The Evidence
Home Up Good Movies The Evidence Cartoon Research War Toy Research Sports Violence D & D Deaths



TV and Other Forms of Violent Entertainment: 

A Cause of 50% of Real-Life Violence

by Thomas E. Radecki, M.D., J.D.

Over 1000 research studies and reports have been published on the issue of violent entertainment and its impact on stimulating human aggression. The earliest research dates from 1933 and is correlational in nature. However, with advances in psychological research techniques, numerous laboratory and field studies looking at the impact of violent entertainment have been able to document a cause-effect relationship (Siegel, 1956; Bandura, 1956; etc.).

In 1982, an expert panel of seven leading aggression and television aggression researchers appointed by the U.S. Surgeon General determined that the evidence was "overwhelming" that television violence increased the tendency towards anger and aggression in the viewing public. This panel, when challenged by psychologists under the hire of the American Broadcasting Company, stated that the scientific evidence that TV violence has a harmful effect on viewers is as strong as the evidence that cigarettes cause lung cancer.

In a 1983 report for the U.S. Department of Justice, Thomas Cook, an expert in psychological research methodology but not an aggression research concluded after an extensive study that virtually 100% of aggression research now agree that there is a cause-effect relationship between the consumption of violent entertainment and an increased tendency towards anger and aggression. He stated that the only legitimate area of debate now is the extent of the impact.

In 1984 the U.S. Attorney General's Task Force on Family Violence concluded that the evidence "was becoming overwhelming that both normal adults as well as children" are harmfully affected by TV violence. The Task Force concluded that one of the effects was an increased tendency towards interpersonal aggression.

Dr. George Comstock of the Center for Research on Aggression at the University of Syracuse conducted a meta-analysis of over 230 controlled studies of the impact of TV and film violence. He concluded that a minimum of 10% of aggression in the violence viewing audience of the various studies was attributable to the violence viewed. He emphasized that this was the minimum impact and the actual full impact of violent entertainment is bound to be greater.

Evidence is presented in this paper that a reasonable estimate of the full impact of TV violence on American society is an increase in real-life violence by at least 50% above where it would otherwise be.

Violence Trends in History

Ted Robert Gurr in the book Violence in America notes that epidemics of domestic violence in European history since 1200 have always been linked to periods of war. Epidemics have had increases in violence by as much as 300%. However, these epidemics have always passed as peace was restored. Historians attribute times of war as being times when violence is glamorized. Gurr notes the current world pandemic of violence, the first in world history, can be attributed to a fantasy war of violence being spread in TV and film entertainment. He details dramatic increases in violent crime in the U.S. starting after 1957 with an ongoing increase through 1974. Violent crime temporarily reversed itself slightly but has continued a much more gradual climb since 1978. Violent crime has increased each year in the U.S. again since 1984 although the increases have been smaller in extent than the rapid growth in the 1960's and early 70's.

Television first went nationwide in the U.S. in 1952 with the average American obtaining a TV set in 1956. Television had an estimated 8% violent entertainment in the early days with no program in the top 20 in popularity high in violence. In 1956, with the introduction of the adult western in prime-time, TV violence took off in popularity and the percentage of prime-time programs with themes high in violence never again decreased below 30%. 1956 was also the year when the violence epidemic in the U.S. first got underway. Violence had decreased throughout the 1800's reaching its lowest levels between the 1930's and 1955.

Since 1955, the levels of murder in the U.S. have increased by over 200% per capita with rape and other crimes increasing by over 500% per capita. The violence epidemic in Canada started in 1960 and experienced a 400% per capita increase in murder and 1000% increase in attempted murder. Perhaps medical advances have helped restrain the increases in murder rate in comparison to other crimes.

An early study documented an increase in petty larceny in the U.S. in 1952 with the spread of the TV signal nationwide by using cities which had television in 1951 as control cities. However, anthropological research suggests that the problem of violent entertainment predates history. Sipes (1973) compared 10 violent and 10 non-violent cultures which had been studied by anthropologists in the early 1900's. These were cultures isolated from the world's mass entertainment. Sipes studied the sporting entertainment of the cultures involved. The only violent culture that did not have a sizable level of violence in their sporting activities was an African culture at war virtually 364 days a year that didn't have any sporting activity. Only two of the ten non-violent cultures had the least suggestion of violence in their sports (e.g. a log-rolling contest where contestants attempt to knock the other opponent off with a pole). Sipes speculates that these two cultures had been at peace only 150 years and that there had not been enough time for the violence to totally die out.

