Media & Violence
by Marjorie Heins
Paper Presented at the New York Hilton
International Conference on Violence in the Media
October 4, 1994
Last year, you may recall, Attorney General Janet Reno warned television executives that if they did not start seriously censoring violent content in TV programming, the federal government would do it for them. Reno blithely asserted that a federal censorship law would not violate the First Amendment. When asked what kind of violent content would be censored -- for example, football games? -- she replied that no, she would exempt sports, since they provide a healthy channel for aggressive impulses.
This little vignette nicely illustrates the problem with even trying to define what we mean by media violence, no less with attempts to suppress it. Is violent football harmful? More harmful than violent fiction or news reporting? Who decides what's a healthy channel for aggression and what's a public health menace? Should the violent Rodney King video be censored, or violent war movies? Should we only allow violent depictions where the aggressor is punished?
The multitudinous social science studies and reports that are so frequently relied upon to justify demands for censorship are all over the lot on this question. They have so many different definitions of violence or aggressive behavior (many include verbal aggression, rudeness, and so forth), that it's impossible to tell what's being measured from one study to the next. But this doesn't stop Ms. Reno and countless others from jumping on the censorship bandwagon and targeting "TV violence" as the latest demon to be blamed for our societal woes.
Scapegoating speech -- whether on TV, in movies, in popular music, comic books, detective novels, high-toned erotic art, or sleazy pornography -- is a perennial American pasttime. Its psychological roots are complex and deep. Politicians and direct mail fundraisers love it: they can wax indignant about images and symbols, gain headlines, dollars, and votes, without having done anything serious to address our social problems. (Academics also seem to love it: witness the seemingly interminable flow of research grants and conferences on the subject.) Cartoonist Don Wright said it all in a comic strip published last October in _The Palm Beach Post_:
- Ms. Reno, what really happened in the Waco disaster?
- Beavis and Butthead did it.
- No jobs!
- Beavis and Butthead did it.
- You shot people? . You're pregnant? . Skin cancer?
- Beavis & Butthead did it.
There is nothing new about the phenomenon of scapegoating art and entertainment, and if I had more time, I would treat you to a leisurely history of the subject, from 19th century bluestocking Anthony Comstock, with his New York Society for the Suppression of Vice, who not only prosecuted birth control pioneer Margaret Sanger and other sexuality and reproductive rights experts, but also went after "blood and thunder" publications that ran graphic crime and sports stories; through the Hollywood Production Code, which banned depictions of crime that "might arouse sympathy or inspire emulation"; to our own Tipper Gore and the Parents Music Resource Center, complaining in the mid-1980s in U.S. Senate hearings that rock music promoted sex, violence, profanity, drug use, suicide, and the occult. In between, of course, we had the 1954-55 U.S. Senate Comic Book Hearings, where psychiatrist Frederic Wertham attracted political and media attention with his scientific-sounding claims that exposure to comics led to a life of delinquency and crime.
Television violence has also been blamed for social ills at least since 1952 when Congress held its first hearings on the subject. Even before we had experts armed with research grants and multiple regression analyses wading into the picture, mental health professionals such as the Mayo Clinic's Dr. Walter Alvarez blamed delinquent behavior on TV pictures of "murder, violence, shooting, hanging, kidnapping, and the doings of wicked witches."1/
It is now thirty years and many studies later -- though not 3,000, as the organizers of this conference and others have claimed. (Many articles on the subject simply review or interpret the hundred or so actual independent studies; other articles included in the inflated estimates concern television generally, not media violence.)2/ Often, one is accused either of hopeless bias or pathological denial syndrome if one dares take issue with the now supposedly well-established "proof" that media violence causes aggressive behavior. I was sufficiently intimidated by this presumption (so well reflected in the brochure for the present conference) that it wasn't until this summer that I finally sat down and went through some of the social science. And to tell you that the "experts" disagree, and that the literature is ultimately inconclusive on the actual question of "causation" would be to understate the case.
