Movie bad guys, by the numbers.

Warning: This post mentions major plot points from the 2017 movie Unlocked.

Last summer, in a rambling post inspired by a scene from Robert Altman’s The Player, I wrote about my friend who’d been complaining that Muslims were stereotyped as the bad guys in Hollywood films. I demurred that

even after a decade and a half of Middle Eastern war and unrelenting media attention to Middle Eastern terrorism, in the movies Middle Easterners were stalled in the number four bad guy spot behind Russians, Nazis, and rich WASPs – maybe even five, after Latin American drug lords. But my friend seemed to doubt me.

I went on to wonder whether our argument could be settled by numerical analysis. Could one analyze a large volume of films, determine who were “the bad guys”, and prove scientifically that Hollywood had been treating certain groups unfairly?

I attempted to define the parameters of the experiment:

One would need to examine all movies (caveat: define “movie”) over a given period, identify the main bad guys (caveat: by what criteria?) and somehow sort them (caveat: actors, or characters?) by ethnicity and religion.

I now realize I was understating the difficulty. Consider only my first caveat, defining the data set. Do you limit your investigation to American-made films, and if so, in the era of international co-productions what constitutes “American”?…or for that matter, in the era of Netflix and video-on-demand, a “film”? You could make a case for restricting your analysis to big-budget movies, as they more accurately represent studio conventional thinking. Or you could ignore budgets, and focus on the highest-earning movies, as they’re likeliest to reflect audience prejudices. Or you could include as many movies as possible, including little-seen indies, as they represent the widest possible sample of filmmakers.

Your choice will skew the results. If your sample is heavy on big-budget, theatrically released movies, you’re going to find a lot more superheroes shooting Nazis with laserbeams; the more you expand it to cheapo direct-to-DVD fare, the more Mexican cartel members you’ll see getting kicked in the face by guys in blue jeans.

But suppose you cracked all the above problems and carried out an accurate and objective census of bad guys: what percentage would qualify as “unfair”? What does science tell us is a proportionate depiction of Middle Eastern villainy?

***

Netflix recently made available a pretty generic spy thriller called Unlocked, starring Noomi Rapace, Orlando Bloom, Toni Collette, John Malkovich, and Michael Douglas. It’s ostensibly about Islamic terrorism, but none of the main actors plays a Muslim. In the end we discover that the evil mastermind is one of the top-billed stars – a CIA agent secretly helping advance a jihadi plot in order, he rants, to awaken America to the threat of biological weapons.

I’d seen enough movies of this type – i.e., more than one – to predict that it would be something along these lines: the only question was, would it be Douglas, Malkovich, or Collette who turned out to be the villain? This insight didn’t rely on parsing Hollywood’s racial politics; only awareness of Roger Ebert’s Law of Economy of Characters.

I could use Unlocked as a data point against my friend’s argument that Middle Easterners are negatively stereotyped: all the main bad guys, even the leader of a jihadi cell, are white men; of the five non-white Muslim characters, one is clearly good, three are ambiguous but portrayed sympathetically, and only one (fairly minor) is an outright villain.

But if I wanted to make the opposite case, those three ambiguous Muslims could easily be roped into the “bad guy” column; and it’s true that all the Muslims in the movie, good and bad, are defined by their relationship to Islamic extremism.

In short, like many movies on this theme, Unlocked could be pigeonholed – stereotyped, if you will – equally well as anti-Muslim paranoia or anti-American paranoia.

Poking around for reviews of Unlocked I came across this one by a writer who thought it was not just a good but a “great thriller”, and who was “pleasurably surprised more than once by sudden twists in the plot”. But even this credulous viewer found something to roll his eyes at:

The only real flaw it has is in following a very hoary cliché. Cynical viewers would guess from the beginning that the heroine’s black friend is marked for death.

As soon as we see his happy home life, and watch him playing with his beloved infant daughter, we know his fate is sealed…

This “flaw” didn’t even register for me. Is “black sidekick with happy home life is doomed to die” more or less of a cliché than “CIA heroine’s mentor is secretly the bad guy”? Could we conduct a numerical analysis and find out?

I doubt it. Movie-watching isn’t a science. We see the stereotypes we’re interested in seeing.

***

Pursuing the line of thought described in my earlier post, last summer I downloaded ten years of box office returns from the website Box Office Mojo and attempted to answer what I believed was a straightforward question: In the previous decade, had there been more movies about the “Global War on Terror” (henceforth GWOT), or about World War II?

I predicted that WWII would be the clear winner. In spite of (or because of) the ubiquity of real-life Middle Eastern violence in our newsfeeds, and the central place of Islam in our current ideological squabbles, in our fictions we prefer to go on reliving the clear-cut ideological and military triumphs of our grandparents.

I started with the top 200 movies, by North American box office receipts, from each year 2007-2016.

I threw out all documentaries and animated movies.

I disregarded country of origin but excluded a few foreign-language films for which there was little information online.

Then, using Wikipedia plot summaries for the 1686 movies remaining in my sample, I attempted to identify and categorize every war movie.

Finally, having devoted many evenings to this time-consuming project…I chucked the whole thing out.

I realized that my survey was absurdly susceptible to manipulation. Depending on how I defined “war movie”, I could make the case that WWII movies greatly outnumbered GWOT movies…or the exact opposite.

Here’s a table – which should not be regarded as in any way scientific – illustrating what I mean:

war movies wwii versus gwot 2007-2016

Click for PDF.

Movies marked red take place primarily in a war zone.

Movies marked yellow include one or two battlefield scenes, or explore the causes or consequences of war, or deal with war in a comedic or fantastic way…but most people wouldn’t think of them as “war movies”.

Movies marked orange could have gone either way.

Using a strict (red) definition of “war movie”, there were more than 1.5 times as many WWII movies as GWOT movies. (16-10)

Using a loose (yellow) definition, the GWOT movies outnumbered the WWII movies by an even greater proportion. (43-25)

But those results are next to meaningless. I could have expanded the definition of “war movie” still further by hauling in the innumerable action flicks about ex-Green Berets fighting bad guys on U.S. soil. Or limited GWOT movies to only those involving declared wars in Iraq and Afghanistan.

I could have applied a higher or lower box office cutoff, or used some arbitrary criteria to exclude “non-Hollywood” films, or performed any number of subtle manipulations, to get whatever results I wanted.

My effort wasn’t entirely wasted. It has made me even more skeptical about dubious claims of scientific objectivity, and the journalists, bloggers, and social media stooges who unquestioningly pass those claims along.

Having said that, I can scientifically prove that there is a shortage of movies about the surprisingly busy sex lives of struggling middle-aged male writers. My study is forthcoming.

M.

Advertisements

1 Response to “Movie bad guys, by the numbers.”


  1. 1 Dolf van Deventer May 5, 2018 at 6:39 pm

    I like anti-muslim movies so why the fuss?


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s