Violent entertainment is not a new phenomenon or concern. However, because of television and more recently VCR and cable technology, the number of hours of violent entertainment per capita has increased from perhaps 2 to 3 hours per week to 10 to 12 hours. The intensity of the violence has also increased.

Field Studies of TV Violence

Dozens of field studies have examined the impact of violent entertainment on normal children and adult viewers. Most of the studies fail to be able to measure the full short-term and long-term impact of violent entertainment because of their research designs.

Perhaps the most impressive study that attempts to measure the full-impact of violent entertainment is that of Centerwall (16). Brandon Centerwall of the University of Washington Department of Psychiatry took an epidemiological approach. He measured homicide rates in populations of white European origin in the U.S., Canada, and South Africa. South Africa was chosen because television was not allowed into the country until 1975 due to fears by South Africans of Dutch descent that it would give South African of English descent an unfair advantage.

Centerwall found that per capita murder rates soared 100% both the U.S. and Canada roughly 15 years after the introduction television and, as previously noted, starting at the time of the increase in violent programming in both countries. No increase in the per capita murder rate occurred in South Africa. A subsequent report noted that South Africa, while experiencing no increase in the white homicide rate from 1945 to 1975, had a similar 100% increase in the 10 years of time after the introduction of television and TV violence. Centerwall examined multiple other possible explanations for the time of these increases and was unable to find any except of violent entertainment.

Centerwall states, "Television is a factor in approximately 10,000 homicides each year in the United States. While TV clearly is not the sole cause of violence in our society, and there are many other contributing factor, hypothetically if television did not exist there would be 10,000 fewer homicides a year" (Seattle, AP 4/8/89). It is worthwhile noting that some amount of violence was already present in American films and sporting entertainment before the advent of television. It is virtually certain that this glamorized violence also had some impact on society. Thus, the full impact of violent entertainment would have to take this element into account as well. Roughly 15% of best seller books, 4% of radio time and 40% of Hollywood films featured themes of glamorized violence. While the violence was less intense than much of the modern horror and war film violence of the present, intensely glamorized violence has been an element in Hollywood entertainment since Thomas Edison's first film, The Great Train Robbery.

A series of eight field studies by Leonard Eron and Rowell Huesmann have looked at the long-term impact of violent entertainment on children. Three-year studies in Chicago public and Catholic school and in schools in Finland, Israel, and Poland found evidence of harmful cause-effect relationships between the preference for and consumption of violent TV entertainment and an increased tendency towards hostile aggressive behavior in normal boys and girls(15). In the Netherlands and Australia, a positive correlation was found, but the research could not demonstrate a cause-effect relationship.

The longest study ever done on the impact of violent entertainment is a 22 year study of New York middle-class school children.16 This study, also by Eron and Huesmann, found very strong evidence of harmful effects. At the time of ten year follow-up, the researchers found that, "the best single predictor of how aggressive a young male would be in late adolescence was the violence of the television programs he watched when he was in middle childhood." Eron notes that the cumulative effect over time can become large, since violence viewing stimulates aggression and aggressive children turn to violence to validate their own behavior. Thus, a bidirectional causal process model seems the most plausible.

After 22 years, the researchers still found that a diet of and preference for TV violence in childhood was a good predictor of adult criminal behavior. The quartile of school children consuming the largest amount of violent entertainment at age 9 were convicted of criminal offenses 150% more often than the quartile with the lowest consumption of violent entertainment. Although there was some preference for violent entertainment by children who were higher in aggression at age nine, after controlling for this and other important variables, a diet of TV violence still was a quite good predictor of convictions for crime in adulthood. There was also an important correlation between TV violence in childhood and convictions for driving while intoxicated in adult males. In females, the consumption of violent entertainment in childhood was related to an increased use of physical violence against their own children once they had developed their own families.