Before the 1960s, social scientists generally viewed entertainment as having, if anything, a cathartic effect on viewers, as Aristotle explained in his _Poetics_ more than two thousand years ago. Aristotle's theory was that the audience to a creative work experienced catharsis through feelings of pity and terror; this was translated by social science jargon as the "drive-reducing function of fantasy"3/ or as "vicarious hostility reduction."4/ In any case, fantasy or other expressive material was not thought to inspire imitation. As the authors of one early study put it, angered audience members may establish a strong association between their own frustrators and the fantasy victim of aggression. Theoretically, at least, their instigation to aggression should lessen to the extent that they associate the attacked -- and injured -- person on the screen with their own tormentors. But in addition, the film . may make audience members forget about the thwartings they have suffered. Carried away by the movie, not brooding on the injuries done them, their anger might well subside fairly quickly. Their instigation to aggression has weakened because they have not kept themselves stirred up."5/
This changed in the 1960s with the emergence of "social learning theory." Laboratory studies, involving specially created films, Bobo dolls, and other paraphernalia, were said to establish a causal connection between aggressive material and aggressive behavior among children. That is, children exposed to the film or cartoon showing an adult hitting a Bobo doll were shown to be statistically more likely to imitate the adult afterwards, under laboratory observation (also after being frustrated, and with none of the usual social constraints on aggressive behavior), than were children in a "control" group. The lab scientists with their carefully orchestrated experiments now announced that they had disproved Aristotle!
But as many scholars have pointed out, the lab setting made the variables much easier to control than in real life; and the experiments were questionable for just this reason. In the scientists' words, the "artificiality of the circumstances, . and the brevity of both the TV exposure and the effects being measured,"6/ the "methodological elegance" of the lab studies, their ability to manipulate the variables being tested, was precisely what made the results dubious. In the inimitable language of the field: "methodological elegance is often inversely related to ecologic validity."7/ Indeed, most researchers will readily admit that laboratory results cannot be extrapolated; what happens in a lab doesn't necessarily happen in real life.8/
The early lab experiments nevertheless formed a primary basis for two influential reports: the 1972 Surgeon General's Advisory Committee on Television and Social Behavior,9/ and the 1982 survey sponsored by the National Institute of Mental Health.10/ Although often touted as supporting the view that violent TV was a national health problem, the reports were actually much more circumspect. The Surgeon General's report noted only a "preliminary and tentative indication" of a causal relation between TV violence and the real-world variety; and said: "The effect is small compared with many other possible causes, such as parental attitudes or knowledge of and experience with the real violence of our society."11/ The report acknowledged that any causal relation operates only on children already predisposed to aggression; the evidence did not "warrant the conclusion that televised violence has a uniformly adverse effect . [or] an adverse effect on the majority of children."12/
In short, the lab studies showed at most that aggressive material may sometimes induce object aggression (Bobo dolls, popping balloons), symbolic aggression (toy guns) or verbal aggression -- but not actual physical aggression against another person.13/
Because of the lab studies' "modest external validity"14/ (i.e., very limited applicability to real-world behavior), scientists began to fashion field, correlational, and longitudinal studies to test the presumed pernicious effect of aggressive material in the media. But the results were also inconclusive and often contradictory.15/ As one article reported, any statistical relation between aggression and TV violence viewing almost completely disappears when corrections are made for such other factors as a child's intelligence and preexisting level of aggression; the social learning hypothesis could not be supported in field experiments.16/ Or as another survey of the lab, field, and longitudinal literature concluded in 1986: "As convenient as it would be to have one of the `causes' of criminal behavior neatly pinpointed, the research evidence to date does not allow this conclusion."17/
Attempts to find a statistical correlation between television viewing in general and antisocial behavior also yielded mixed results. Some studies compared the crime or homicide rates in certain areas before and after the introduction of TV. (The pre-TV communities were coyly designated "television virgins.") One much-touted study found an increase in the South African homicide rate after TV's introduction in the 1970s, and claimed that other likely explanations (urbanization, economic conditions, alcohol consumption, availability of firearms) did not account for the change.18/ Whether one or a variety of other political and social factors not considered by this experimenter might yield an explanation for the South African homicide rate increase during those years (for example, increased opposition to apartheid, and consequent police repression), one might question the scholar's eagerness to blame it on TV. Computers, Burger Kings, Pizza Huts, and Hondas also were probably introduced in South Africa around this time, yet no one suggested that they caused the increase in reported homicides.
Moreover, this was a correlational study on the effect of TV in general, not on any particular TV content. It was criticized for vastly overstating the statistics.19/ Nor could its results be consistently replicated. Another study found a much _lower_ rate of aggressive behavior among sixth graders in a Canadian community _with_ television than in a "television virgin."20/ Yet proponents of censorship still cite this discredited study as proof of their claims.
