of hand and lead to head bumps, but most authorities agree that it’s normal, healthy, and generally conducive to more confident kids. Profiles of violent adolescents don’t generally show any exorbitant amount of aggressive play early in life and, in fact, often show the opposite: violent teenagers often had trouble bonding with peers in normal childhood play. It may be that John McCain–war hero, maverick politician, and presidential contender–spent his whole childhood playing as quietly and cooperatively as the kids who’d just finished watching Barney in his video clips, but I doubt it. If he was anything like the other risk takers I’ve known of his generation, he spent his formative years watching Hollywood gun battles and playing at least as aggressively as any modern preschooler in the wake of a Power Rangers episode.
Normally, when we see kids playing superhero, most of us don’t see it as much of a danger. But when we fall back into anxious discussions of youth violence, we lose sight of that context. We suddenly see only a polar opposition of “aggression” and “nonaggression,” and against our own better judgment we start to see all aggression as part of one destructive continuum.
The cornerstone of all media-aggression experimentation was the Albert Bandura study of 1963, which showed that children who had watched films of someone punching an inflatable clown doll subsequently punched an identical clown doll more often than a group of children who had not. Hundreds of similar experiments, with similar results, have followed over the decades since then. Clearly there’s a basic truth to these experiments;we all know from experience that if kids see something exciting on TV they’re likely to imitate it in play immediately afterward. But we need to step back from the interpretation put on those studies by generations of commentators, get outside the terminology that speaks of “heightened aggression,” and look at what the kids are actually doing:
They’re punching an inflatable clown.
My mother bought me one of those clowns. It was probably the same year as the Bandura study–an irony she appreciated decades later, when I reminded her of how psychological opinion had turned her against violent TV and toys later in the 1960s. I loved that clown, until I got bored with it. I loved the feeling of exploding with all my tiny strength into its springy face, knowing that no harm could come either to the clown or my hand, and watching it bounce back with a grateful smile. That clown loved being hit, I was sure of it, and the bond he and I forged over my delight in cutting loose and his in being hammered was one of the most satisfying of my life for a week or two. I felt powerful. Later my mother told me that I was shy about punching it hard at first, until my dad stepped in to show me how. It was the best thing he could have done.
There is no evidence to suggest that punching an inflatable clown has any connection to real-life violence. There is no evidence that kids who love to punch inflatable clowns are more prone to playground aggression or later delinquency. There is anecdotal evidence that clown-punching is beneficial when it has any effect at all. And yet incidents of clown-punching have been jotted down by generations of researchers as evidence of “heightened aggression.”
More sophisticated experiments have attempted to isolate more antisocial, more “real,” aggression by asking children what they feel, in quest of aggressive thoughts. Others have induced young adult subjects to administer electric shocks to imaginary opponents. But the results, though they disturb us, reveal no more than the clown-punching data. As Dr. Richard J.Borden demonstrated in his study of such experiments, the subjects are always led to believe that whatever they do or say is ultimately safe, fun, and approved of–even desired–by the researcher. Still others have attempted a physiological approach, gauging subjects’ production of norepinephrine and other neurochemicals associated with aggression, or measuring activity in areas of the brain linked to painful or anger-inducing memories. Those too only record reactions in the stressful environment of the laboratory.
However, even if the measurements of all these tests match exactly what children experience in normal life, they may still be measuring the opposite of what the media-aggression hypothesis leads us to expect. What such experiments measure is general arousal.“When we are aroused,” says Stuart Fischoff, “we do everything harder–talk louder, walk faster, play harder, argue more intensely–no matter what the stimulus is.”All emotions are intensified; if aggression is the one we’re measuring, then we’ll find aggression intensified. One study found that a group of adults who watched a violent movie were more aggressive afterward than those who watched a quiet movie. But those who watched an erotic movie were more aggressive than those in the other two groups. Since physical action usually arouses children, it will also elevate their aggression. It will probably also make them laugh louder, jump higher, and think faster–but we’re not in the habit of looking for such positive effects.