The study thus found that the harmful impact of television violence even carried over into the next generation with the offspring both consuming more violent entertainment and having increased problems with aggression themselves. Eron concludes, "The continued viewing of TV violence by children can have a lasting effect on their character and personality, leading to serious criminal behavior and antisocial violence of all types."

Four more important field studies, these with children three to nine years old, have been completed by Dorothy and Jerome Singer at Yale University(17,18,19). Parents kept careful logs of all programs viewed during two separate two week periods of time nine months apart. The children were followed from nine months to three years. A cause-effect relationship was found to exist between the violent TV programs consumed at home and an increased rate of assaultive behavior on the school playground. The correlation between TV-viewing and aggression accounted for between 10 and 25% of all the variance in aggression. An interesting finding was that it was the parents of the more seriously affected children who actually watched more television with their children. These parents were also more likely to take their children to contact sporting events. Even the saying of nighttime prayers proved of no beneficial impact in decreasing the harmful effects of violent programming. Such programming was also found to have a harmful impact on the imaginativeness of the children viewers.

A field study by William Belson with 12- to 17-year-old adolescent boys in London, England used in-depth sociological interviews with 1600 randomly selected juveniles (20). Here too, a harmful impact from violent entertainment was documented with increases in serious violent behavior being, in part, causally linked to both violent TV and movie entertainment. Realistic violence, cowboy violence and violence by good characters had the strongest effects. Cartoon violence was found to increase violent behaviors of a more minor nature in these adolescent boys.

A careful study of 183 adult males by Roderick Gorney and David Loye of UCLA randomly assigned husbands to one of five viewing schedules for a one week period(21). The wives kept diaries of helpful and hurtful behaviors committed by their husbands that week without being aware of what programs their husbands were watching. The husbands completed serial mood surveys throughout the week. The surveys contained questions measuring hostile feelings as well as other moods. The groups that watched their usual TV programming and the groups who watched assigned violent programming experienced no change in their levels of hostile feelings as the week progressed. However, the groups assigned to watch either non-violent or pro-social programming on their TV sets reported a 25% decrease in their levels of hostile feelings and were reported by their wives to have committed 37% fewer acts of hurtful behavior.

Of interest in the Gorney and Loye study is that the average number of hurtful behaviors in normal middle-class husbands was 3.5 events per week around the home. These are the typical minor losses of temper that are quite common in the American home and often the first events in the chain of behavior when more intense forms of violence occur. Men in the groups assigned to watch non-violent or pro-social TV averaged only 2.5 hurtful behaviors in the week of the study. Although this 37% decrease in hurtful behavior sounds impressive, it is important to note that the average person would be unable to note such a small change in angry and hurtful behavior without keeping very careful records over a period of many weeks. With typical male viewers averaging roughly 10 hours of violent entertainment per week, it is interesting to note that this study found that the elimination of this 10 hours of entertainment resulted in an average decrease of one hurtful behavior or loss of temper. Although this is a very small effect, when it is multiplied by 52 weeks a year and some 200 million American TV viewers, it implies an impact of major proportion.

A review of the 67 best studies on TV violence through 1976 was done by Scott Andison (22). Three-quarters of these studies found harmful effects. The more recent and more carefully designed and controlled studies were more likely to detect harmful effects. The age of the viewer made little difference. Indeed, the studies on the whole found a non-significantly greater percentage of harmful impact on college students than on preschool children. Of the studies finding a harmful impact, the average impact was a 25% increase in the parameter being used to measure the aggressive or violent attitudes or behavior.

Another field study with adults was completed by Neil Malamuth at the University of Manitoba (23). This study used college students who were enrolled in both a film course and a psychology course. Before and after surveys of rape attitudes were given in the psychology course and students in the film course were randomly assigned to attend either of two sexually violent or two non-violent theater movies in local theaters, without the students being aware that a study was taking place. The researchers were able to document sizeable increases in the acceptance of the rape myth, e.g. that women want to be raped, and an increase in the willingness to rape a woman, if certain of getting away with it. These increases were causally linked to the consumption of sexually violent movies of a non-graphic nature, which have since shown on both national Canadian and U.S. broadcast television.

