It has been an interesting week, I have been at a small conference in the US on cyber security and the question frequently asked is what are you working on? I think the assumption was that I would reply with something in the realm of cyber security, but that would be too clichéd for me. This week my research focus has been video games (we prefer to use the digital games since it encompasses all forms of the gaming industry).
Digital games have surpassed movies as the most profitable dimension of the entertainment industry. Due to this shift, an interesting question is if there are international relations implications for this development. Along with my co-author Phillip Habel, we argue that there are some key considerations that can be examined by focusing on games as a transmission and framing device for enemy images. Our paper for the Southern Political Science Association investigates enemy images in digital games. To this end, we coded 37 first person shooters (FPS) from 2001 to 2013 along many dimensions to investigate this question (sorry, we did not code Duck Hunt).
While still a work in progress, I need to code a few more games, we are interested in feedback as this project develops. (Also plan on adding codes for the nationality of the developer of the game)
The results to the project demonstrate some interesting trends. Of the 37 games coded, the US represents the protagonist 20 times and the UK 1. Genetic and generic humans are represented 16 times. There is a remarkable lack of diversity in the gaming industry. While the West is the main consumer of games, there is a lack of imagination in who is seen as the main actor in such games.
North Korea and China, nation-states that might be considered more pressing and relevant rivals to the West only show up as enemies in games once each. On the other hand, Russia shows up as the enemy 11 times. Either as a future Russian Federation or Ultra Russian nationalists, the most prevalent outcome for an enemy is the Russian context. Terrorists likewise might be considered the most promising enemy in digital games, yet they still are not as prevalent as one might think appearing seven times, mostly in various Tom Clancy series. Tom Clancy series also tend to use Latin American terrorists rather than Middle Eastern. Iraq, Iran, Libya, and Serbia do not appear as enemies at all, but Iraq, Iran, and Pakistan do sometimes appear as locations in campaign modes of some games. Historical enemies, such as World War II battles, also appear often (yes, I did just want to have an excuse to post riod-Hitler with machine guns). 
One might also think that the most successful games (I only coded games with more than 1.5 or 2 million in sales) might be off world alien adventures such as Halo or what I term monster games like Resident Evil and Left 4 Dead, but we do not see very many of these games represented based on sales. Only eight games feature aliens and six feature monsters. This might change with the success of Walking Dead the TV show and game, yet we don’t see this working so far in the FPS world (we did not code third person viewpoint shooters like Grand Theft Auto).
There are various reasons for these results, but I am interested in hearing what the public might think of these outcomes. The “Russia” result is likely because they are ongoing rivals with the US, it is hard to shake Cold War images, and the fact that China represents a large market for video games and other entertainment forms that is best not annoyed (see my book on Tibet and China for similar themes). The reason we see few games covering terrorism is likely because these fights might hit too close to home and lose the element of fun important in games. Medal of Honor Warfighter was a huge failure for this reason, killing the entire franchise.
Regardless of the reasons behind these findings, video games represent an interesting avenue of research. They certainly can be transmission modes for ideas. This is not to say that games will make people kill, but who the enemy is in these popular games tells us about who we fear, what we value, and the future of international interactions. I look forward to comments on these ideas and what others might do in this area of research.
Here is a draft of the paper if you would like to read the whole thing and see the tables.
I am reluctant to believe that video games are a true reflection or transmitter of ideas. The industry is often directed by a small group of people, trying to maximize earnings, thus the lack of diversity. This means that they copy some successful business model rather than represent the idea of an enemy, on this point I am thinking of Call of Duty and its copycats. People usually buy this kind of games based on its popularity rather than the premise of the game itself, although this is one possibility and I could be wrong. On the other hand, the success of some of the games might not depend on the enemies portrayed but on the gameplay of the actual game.