General arousal can make someone who is already inclined to be aggressive more likely to act out. When action entertainment is what someone uses to arouse himself, then that entertainment does contribute to a higher likelihood of aggression. But arousal also ebbs quickly and, in doing so, can relieve tensions and leave people more relaxed. Those experiments with clowns, questionnaires, and electrodes are frequently cited as evidence that the catharsis theory does not apply to media: if children play more aggressively after watching aggressive images, then surely the images are stimulating their aggression rather than aiding its release. But, as anyone who’s paid attention to children watching TV and then playing should know, the watching and the playing aren’t two separate events but steps in the same process. Catharsis requires the emotions to be stimulated before they can be released. Just watching a video usually won’t bring about a release of tension or anger, but many times I’ve seen a roomful of kids get wound up over a Pokémon scene and then jump off the couch and reenact it physically until they’re tired and relaxed. I’ve gone to action movies with sullen middle schoolers, seen the big-screen explosions and the seat-shaking Dolby sound effects shock them out of their funks, watched them bop around chattering happily about it afterward. Kids play through the images in their heads, try on what they’ve seen to make sense of the emotions, expel their energies through acts and words.
Many studies find raised norepinephrine levels in young people right after they’ve played violent video games. Similar conditions, however, are found in people right after playing a close tennis game or finishing a quick sprint. Relaxation comes later. Few experimental studies have bothered to look beyond the immediate aftermath of stimulation by make-believe violence, but those that have show important results. Penny Holland’s recent studies at British preschools found that when kids were allowed to play with toy guns, their games became more “aggressive” in the short term but that “the atmosphere in the room was notably more relaxed later in the day.”
Yet, even the question of whether a catharsis is achieved is secondary. The primary question is this: if mock-aggressive play is good for children, and if these experiments demonstrate that entertainment violence inspires children to engage in mock-aggressive play, could it be that the experiments actually demonstrate that entertainment violence is good?
Some field and correlative studies look only at acts of serious violence, but many cast the word “aggression” over very mild and even desirable behaviors. A report may refer to a five-year-old yelling at her friend for taking her stuffed animal as an incidence of “aggression.” It is aggression, of course, but what meaning are we taking from this? A child yelling about an injustice may not be showing the best problem-solving skills, but surely such “aggression” is far healthier than passivity or silent resentment. I think of the “peer nomination” used in so many studies, and I wonder if some of my most successfulpeers would have been branded “aggressive” if our class had been interviewed as third graders.
Even if the research means just what the conventional interpretation says it does, and entertainment violence does lead to increased aggression in 10 percent of children, we still don’t have a complete picture. What about the other 90 percent? Do we ignore their needs simply because they aren’t becoming more aggressive? What if even the 10 percent who are experiencing heightened aggression are picking up psychological or emotional benefits along with it? What if most of them are channeling that aggression into self-assertion, healthy competition, increased energy, determination, and courage? By failing to consider the meaning of children’s behaviors, we do worse than render the research useless. We risk reading it upside down, seeing only negatives where we should also see positives, and so taking tools from children that may be helping them deal with the very stresses that concern us. That’s why arguing about the “validity” of the research will never be anything but a pointless cycle. Ask of anything powerful, “Can this be harmful sometimes?” and we’ll come up with a “yes.” If we ask no more than that, the answers will never take us to the whole truth.
If we keep asking the same questions, we’ll keep getting the same answers. We’ll keep falling back on our usual public responses, the old familiar calls to limit the quantity, explicitness, or availability of entertainment violence. Caught in our repetitive arguments, we forget that we’ve been here before. We seem to have forgotten, in fact, one of the most important lessons of the recent history of American popular culture.