Community & National Studies

Another type of field study looks at what happens to communities before and after the introduction of television. In two Canadian studies, increases in aggressive behavior were found to occur within relatively short periods of time after the introduction of television (23,24). One U.S. study looked at rates of criminal behavior in 1951 and 1952 as television spread to many new parts of the United States (25). Cities that already had television in 1951 were used as controls. A significant increase in petty larceny was found to be related to the introduction of television into new American cities, despite that fact that only 10% of Americans had purchased televisions at that time and despite the fact that TV violence was relatively low by modern standards. Eron's studies have been able to link increases in anti-social behavior to the TV violence viewing, per se, in countries with similarly low, but significant levels of violent TV entertainment. A recent report from China states that increases in anti-social behavior and decreases in positive school study habits have accompanied the introduction of television there as well (26).

Yet another type of field study has looked at local or national daily homicide or suicide fluctuations and compared them to newspaper reports of suicides (27), heavy-weight boxing matches (28), soap opera suicides (29), and reports of capital punishment (30). Homicides have been reported to be causally linked to the heavy coverage of major heavyweight boxing matches, while suicides have been linked to suicide themes in soap operas or heavy newspaper coverage of real suicides. TV network psychologists have rebutted the soap opera study (31). Media reports of capital punishment or major prison sentences for murder have been linked to decreases in daily murder rates, although the differences were not significant between the two types of punishment.

Attitudinal Surveys

Attitudinal surveys of adults in field studies have linked the consumption of greater amounts of television programming to a wide variety of undesirable attitudinal changes in viewers. These include an increased desire to restrict civil rights, increased levels of racial bias, an increased fear of criminal violence out of proportion to reality, an increased favoring of spending for military and police forces, and a decreased willingness to support non-military aid to foreign countries (32). Gun ownership, and increased numbers of watchdogs and doorlocks have also been linked to heavy TV viewing. Heavy TV viewers to consume more hours of violent entertainment, both on a volume basis and also because light viewers tend to discriminate more against violent programming than do heavy viewers. Research by Pingree has found that controlling for the types of programming viewed further strengthens Gerbner's findings and links these more closely to the violent program consumption (33).

Sexual Violence & Adults

An excellent series of studies by Edward Donnerstein of the University of Wisconsin has linked the viewing of popular horror movies, including Friday the 13th, Vice Squad, and the Texas Chainsaw Massacre to an increased acceptance of rape violence against women (34). College students from the University of Wisconsin were screened so only mentally healthy students were selected for the studies. Five such horror movies were viewed over a two week period. Students viewing the sexually violent movies became less sensitive to violence when asked to review a realistic rape trial. They showed less empathy, saw less injury to the victim and considered her more worthless when compared to male students viewing non-violent films. They also became desensitized to the violence in the films as they viewed more of them, enjoying the films more, while seeing the films as less degrading to women and less offensive.

Indirect & Other Effects

TV violence undoubtedly interacts with other factors in the causation of violent behavior. Role models observed in the home can have powerful effects. Studies show that children growing up in families high in hostile aggressive behavior are much more likely to develop problems with aggressive behavior themselves. Of course, TV violence can even play a role in these secondary influences by affecting the behavior of the parents ,as shown in the Gorney and Eron studies. Thus, the unconscious effects of TV violence can cause parents to be more likely to lose their tempers, which can then, in turn, have a harmful effect on their children.

Alcohol and Drug Usage

Alcohol consumption is another important factor in considerable violent behavior. Even here certain TV programming and advertising can play a harmful secondary role. The average child will see alcohol consumed on TV 75,000 times before he is of legal drinking age (38). The typical adult will view 5000 drinking events per year on TV and 99% of these portrayals will be neutral or favorable. Whereas an estimated 50% of serious real-life violence is associated with alcohol consumption, only 1% of televised violence is presented in this way (38). A study by Atkins has linked alcohol advertising to increased alcohol consumption in adolescent and young adult viewers (37). Programming portraying alcohol in a favorable light increases attitudes favorable to alcohol consumption in young viewers (35,36).