On another note, I think it would be interesting to compare sell numbers with other genres of video games, such as Role Playing Games. These games also represent enemies and have a wider variety of settings, yet the violence is fairly similar with FPS. The difference might be that it is not a genre so focused on the american market and thus representing a more global perspective about interactions in the international system.
Finally, I would like to say it is a very interesting topic and I am waiting to hear more about this.
The interesting thing to do in the future would be to use this paper as a baseline and get funding for an experiment regarding the transmission hypothesis. At this point its nothing more than a possibility and we could not hope to get funding for such a project until this basic paper is published first.
Going to speak a bit more about different genres in the paper, this comes up often, but the focus will be on FPS since there is this combat theme, the audiences are much bigger, if the transmission hypothesis has any validity, it would be here.
I am sure you are right about the marketing, thing is that these games have been much more successful they the developers probably ever thought. I started playing them as word of mouth filtered out, never once noticed an ad.
Brandon, that’s a really cool project. Some random thoughts (I havn’t read the paper yet):
– Why only FPS? If GTA and Fallout were in that sample, you’d see much more US self-criticism. Or all of the fantasy stuff, e.g. Elder Scrolls. Or Assassin’s Creed. (See Solaire’s comment. Also: FPS and RPG have come closer, with many interesting hybrids out there…)
– What about the Far Cry and Crysis franchise? (Maybe they didn’t sell enough copies, I don’t know?)
– This might just be the product of the trend towards US-centric war-sim FPS… if that’s an interesting finding on its own, OK, but I’d be careful with generalizations on the gaming industry. A couple of years ago, you wouldn’t have found a single zombie game — these things come in waves. (Again echoing Solaire here.)
– If you look at multiplayer, there’s a lot of successful war sims, e.g. World of Tanks. There you can pick basically all sides, regardless of where the developer comes from. It’s just about making it appealing to the largest possible audience. (Would be interesting to see whether players stick to their real-world home country or not.)
Thanks for the comments, we do note the waves issue in the new version of the paper. I am curious if the success of the Walking Dead show and storybook game will lead to a reevaluation of the topic in FPS. We can only look at the current sample we have, things might change in the future.
I look forward to seeing how this sharpens up. If the question is about simulations that depict enemies, I am reminded of the passage in Der Derian’s Virtuous War where he asks where the enemy missiles are coming from in a missile defense simulation, and gets the uncomfortable non-response.
I think it’s probably a mistake not to code/study GTA (or related games like Red Dead Redemption). A significant portion of what happens in the freeplay online could be considered terrorism (especially in GTA IV with all of the sessions at the airport where everyone is armed with RPGs). Moreover, it is amazing how many of the players are from KSA and UAE (as evidenced by their gaming name prefixes and use of Arabic to communicate)… at least during the odd hours when I am “researching.” The GTA games provide an opportunity to either “experience” a completely chaotic version of America or (perhaps) to take out aggression against America as an enemy. Just a thought… now back to car jacking… umm.. I mean working.
Thanks, I keep debating the GTA/ACreed issue myself. One problem is we are already at 9500 words with just FPS. 2nd is I don’t play 3rd person games, just can’t get into them for some reason. Gave up on Fallout pretty quickly having to look for this plant or that berry. I think feasibility at this point limits our ability to tackle the question. But if we get money for an experiment I would make a book covering FPS – 3rd person – Strartegy/Wargames – Transmission experiment. I think a sabbatical playing GTA and Creed is in order at that point :)
I’ll send you some comments but I think:
1) you need to reconsider your coding — “monsters” and “aliens” have human-relevant representational content, and if you don’t want to go there, not all monsters are created equal.
2) you need to embed this in theoretical/empirical work on popular culture and representation. There’s a dialogue out there about this stuff in IR, including my own work, that of Weber, Neumann, Shapiro, Weldes, etc.
