Media Violence versus Real Violence

In the days since Wayne LaPierre of the NRA blamed the Sandy Hook massacre on violent movies and video games (in particular, for some reason, Natural Born Killers, a film that came out 19 years ago and was a critique of the media's obsession with violence), a number of people in the entertainment industry have been asked about whether their products contribute to real-world violence, and they've seemed extremely uncomfortable answering the question. They seem to have no idea what the answer might be. As it happens, this is a question that has been studied extensively, although the research is a bit ambiguous and unsatisfying. Nevertheless, I thought it might be worthwhile to go over just what evidence there is for the assertion. So if you're a Hollywood big shot, read on so you'll have some idea what to say next time the question comes up.

But before we get to that, I was prompted to write this by seeing this interview Quentin Tarantino did with the UK's Channel 4. When the interviewer asks why Tarantino thinks there's no connection between enjoying movie violence and enjoying real violence, Tarantino refuses to answer in a way that suggests to me that inside his head he was panicking. "Do not ask me a question like that. I'm not biting. I refuse your question," he says. "I'm not your slave and you're not my master. You can't make me dance to your tune. I'm not a monkey." When the interviewer presses him, Tarantino says "I'm shutting your butt down," and "It's none of your damn business why I think that." The fun starts around the four minute mark:

You'd think that given his oeuvre, Tarantino would have a more thoughtful answer to these kinds of questions, but apparently not. So let's get on to what the research says. Like all media effects research (and much other social science), the studies in this area have a particular challenge, which is that the methodologies that can offer clear evidence of causality don't represent the way people use media in the real world very well, while those that measure things in the real world make it hard to unpack causality. You can do a lab experiment and know you've produced a particular effect with your stimulus, but does that mean it works that way? To see what I mean, take the "Bobo doll experiments," which were conducted in the 1960s by Albert Bandura. To simplify it somewhat, Bandura had kids watch a video in which an adult bashed a big inflatable doll with a mallet, then put the kids in a room with a big inflatable doll and a mallet, whereupon they picked up the mallet and bashed the doll. Bingo: media violence leads to real-world violence!

Again, I'm simplifying those experiments a little (there's an explanation you can read here), but only a little, and you can see the problem. Subsequent researchers have tried to create more realistic scenarios to see whether watching violent films or playing violent video games can produce aggression, with varying degrees of success. There have been hundreds of studies in this area; here's a brief summary of what they've found: First, laboratory studies do find that watching a violent film or playing a violent game can, in the short term, increase aggression. However, the highest correlations are found on things like "arousal" and "aggressive thoughts," while the lowest correlations are found on actual behaviors. Second, lab studies tend to show higher correlations between media violence and aggression than surveys or observational studies attempting to capture aggression in the real world. Third, surveys and observational studies suffer from the causal direction problem: If you find that, for instance, people with violent criminal records are more likely to enjoy violent movies, is that because the movies turned them into criminals, or because violent people like violent movies?

Like any widely-studied area, there are lots of different results, many of which are contradictory. But there are a small number of researchers who have produced multiple meta-analyses (which summarize large numbers of studies by many researchers), making the claim that taken together the results are unambiguous and that violent media constitute a public health threat. However, in the past few years their work has come in for some serious criticism, including re-meta-analyses, and among the arguments these more skeptical researchers have made are that the size of the effects produced in media violence studies is, on average, very small to begin with, and the more realistic a study's measure of "aggression" or "violence" is, the smaller the size of the effect (see here for an example).

So to summarize, media violence can make some people more aggressive in some ways to some degree in some situations for some period of time. Like I said, not particularly satisfying, but hey, that's social science.

Now let's try to haul ourselves out of that morass and get back to the question we started with: how artists and entertainers should think and talk about what they produce and its effects on our society. For starters, it's absurd for a movie director or game designer to say, "This is just entertainment; it doesn't mean anything." Art is supposed to affect people; that's the whole idea. It's supposed to arouse us, change our view of the world, make us feel something or think something. The greatest art can change us forever, and less meaningful art may make us feel something only for a moment. And just as television and movies can affect how we think about love or work or death or beauty, they might be able to change how we think about violence, which is why those who produce our culture should be as thoughtful as possible about all the messages they send.

That being said, it's a far, far cry from acknowledging that media portrayals of violence might get some people riled up to saying that they are the primary, or even a significant, cause of crime and violence in our society. Compared to things like genetics, childhood trauma, material deprivation, and the availability of guns, media violence makes only the tiniest contribution to our society's level of violence, if it makes any contribution at all.

Or to put that another way, if exposure to violent media was a significant determinant of real-world violence, then since media culture is now global, every country would have about the same level of violence, and of course they don't. Japan would be the most violent society on earth. Have you seen the crazy stuff the Japanese watch and play? (Two words: tentacle porn. Don't ask.) But in fact, Japan is at or near the bottom among industrialized countries in every category of violent crime, from murder to rape to robbery. There are many reasons, some of them cultural, some of them practical (Iike the fact that it's basically illegal for a private citizen to own a gun there), but the point is that even if all that violent media is having an effect on Japanese psyches, the effect is so small that it doesn't make much of a difference on a societal level.

And the same is true here: even as our movies, television, and video games have gotten more violent over the last couple of decades, crime has fallen dramatically. Because the other factors are the ones that matter.

You may also like