Other Factors

A desensitization to violence due to TV entertainment can have other harmful indirect influences. Reports occur frequently in the mass media of respected political leaders praising violent entertainment, even saying they wish that they could handle certain conflict situations like the heroes in violent programming (39). President George Bush's naming of Arnold Schwarzenegger as his Chairman for Physical Fitness is this type of endorsement. When it was reported in the mass media that Bush had a copy of Schwarzenegger's violent Total Recall, a film of intensely glamorized and sadistic violence, specially flown to Camp David so that he and his wife could see it, this gives a powerful endorsement to violent entertainment.

Entertainers make up the majority of leading role models for normal adolescents and, probably, for adult viewers. A recent survey of the heroes of high school seniors found that eight out of tens of their most respected heroes were entertainers, several stars of violent TV programs and films (40).

A large number of other observational and field studies agree with these results (40a,b,c,d,e,f).

Violent television programming routinely and explicitly promotes a value-system praising revenge and violent retaliation against aggressors as the best and usually only possible way of dealing with such conflicts. Attack rates on television are some 300 times higher than in real life (41). Crimes of intense violence are exaggerated even more than this. The average American viewer will witness an incredible 800,000 acts of violence on TV during his lifetime at current rates. This includes 120,000 attempted murders and 96,000 murders, and 4,000 suicides (42). In real life, most Americans have never viewed even a single murder.

Whereas the traditional ethics of western civilization discourage hatred and revenge, violent programming often portrays these in a positive light. The viewer learns to hate his enemy, instead of learning to try to understand his enemy. Instead of attempting to convert his enemy to peace and non-violence, the viewer falsely learns that the best and usually only way to deal with violent conflict is with more violence. Whereas, is reality non-violent means of dealing with conflict is, on the average, much more successful, the viewer is taught the opposite message. Indeed, police officers on television use their guns an incredible 800 times more often than in real-life and this usage is portrayed in a very positive light (43). Even this report may be an underestimate. Whereas the average police officer in Chicago fires his gun once every 27 years, The two TV killer cops on Miami Vice shot 43 people dead in 18 episodes. This was five times as many people as the entire Miami police force killed in a full year of Police work. The Miami Vice heroes were 12,000 times more violent than real life police officers and if every police officer in the nation imitated the TV killer cops, all of the rest of us would be dead in just six months. In Tango & Cash, Sylvester Stallone and his sidekick as Los Angeles police officers, killed 17 people. This was more people than the 7500 real Los Angeles police officers killed in the year the film was released, 1989.

Sub-Types of Violent Entertainment

Sports Violence

Reviews of specific types of violent entertainment have repeatedly found harmful effects. A review of 48 studies of sports violence has found harmful effects on participants or viewers in 45 of the studies (44). The viewing of a hockey match with considerable roughness and fighting was found to have a more harmful effect than the viewing of a relatively clean hockey match (45). The viewing of boxing matches has been linked to increases in the tendency towards retaliatory aggression in normal college students (46). High school football players have been found to become more prone to retaliatory aggression at the end of the playing season when compared to the beginning of the season, using students in other physical education activities as controls (47).

Countries participating more heavily in contact sports in the Olympics have been found to be more likely to be involved in wars, when controlled for population and population density (48). Non-violent cultures have been found to have almost a total absence of contact sports, while almost all violent cultures were found to have contact and combative sports (49).

Cartoon Violence

A review of 37 studies by NCTV on the effects of cartoon violence on children and adolescents reports that 33 of the studies found at least trends towards increases in aggressive and anti-social behaviors. Most studies showed definite significant and causally related increases in anti-social behavior related to violent cartoon programming (50). The studies cover 4300 children in six different countries: the U.S., England, Lebanon, Canada, the Netherlands, and Australia. The cartoons studied include Mighty Mouse, Tom & Jerry, Roadrunner, Bugs Bunny, Batman, Superman, and Popeye.