Fine, pulling out Nexon and Neumann…
Not sure what you mean by the monsters and aliens point. Monsters ends up being Resident Evil and Left 4 Dead while Aliens is Halo and Gears of War
I’m not a gamer myself but I think back to my own childhood in the mid to late 60s, at the age when kids would play “army” outside, and for us, the “enemy” was always the Germans, even though WWII was 20-25 years in the past. Sounds to me like computer games are showing about the same time lag.
“The reason we see few games covering terrorism is likely because these fights might hit too close to home and lose the element of fun important in games. Medal of Honor Warfighter was a huge failure for this reason, killing the entire franchise.”
Silly me thought that Warfighter was a failure because of the clunky controls, broken VOIP, nearly impenetrable layers of menus, and the refusal of EA to fix those things. Instead, it was just that we weren’t shooting virtual Russians.
Thanks, noted in the text.
Medal of Honor was a failure because it was a terrible game, not because it was set in Afghanistan. The first of the two MoH reboots shipped with a broken campaign, which sunk any chance of its successor to find an audience. The fact that it was equally broken didn’t help it.
To me, the failure was the multiplayer component was horrible because of the setting. But that was my perception.
If you think that Warfighter failed because of who you were killing in the game, then you don’t understand how gamers actually evaluate things. The game failed because it was a boring, on-rails series of action setpieces and gimmicks.
That description also fits Call of Duty, which saw its sales stall out this last time around IIRC. Hardly anybody plays these for the campaign however. I mean I do, so I don’t play many, but that’s because I suck at FPS.
Right, but Warfighter’s multiplayer didn’t have the game-culture cachet of things like CoD and Battlefield, which continue to sell well despite terrible single-player campaigns. I think everyone recognizes that unless you’re CoD or BF, you’re going to have few people buying for the multiplayer aspect. This is why games like Killzone and Resistance focus much more on the single player mode.
Hi, I think it’s really great that you are pursuing research of this kind. I also believe that there are a number of issues with your data selection that should be resolved before drawing any conclusions.
First, a correction: The franchises Resident Evil, Gears of War, Socom and Splinter Cell are all Third-Person shooters, not First Person Shooters. You should either remove these games from your study or broaden the scope to action games in general. However, this would then grow to include games like GTA, Uncharted, Max Payne, etc. If you keep FPSs, you may want to include an explicit definition of what an FPS is, I could not find one in your draft.
Also assuming the FPS focus, there are number of important games not yet on your list. You did mention that you still had some to add, I only mention because these may change your results. The franchises Half Life, Crysis, Deus Ex, Far Cry, Doom and Stalker all contain games that match your criteria (> 2 million sold, since 2001) and contain prominent alien, monster and mercenary (not state affiliated) enemies. Their omission weakens your claim that “we did not preselect any games based on who the enemy was—our selection criteria was motivated by sales alone.”
In your Data and Method section, I appreciate you explain your reasons for many of your decisions like inclusion based on sales figures, the time window and decision to include only western games. It would help your argument, though, if you included some examples/data to back up these decisions. For example, I would feel more comfortable with the exclusion of non-Western games if you had some sales figures for the highest selling games produced outside of the US, and could demonstrate they did not approach your other titles.
One more thing that concerns me about the inclusion by sales is bias in favor of recent games. The games industry has grown tremendously in 12 years and a bestselling game in 2013 has many more sales than one in 2001. The top ten selling games in Table 2 all come from 2007-2013, the second half of your window. Only 10 out of 37 titles comes from the first half. Finally, as an example, Deus Ex is commonly regarded as a contender for “best game of all time” and has had profound impact on the industry. It was released in 2000 and has sold somewhere over 1 million copies. To accurately represent earlier games, the size of the market needs to be corrected for, possibly by using market share instead of units sold.
I hope some of these points help. I really appreciate that things like this are being studied, just want to make sure the information is accurate.
Thanks much, I did mention that the coding was not complete (basically was stuck doing it on an Amtrak), thanks for highlighting the ones I missed.
Thanks for all the comments, taking another pass at the paper and I will dive into them soon.