Two Sides of the Argument: Does TV Violence Cause Real-Life Violence?

The question of whether watching violence on TV or in the movies or playing violent video games, leads to actual violent behavior, especially in children, has been hotly debated in recent years. Huesmann and Taylor (2006) take the position that there is a direct and significant causal link between viewed media/entertainment violence and actual violent acts; Freedman (2002) takes the position that no such link has been proved and that the methodology of those claiming such a link has been seriously flawed.

Both authors note that the “pop psychology” view is that such a connection does indeed exist. They both remark on the horrific Columbine and other school mass shootings as having galvanized the public and made this a very “hot topic.” Huesmann and Taylor note that the perpetrators of the Columbine shootings loved to play violent video games, but stop short of inferring a causal link from that alone; after all, most young boys play video games, most of which are violent, so concluding that video games cause violent from that datum alone is akin to saying that eating cheeseburgers causes violence because 99%+ of those who commit violent acts have eaten a cheeseburger at some point in their lives. Thus, both authors recognize the need for scientific research on this problem. Where the authors diverge is in their opinions of the existing research and the conclusions that have been derived from it.

Huesmann and Taylor (2006) cite a plethora of studies showing a correlation between viewing violent material and hostile behavior immediately afterward. Due to the sheer number of studies done and the fact that the majority of them reached similar conclusions, this is the strongest component of their argument. Significantly, the studies as a whole showed a modest but significant correlation. This was actually a milder cause-effect link than many people might intuitively consider and adds to the validity of the collected studies, in that it wasn’t said that every time a kid watches an episode of “Road Runner,” he wants to go beat someone up. Nonetheless, it should be noted that the validity of such studies depends on how rigorously variables were controlled for, and the authors do not mention any litmus test in that regard.

The authors, in mentioning the link between real-life violence on the news and subsequent violent acts by viewers, note the “Marilyn Monroe effect” whereby immediately after highly publicized suicides, there is a significant increase in the suicide rate. Given that suicide is in many respects the ultimate act of violence—against oneself—the connection cannot be ignored. The question naturally arises as to just how many of those people would not have committed suicide without the “inducement” of seeing a famous suicide covered on TV; but after all, the victims can’t be interviewed, so all that can be done is to infer that the suicide rate, all other things being equal, should remain relatively constant, so the measured spikes in the suicide rate can be inferred as causally related to the aforementioned TV coverage. This is a valid point and constitutes another strong argument on the part of the authors.

The author’s citations of longitudinal studies are less convincing. Longitudinal studies suffer from several defects. One is the near-impossibility of precisely replicating them. Another is controlling for variables, which is difficult because of the nature of the data-gathering process. Finally, there is the problem of definitions. The authors quote one study wherein a young boy shoving someone else within a certain time period after viewing violent media was taken as proof that said content had incited him to violence. Fifty years ago, this would not have been seen as remarkable in the slightest. In fact, in terms of definitions, just how should we test the hypothesis that “video games have caused an overall rise in violence”? We set a much lower bar for what constitutes violent behavior than we did in the recent past. Rough-housing on the playground, corporal punishment, and for that matter, wife-beating used to be taken as commonplace and not considered to be objectionable. Recently, we have broadened the definition of “violence” to include shouting and cursing at someone or threatening them verbally. Certainly, redefining violence in this manner will result in an increase in reported and perceived violence even if nothing actually changes. The authors fail to address the simple question: if violent media content causes violence, why aren’t we all slaughtering each other on a regular basis, since such content is now so ubiquitous? In other words, if the connection exists, why don’t we see its logical effects and consequences?

Freedman takes a somewhat different tack (2002). His basic contention is that the causal link is not proved. Since he spends so much time attempting to refute the methodology of studies on violence and the media as well as the intellectual rigor (or lack thereof) of those who cite them in policy recommendations, he clearly feels that the burden of proof is on the “yes” side. He mentions that many who give such recommendations to policymakers have not done rigorous research. He implies that the existing populist view—that a causal link does exist between media violence and real-life violence by those who view it—skews the reportage and distorts the facts. This is a strong argument, in that he makes it clear that those citing such conclusions and recommendations are often not experts in the field and therefore not qualified to recommend actions to policymakers.

Freedman attributes the distortion of facts (as he sees them) to preconceptions and the desire to serve an agenda. He accuses those who state that the causal link is unequivocal of selecting the facts to fit the conclusion that is desired. He goes to considerable lengths to support this contention, and his many examples of decisions being made from a point of insufficient evidence are compelling when taken as a whole. While fairly strong, where this argument weakens is when he fails to give any real reason for such distortions other than the desire of policymakers to pander to the public by supporting the populist view. That virtually everyone who reports on or discusses this issue is simply twisting the facts to serve a political/social agenda seems like an exaggeration.

Where Friedman crashes and burns is when he claims that no proper scholarly review of the research has been done by anyone. Like most absolute statements, this one feels wrong and probably is. It seems more like hyperbole than an honest statement of fact. Given that this is such a hot-button issue, it is vanishingly unlikely that no one has done a rigorous, thorough, and scholarly review of the literature. The impression Friedman gives, in fact, is that such literature reviews do exist but that he doesn’t want to mention them, because they might contradict his point or at least weaken it.

It seems to this writer that Huesmann and Taylor (2006) did a much better job than Freedman (2002) of presenting and supporting their argument. While it may indeed be that the view of media-caused violence is populist, agenda-driven, and unsupported by evidence, Freedman seems to be content to simply mutter, “It’s a conspiracy,” without citing any studies supporting his contention. By contrast, Huesmann and Taylor back up their position by citing myriad outside sources. Anyone with a neutral stance on the issue would very likely be persuaded in the direction of Huesmann and Taylor, simply because they offer more evidence than Friedman, who seems more concerned with refuting the evidence presented by others. This is particularly weak when one considers that, rightly or wrongly, the popular perception is that violence on TV and in video games causes real-life violence and therefore, the burden of proof is on the one who says that that is untrue. Freedman could have attacked the contention, stated by Huesmann and Taylor, that the observed correlations between media and actual violence are significant; in fact, as noted above, though real, they were quite modest. Rather than being dismissive altogether of the data, he should have tried to show how it might not have been all that meaningful. He didn’t bother to do that.

References

Freedman, J.L. (2002). Cited in Cavazos, J. (2013). General psychology supplement PSY 1113 (Issue 20) (pp. 26-42). Cameron University (4th ed.). New York, NY: McGraw-Hill

Huesmann, R. & Taylor, L. (2006). Cited in Cavazos, J. (2013). General psychology supplement PSY 1113 (Issue 20) (pp. 43-51). Cameron University (4th ed.). New York, NY: McGraw-Hill.