Reviews of TV Violence Research

Only two reviews of TV violence out of several dozen since 1976 have not agreed to the presence of harmful effects from this entertainment. One of these reviews was by ABC television network psychologists and was denounced by the Surgeon General (51). The other was by a Toronto psychologist with no background in aggression research (52). These two reviews appear biased in both the selection of studies reviewed and the relative critical stance when evaluating studies. For instance, a field study by Milgram, (53) which found no overall effects from a violent program, is praised in both of these reviews. The study has subjects view a program about repeated theft from a charity collection box. In one part of the study, after the program was completed, the subjects were tested to see if they would steal from a similar charity box after being frustrated. Although no significant differences were found, subject viewing the theft program were twice as likely to think that they were being tested for their honesty by the researchers. In another part of the study, probably the most sensitive part of the study, no significant difference in theft rates was found, although the actual difference was a 50% increase in theft behavior. The increase was not significant because of the low frequency of the behavior in that part of the study.

In another field study cited by these reviewers, this study done by NBC television network psychologists, the psychologists reported finding no long-term harmful influences from violent programming (54). However, other researchers have reanalyzed the NBC data and report finding a small harmful effect despite several drawbacks in the study (55).

Trends in Violent Entertainment

Despite the overwhelming evidence of significant harmful effects from violent entertainment on normal child and adult viewers, the trend in entertainment is towards more violence in entertainment rather than less. Horror and violent science fiction films have increased markedly since 1970 with a 600% increase in box office ticket sales (58) by 1980 and a much larger increase in actual consumption thanks to the advent of pay-cable television and VCRs during the 1980's. Movies portraying war as a positive and exciting patriotic experience have become quite common since 1983.

Television violence has increased markedly since 1955 with the number of hours of prime-time programming dedicated to crime dramas going from zero hours in 1955 to 27 hours in 1985. Although the most significant increase in television violence occurred in the late 1950's, another significant increase has occurred since 1980. The advent of cable television has also introduced high levels of extremely violent movie entertainment into a large number of American homes. Homes with a pay cable movie channel are estimated to consume 50% more violent entertainment then homes without due to the fact that pay cable movie channels average two to three times as much violence as network television itself.

Children's programming has become markedly more violent although the violence peaked in 1985 and has started to recede. The number of hours of cartoons with exciting war themes increased from 1/2 per week in 1982 to 27 hours per week in 1985. This has been primarily due to the introduction of war cartoons produced for toy companies selling war toys. There were 10 such cartoons showing Monday through Friday in national distribution in late 1985 with an additional eight more war cartoons, each subsidized by a toy company, are currently in production for release in 1986. At least two of these cartoons are direct spin-offs of intensely violent adult war movies.

In other areas of entertainment, violence has also been growing rapidly over the past 15-20 years. Violence in pornography has increased from 1% to 19% of all portrayals since 1970 (59). Violent pictorials in the best selling sex magazines increased by 300% between 1973 and 1978 and decreased significantly since. Violence in rock music lyrics has increased by some 115% between 1963 and 1983 (60). And with the advent of music videos interjecting considerably more violence into rock entertainment than contained in the lyrics themselves (60). An intensely violent, hate-filled heavy metal sub-culture has appeared that never existed previously in rock of any other form of music entertainment.

Sports violence has increased with violent penalties occurring 350% more often in Canadian hockey in the 1970's when compared to the 1930's (61). One study found an increase in the tendency to focus on violent action in network broadcasts of professional hockey and football violence when comparing 1976 and 1982 programming (62). Professional boxing has become the fourth most popular televised sport, despite a large number of studies showing sizeable amounts of brain damage in participants as well as harmful effects in viewers (63). Professional wrestling now makes up 3 of the 10 most popular programs on cable television, despite evidence of increases in anger in normal college students after viewing such programming (64). Martial arts, especially that associated with deadly weapons, have soared in popularity with a new emphasis on anger and revenge violence found in many martial arts magazines.

The sales of war toys has soared by 600% when comparing 1982 to 1985. In 1985, war toys made up seven of the 10 most popular toys in America. These war cartoons average twice as much violence as earlier cartoons. They teach children to think of their opponents of unredeemably evil and best dealt with through violence. Non-violent means of dealing with conflict are very uncommon and rarely ever even tried.

Violent fantasy role-playing games have become very common. This area is now a $100 million industry. Some 50 such games are currently on the market. There have already been 125 reports in the media linking a heavy involvement in violent role-playing games to suicides or murders in real-life (66). In several dozen of these cases, the evidence is very strong that the games played a decisive role in the deaths. Although 125 deaths in themselves is a significant number, the much larger impact is likely to be on the millions of young, usually male, participants. These games, which usually are based around planning and carrying out murder and theft, are very likely to be desensitizing millions of young people to very sadistic violence. Unfortunately, no research other than the gathering of the reported deaths has been done on the impact of these games.