In the late 1960s, as crime rates were rising and the war in Vietnam preoccupied us, a groundswell of sentiment against violent entertainment actually succeeded in profoundly altering the landscape of children’s culture. With help from the Federal Communications Commission and the grudging cooperation of the networks, groups such as Action for Children’s Television succeeded in chasing most of the violence out of kids’ programming. The generation of cartoons created during the 1970s, from The Smurfs to Strawberry Shortcake, were designed to emphasize pro-social values and eschew slapstick humor and physical conflict. Even the superheroes of Super Friends did little more than whip up whirlwinds and discuss the importance of peaceful cooperation. Old Bugs Bunny and Tom and Jerry cartoons were edited so that kids would never see anvils hitting heads or cats slamming into doors.
Prime-time action shows like The Incredible Hulk and The Dukes of Hazzard (listed as two of the most violent shows on television in one late-1970s study) featured little action and not a single instance of bodily harm. The mighty Hulk had to content himself with tearing the bumpers off cars, smashing through doors, and sometimes knocking a bad guy into a swimming pool–and even then in only two action scenes per hour-long episode. The old generation of Westerns and cop shows dribbled away, retreated to later time slots, and curtailed their on-screen violence. Although some reruns of shows like Star Trek and Wild, Wild West continued, the grittier older programming was mostly shelved. And there were no cable channels to show anything else.
Violent movies were grown-up affairs then, the likes of The French Connection and Straw Dogs. In those early years of the Motion Picture Production Association code, violence typically brought R and even X ratings, and industry surveys found that theater managers usually enforced the ratings. Kid-oriented movies in the decade before Star Wars were almost religiously nonviolent. Rock music took heat for promoting sex and drugs, but in that post-flower-child era it did anything but glamorize violence. The comic book business had shrunk to almost nothing. The toy industry, which received as much anti-violence criticism as television, turned its back radically on anything suggestive of killing or warfare. Guns, swords, and even ray guns disappeared from the shelves of reputable toy dealers. G.I. Joe became a peacetime “adventurer” overnight without a weapon to his name. The Louis Marx Company, which had dominated the plastic-army- men business all through the 1960s, abruptly dropped its whole line of soldiers. “Action figures” didn’t exist yet, and neither did video games.
We tend to forget now, but for about a decade not very long ago, we truly did give our children the nearly violence-free popular culture that so many critics of the media are pressing for now.“When I was a kid I thought ‘action’ meant Boss Hogg’s car smacking into a hay bale,” says Adam Veeck, a writer born into a liberal home in 1967. “Gradually I pieced together that an earlier generation of kids had had all these awesome war toys and cinematic fistfights. All I could think was what a rotten deal I’d gotten.”
What happened during those years? Crime rates increased. Our national anxiety about violence, as measured by opinion polls, worsened. The kids who spent their formative years in that pop-cultural milieu became the teenagers of the mid-1980s, when juvenile crime rates rose again. The kids who spent their formative years in the 1980s, on the other hand, when action-packed movies, TV shows, video games, and combat toys seemed to be taking over kid culture, became the teenagers of the late 1990s, when those rates plummeted. Obviously, the Smurfs were no more responsible for the crime wave of the 1980s than the Teenage Mutant Ninja Turtles were for the relative calm that followed. But this does cast some doubt on the real-world value of focusing our efforts on restricting media violence.
Something else happened during those years: children’s appetite for make-believe violence seemed to grow. When Star Wars brought fighting, blasting, killing, and the blowing up of big things back to children’s media after nearly a decade without, it hit levels of success never before approximated in juvenile entertainment and launched an action-driven entertainment/toy industry vaster than anyone could have dreamed. It’s almost as if the suppression of entertainment violence drove it to come back more aggressively than ever.
The Joint Statement on the Impact of Entertainment Violence on Children called for “a more honest dialogue about what can be done to enhance the health and well-being of America’s children.” Our children need that dialogue. They deserve it. But it can only truly begin when we come down from that mountain of old research and start asking new questions. Why do so many healthy, nonviolent children love violent stories so much? Why are so many of them excited to such happy play by make-believe combat and destruction? What do their entertainment choices say about their emotional needs? What’s going wrong when they act out destructively? How can we help them make more sense of what they see?
Most fundamental of all, why do they love what they love?