Common knowledge tells us that many of the most violent societies don't even have TV, while highly industrialized countries like England, Japan, the Netherlands, Norway, Switzerland, Germany, and France, with plenty of TV, have far lower homicide rates than the United States.21/ Could it be that social dysfunction, poverty, drug wars, and the wide availability of firearms make the difference?
One of the great problems with correlational studies, of course, is the chicken-and-egg dilemma. That is, correlation is not causation. Does viewing violent television (or any television) _cause_ some people to behave aggressively, or do aggressive individuals simply prefer more violent entertainment? Or are there third factors, like educational deficits or family dysfunction, that account for both aggressive behavior and a preference for violent television?
One longitudinal study (a five-year follow-up on 732 children's behavior and TV-watching habits) concluded that heavy television viewing is primarily related to social and cultural disadvantage, and preexisting behavior disorders, including aggression; the fact that some youngsters prefer more violent television does not show that TV viewing causes their behavioral problems.22/ Another set of scholars concluded in 1989 that no causal effect of television violence on behavior could be shown, but sugggested that "genotypic differences in psychoticism" (i.e., predisposition toward violence in some people) accounted for both real-world aggression and preference for TV violence.23/
Another longitudinal study found that once youngsters' "starting level aggression" was taken into account, the numbers provided no support for the hypothesis that viewing violent TV over the long term causes more aggression. A much more significant predictor of both heavy television viewing and aggressive behavior was children's level of intelligence.24/ Similarly, a study of elementary and high school children found that socioeconomic status, mother's occupation, and father's use of physical punishment, were the prime causal influences on both violent TV-watching and aggressive behavior.25/
One of the more famous longitudinal studies, by Leonard Eron (mentioned yesterday by Congressman Owens),26/ was said to show that people with a childhood preference for violent television turned out to be more violent as adults. But Eron's work has been widely criticized for faulty logic, reliance on slim and inconclusive results, failure to explore possible other factors responsible for violent behavior, use of unreliable self-reporting by parents, inability to keep track of many of the subjects during the intervening years of the study, and failure to explain the lack of effect of TV exposure on girls.27/
Reviewing the social science literature through 1986, Yale Professor William McGuire concluded that many lab and field studies found no real-world effect of TV violence, and those that did reached only a very small level of statistical significance. In correlation studies, said McGuire, the differences were trivial.28/ Similarly, in 1990 the Department of Education concluded after an extensive survey that "a disturbing amount of scholarship had been slipshod or influenced by a prevailing attitude that TV is harmful."29/
One of the less frequently cited studies showed that emotionally disturbed children, who the experts thought would be less capable of processing fantasy information and more likely simply to imitate behaviors seen on TV, in fact responded to nonviolent programming ("Lassie's Rescue Rangers") with more aggressive behavior than when they viewed extremely violent cartoons like "Tom & Jerry," "Bugs Bunny," and "Woody Woodpecker."30/ The surprised researchers hypothesized that the nonviolent Lassie drama, for all its social uplift, caused stress and anxiety because of its suspenseful plot. The conclusion: "commonsense notions about what is harmful or innocent television fare for children are not always on target . forcing a wholesome television diet on children may be counterproductive."
Perhaps the most interesting study, though, showed an almost threefold increase in aggression among previously low-aggression preschoolers after watching . "Sesame Street" or "Mr. Rogers' Neighborhood"! 31/
Even if the social science literature were not so inconclusive, the real problem with using it to develop censorship schemes is that social science proceeds on very different assumptions from the principles of human freedom and responsibility that are embodied in the First Amendment. Blaming the media for our actions reflects a reductionist, determinist view of behavior, rather than a vision of human beings as unique, autonomous, endowed with sensibility and understanding, each processing information and experiences in her own way, and each responsible for her own acts and opinions.