A study of best seller books from 1905 to 1984 by NCTV has found that in this area that there has been a marked increase in the proportion of bestseller books with violent themes and that this increase has occurred in the past 20 years. In the first half of the century only 15% of best sellers had themes of violence. Violent themes grew rapidly in the late fifties and Late 60's and 70's. Now over 70% of best sellers feature themes of intense violence. Sexual violence, satanic, and slasher horror themes were non-existent in best sellers in the first half of the century.


Case Studies

There are hundreds of reported imitations of violence on television and even in books. There have been at least 44 Russian roulette deaths directly connected with the viewing of the Russian roulette scenes in the Deer Hunter movie (72). An increase in self-immolations, accounting for 30 deaths more than expected over a 6-month period, has been linked to heavy news coverage of similar suicides on English television (73). An increase in suicides above the expected level occurred during the heavy media coverage of the Marilyn Monroe suicide accounting for some 216 deaths (74). Some 150 suicides have been connected with a popular best-seller Japanese book in the late 1970's (75). Four murders have been linked to the Rambo, First Blood movie and dozens of murders have been linked to various violent and horror films (76).

Violent fantasy role-playing games have been linked to 125 homicides and suicides (77). Dungeons and Dragons, the most popular of these games, has played a role in at least 12 murder trials to date as well as one rape trial. I, myself, have been involved with eight of the trials. In examining individual cases, it is often quite impressive how deeply involved in many forms of violent entertainment the defendants have been. It has also been impressive how often the defendants think of different ways to kill people and how obvious the connections are to their entertainment consumption.

In these cases it is quite frequent to find a heavy consumption of intensely violent revenge, war, and horror films. Heavy metal music is often popular as is a fairly deep involvement in the martial arts, including an involvement with martial arts weaponry. Often the reading of survival guerilla war materials has been present, including pro-war mercenary soldier magazines which are currently available in many supermarkets, etc. Although these individual cases may not prove a cause-effect relationship on a larger national basis, they certainly bring home the lessons of the experimental research.

In my own psychiatric practice, a simple questioning of patients as to their TV, film, music and other entertainment habits often leads to interesting findings. Several women have linked specific, violent sexual pornographic material to physical and sexual abuse they have suffered at the hands of their husbands or boyfriends. One male patient and one adolescent boy clearly linked specific pornographic material to playing an important role in leading to sex crimes that they committed. Very frequently I find cases where a diet of violent entertainment appears to be playing a significant role in problems of anger and loss of temper. Working to sensitize patients to this problem and to eliminate such inappropriate consumption should be a standard part of complete therapy program.

I have had three patients in four years link suicide attempts to specific TV or film programs they had just seen. In one case a mentally normal, but frustrated 12-year-old girl took an overdose after seeing a TV program with a similar overdose. In the program, the overdose had led to a better understanding between mother and daughter. In a second, more recent case, a 15-year-old adolescent had seen Rambo, First Blood II several days in a row. He had gotten into the Rambo mystic after the First Blood film, often playing war with his friends. After the repeated viewing of the second film, he went immediately to the Army recruiting office, a phenomenon which was reported to have occurred in many parts of the U.S. When he was turned down because of age, he went home in a very frustrated condition. He got into a fight with his mother and attempted to commit suicide. Upon admission to the hospital, he was noted to be wearing a Rambo-type headband and had a U.S. flag hanging from his waist. In a very frustrated manner, he explained to me that he just wanted to be a soldier, "kill communists and make my parents proud."

Obviously, these are all unusual cases and in each case the violent or sexually degrading materials were only one factor out of many. However, so many cases where entertainment violence was obviously involved in just one general, unselective psychiatric practice strongly supports the research suggestions that the heavy American diet of murder and rape fantasies is having a significant and harmful impact on our society.

What Should Be Done?