Social science deals with generalizations and probabilities, not evaluation of particular works of art and the responsibility of their creators for their multifaceted impact on particular individuals. Social science finds significance in relatively small statistical difference -- for example, five people out of a hundred -- while the majority of people in a study show no change in behavior or attitude as a result of watching even a steady diet of gore. (As Professor Gerbner mentioned yesterday, even those who think there's a causal connection -- as opposed to simply a correlation -- between viewing violence and acting it out, put the magnitude of the causative effect at about 5% -- hardly a compelling justification for depriving the entire population of ideas, information, or creative works.32/
Thus, for example, viewing a TV documentary about suicide (or a production of _Romeo & Juliet_) might trigger a statistically significant number of teenagers who were already suicidal; but the vast majority of viewers would not take their own lives in any event.33/ For those few who do, their predisposition was sufficiently strong that _Romeo & Juliet_ can hardly be blamed for their demise.
In any event, censoring _Romeo & Juliet_ is neither constitutionally acceptable nor an answer to reducing teenage suicides even among the already predisposed. And if you think my example is far-fetched, it isn't. _Romeo & Juliet_ has been a target of censorship efforts in public schools not only because of its double suicide but because it allegedly encourages other violence, teen sex, and disobedience of parental authority.34/
The First Amendment guarantee of free speech assumes this principle of human individuality and responsibility. In general, speech is punishable only where identifiable individuals are harmed in specific, tangible ways -- e.g., defamation, invasion of privacy -- or where the speech in question is really a verbal act, e.g, committing perjury, bribery, or extortion, threatening somebody with bodily harm, firing someone from a job for discriminatory reasons.
Obscenity is an exception to the First Amendment that admittedly doesn't fit into this scheme, is really not justifiable on any legal or analytical ground, and is an oddity of constitutional law that perhaps can be best explained via Anthony Comstock and his English Victorian counterparts and our long Anglo-American tradition of schizophrenia about sex. To extend the obscenity exception to violence, however, as some urge, would be a project so amorphous and vast as to leave little of "the freedom of speech" in place.
In this area (violent content), the Supreme Court has settled on a line that protects all communications, including those that may be thought to promote violence or otherwise convey bad ideas, unless a particular communication amounts to actual, direct incitement to specific violent or criminal acts.
Drawing the line at actual incitement is critically necessary to a free society. In the old days before the incitement standard, we had the so-called "bad tendency" test in First Amendment law. If speech (usually radical political speech) was thought to have a tendency to induce bad thoughts or behavior, it could be suppressed. This was a great tool for repression, since the judgments of prosecutors, judges, and other government officials as to what speech has a bad tendency will pretty predictably favor the status quo and disfavor critics, protesters, muckrakers, and other outsiders whose speech threatens those in power or seeks social change.
As Congressman Rangel reminded us yesterday, without the freedom to televise violence, including excessive and sadistic and racist violence, the American public might never have been galvanized to support the civil rights struggles of African Americans in the 1960s. The same might be said of napalm and other atrocities during the Vietnam War. (Compare and contrast the antiseptic, government-censored coverage of the Persian Gulf War.) Do we really want to depend on the Federal Communications Commission to decide whether depictions of napalm, or police dogs in Birmingham, is "good" or "bad" violence? The last thing this society needs is for government to decide how much and what type of media violence its citizens should see.
What's so dangerous about present-day demands for censorship, whether of "pornography" or of "media violence," is that they quite explicitly advocate a return to the bad tendency doctrine, this notion that expression should be silenced if it's thought to reflect or promote bad ideas. There's no quicker road to thought control and loss of freedom than this philosophy. As a federal court of appeals wrote in 1987 (dismissing a parent's claim that a magazine article on autoerotic asphyxiation caused her son's death): "If the shield of the First Amendment can be eliminated by proving . that an article discussing a dangerous idea . helped bring about a real injury simply because the idea can be identified as `bad,' all free speech becomes threatened."35/
Or as Judge Frank Easterbrook wrote in the federal appeals court decision striking down an anti-pornography ordinance authorizing civil actions to ban "sexually explicit subordination of women, whether in pictures or in words": "any other answer leaves the government in control of all of the institutions of culture, the great censor and director of which thoughts are good for us."36/
There is another equally serious constitutional problem with giving government the power to censor speech that it thinks contains bad ideas -- or even "causes" some small portion of the population to be behave aggressively. This is the famous constitutional vice of vagueness.