Given that the problem is a very serious one, effect societal actions are clearly indicated. Unfortunately, the violent entertainment industry has become a $20 billion a year giant. Many people think that the first amendment precludes any significant action. This is certainly what the entertainment industry would like people to think.

The answer to the problem is not any one single step but a large series of reasonable steps that any responsible consumer democracy should take. The TV Violence Act of 1990, sponsored by Sen. Paul Simon, was signed into law by President Bush last year. It is a very mild piece of legislation, but the first ever to address the TV violence issue. It merely gives the TV and cable industry a three year exemption to anti-trust legislation limited to allowing them to meet and take responsible industry actions to correct the TV violence problem. In the six months since the passage of the law, the industry has refused to meet.

1) Probably the most powerful single step would be for every TV channel to be required to carry several public service announcements each day to teach viewers that TV violence is likely to have a harmful, subconscious effect on them personally no matter what their age or level of intelligence. The PSAs would teach viewers that it is not whether you can differentiate fantasy from reality, but what kind of fantasies you are having that determines your future tendency to be sensitive to the problem of violence and to be able to deal with stress in non-violent manners. These PSAs would provide some balance to the extensive promotion of violent entertainment throughout the viewing day.

2) A second important step would be reasonable codes of ethics on the part of the industry that they would follow, especially in regards to entertainment directed towards children. Such a code should prohibit programming that has been shown or that is similar to programming that has been shown to increase aggression in normal children.

3) A third step would be the establishment of honest public film rating system or systems. This would replace the current uninformative and extremely lax industry run rating system. It would also allow the film ratings to be enforced at theaters and in video stores. Many of the currently violent R-rated films would be restricted to adult audiences only and shelved with the pornography. Such ratings would also bring pressure on pay cable movie channels to schedule films of extreme violence only after 10 P.M. and with warnings that these films are likely to have harmful effects even on normal adult viewers.

4) War toys and video games would be required to carry labels explaining that more than one dozen studies have found harmful psychological effects with increases in violence in normal children.

5) Schools, psychiatric units, churches, prisons, and other institutions which are responsible for teaching and supervising individuals would conduct regular educational programs sensitizing people to the problem of violent entertainment. These institutions would act responsibly to assure, as much as possible, that the unhealthy consumption of violence did not occur while individuals were under their care.

6) The government would set of goal of say a 75% decrease in the consumption of violence and would take every economic means compatible with the first amendment to attain that goal. The government would monitor the levels of violence on television on a regular basis. It would fund projects designed to decrease the consumption of violent entertainment.

7) Government would fund the production of pro-social entertainment that taught peace and non-violence and attracted viewers. It could do this by by production rewards that increased with these elements. Government would cease running its own advertising on programs that glamorized violence.

8) The government would shift the Pentagon film assistance from its current policy of aiding only films that makes the U.S. military look good to only films that do not glamorize violence. State film commissions would stop aiding the production of violent films.

9) Boxing as a sport would be banned. Efforts would be made to eliminate the fights of ice hockey and officials in other sports such as basketball and baseball would be pressured to eliminate displays of aggression.

10) Within the field of psychiatry, in view of the seriousness and frequency of problems with violence, JCAH would require all in-patient units to have a regular weekly course on selecting healthy entertainment. The course would probably take the form of a one hour didactic session with question and answer and discussion period. A videotape accompaniment could further be added and increase the length of the course to two hours. Psychiatrists and psychotherapists should include an evaluation of entertainment consumption in standard history taking and should encourage all patients to clean up their entertainment diets. The AMA and APA actually encouraged all physicians in the late 1970's to urge their patients to pressure the sponsors of violent programs to cease their sponsorship. If the APA and AMA are really serious about their commitment and in view of the fact that the 1979 recommendation did not result in remediating the problem, then these bodies should form an ongoing panel to work at monitoring glamorized violence and degrading sexual themes as well as other harmful themes, e.g. alcohol and nicotine consumption, in popular culture. The bodies would have as a task helping direct efforts to effectively decrease violent entertainment.

Since the intensity and consumption of violent entertainment appears to still be growing at least in young people, the effects of this diet of violence is likely to be with us for a long time. Research does suggest, however, that a significant proportion of the harmful effects are short-term and a change in diet could lead to a relatively rapid change in national consciousness.