A December 1993 report from North Dakota Senator Byron Dorgan -- a censorship advocate -- nicely illustrates the point. Dorgan's report summarized the results of a one-week survey of violence on prime time television conducted by college students earlier in the year. Among the shows found to contain the highest number of violent acts per hour were PBS' "The Miracle Worker," "Civil War Journal," "Star Trek 9," "The Untouchables," "Murder She Wrote," "Back to the Future," "Our Century: Combat at Sea," "Teenage Mutant Ninja Turtles," and -- my favorite -- Alfred Hitchcock's classic thriller "North by Northwest." This list ought to give some pause to those who would endorse Rep. Owens' idea that we can identify bad programming simply by counting the number of violent incidents a show contains. As Human Rights Watch Associate Director Gara LaMarche recently wrote,
Just as I don't trust government censors to make a distinction between sex education material and "hardcore" pornography, I don't trust them to sort out _Schindler's List_ from _Nightmare on Elm Street_, or whatever it is that [those advocating censorship] would like to curb.37/
The point could be made many times over: Are all horror films to be banned, or made off-limits to minors? Are _Bonnie & Clyde_, _The Accused_, _The Burning Bed_, and _Psycho_ to be banned on the theory that they encourage the very violence they so graphically depict? Do we really trust censorship boards to decide whether any work, from _The Accused_ to _NYPD Blue_, carries a politically correct message condemning rape or mayhem, and whether the most vulnerable and least media-literate in society are likely to see it that way?
War stories from _The Iliad_ to _Apocalypse Now_ are endemic in our culture, and crucial to our understanding of our history. From the atrocities in Bosnia to police brutality at home, news, docudrama, and fiction are replete with violence, all of it arguably more anxiety-producing and desensitizing than Bugs Bunny, James Bond, or _True Lies_, works that certainly might be said to employ excessive aggression and trivialize its harms -- or, in Professor Gerbner's description yesterday, to constitute "happy violence."
Of course, the Bible has been cited more often than any other work by psychopaths and other misfits as justification or inspiration for violent behavior, from wife and child-beating to mass murder.38/ The fundamental point is that, short of actual incitement (the proverbial false shout of "fire" in a theater), each of us responds to media messages in a personal idiosyncratic way that is driven by our own individual psychology, upbringing, and other life experiences. The influence of any work depends on how it is mediated by the mind of each viewer.
One of the most thought-provoking studies in the field (co-authored by yesterday's speaker, Professor Gerbner) suggests that the pervasiveness of television in our lives, including television with violent content, produces an effect quite different from straightforward imitation of the acts portrayed on screen. Instead, television primarily teaches certain "concepts of social reality," in particular that the world is violent and frightening; it fuels feelings of anxiety, insecurity, and mistrust.39/
This sense of "danger, vulnerability, and general malaise" in turn invites exploitation and repression. That is, fearful people are "more dependent, more easily manipulated and controlled, more susceptible to deceptively simple, strong, tough measures and hard-line postures --both political and religious. They may accept and even welcome repression if it promises to relieve their insecurities and other anxieties."
Perhaps, then, the most pernicious effect of television is not that it makes us violent but that is makes us neurotic, so much so that we start demanding censorship of . of course, television.
Prof. Gerbner suggested a remedy for the problem: demand diversity and access; break up the oligopolies that, he argues, pose a greater threat to free speech at the moment than government. Or, "more speech," as Justice Brandeis said long ago. Given the compelling good sense of this proposal, it's particularly depressing to see that these days our few alternative electronic outlets are shrinking: PBS going corporate and noncontroversial; Congress passing legislation to destroy the "electronic public forum" of public access cable television.
As you can see, I do not claim that television has been a positive influence on society, or on our cognitive faculties. In addition to perhaps making us anxious, it makes us stupid. (Not for nothing is it called the idiot box.40/) This is due partly to its sound-bite style, and partly to what the social scientists call "displacement": i.e., TV has displaced more mentally demanding activities like games, hobbies, reading, socializing, creative thinking, fantasizing, and even sports.41/
This is actually consistent with one of the strongest findings from our social science friends; i.e., whatever minor, marginal impact TV, or violent TV, might have on aggression in some small number of predisposed individuals, a major determinant of both heavy television viewing and antisocial behavior is educational deprivation and violence in the home. Would not our government resources be better spent, then, on combatting domestic violence and rebuilding our crumbling public education system (including programs in cognitive thinking and media literacy), than on constitutionally dubious and socially ineffectual censorship schemes? Americans need to find the political will to undertake these real-world solutions, and stop blaming the book, the movie, or even the idiot box.