A few years ago, the US Army launched its Esports Team (which I call the USAE for brevity’s sake). And within a couple of years, they had (most likely) accidentally violated the First Amendment when a soldier-streamer banned a bunch of people trolling his chat as part of a larger clowning action that many gamers undertook to mess with the program. They stopped streaming. And, after they resolved a few things behind the scenes, they tried to go back, and had a pretty bad experience again. They scrubbed a bunch of old videos and left up a few links to send potential viewers to other accounts and pages they maintain. They haven’t been back on Twitch since.
I am skipping over a lot here (sorry; I need to lampshade it), but that incident more or less formed the basis of my research agenda during the last five years or so. Specifically: I am trying to figure out, what exactly is the US military trying to accomplish by positioning troopers as gaming personalities? What else are they trying to do? Does it work? Is any of this okay? These are surprisingly hard questions to answer, but I’m trying. My latest effort is trying to see what the remnants of the USAE’s Twitch channel. They left a few videos up, and I decided to look at all of them, hours and hours of footage, to see what they are trying to communicate about themselves.
Foust, J. (2026). Paratexts for strategic communication: The US Army Esports Team on Twitch. Media, War & Conflict, 0(0). https://doi.org/10.1177/17506352261495488
It turns out, studying Twitch is hard. You have to account for the streamer you’re watching, since streaming necessitates making the player part of the media, plus the audience, which is also part of the media one consumes when watching streamed gameplay, and you need to track how things in the chat affect things on screen, and in what way. Plus, you have gameplay being played underneath all of that, which itself can drive interactions. There are methods available for individual aspects of this analysis, but few have tried to combine a bunch of them in one place. I tried it, and found it is a nice way to analyze what streaming is like and how it might influence audience perceptions.
That’s all fine, and it would make for a tidy little unremarkable paper. But it turns out, the USAE isn’t just streaming; they are recruiting. They aren’t just playing for fun with friends, or playing to entertain or inform an audience, which are the two most common reasons people stream. The USAE was trying to get viewers to become interested in Army Service. That’s unusual. That it taking entertainment media into a different space, into strategic communication. That’s where states and large organizations are trying to persuade populations to think and believe and react to things, not just a streamer being silly on camera. It turns out, no one has analyzed a government using game streaming in this way (and in fact, as far as I can tell no one has used game streaming this way).
So, this isn’t a tidy little unremarkable paper. It is actually using a combination of methods in a new way to make two parallel points: the first is characterizing the actual videos the USAE have left on their Twitch page. But it is also developing a new methodology for understanding the instrumentalist purposes of streamed media, with an eye toward being able to apply this toolset in other streaming contexts (more on that in a future post). And at first, that methodology is the part I was really excited about because it combines transcription, textual coding techniques from interview-based work, video game analysis, and strategic communication analysis. Maybe if you aren’t in this field it’s hard to convey just how weird that combination is, but maybe trust me from the 15 months this spent in peer review: it’s weird, and people aren’t quite sure what to make of it yet.
But it also turns out at the US Army is (was) doing something really unusual and remarkable as well. They were trying to position their Esports soldier-players as gaming personalities, mirroring how other Esports team building processes might function. And this makes sense: it’s not exagerration to say that capital-G Gamers are probably more aligned with military values and the right wing politics that support them than the median American, or even the bulk of gamers who are not like that. There should be a lot of affinity to build off of. As alluded to above: this was mistaken, pretty badly. And, as a result, the USAE scrambled for quite some time to get their footing online and settle into a more appropriate communications posture the gaming public would be more willing to accept.
That’s the real core of this paper: seeing how the US Army was figuring out how to engage with gaming communities in real time. I found you could categorize their talk into three different streams: Army Talk, Gamer Talk, and Stream Talk. The Army Talk includes the rare moments when someone on screen makes an explicit pitch for enlisting in the Army — it happened a couple of times. But more often, Army Talk was focused on the annoyances of service, including an extended conversation about where soldiers can audition for the team while complaining about being deployed. This is kind of interesting, right? The recruiting program ends up providing “unfiltered” glimpses of griping, being frustrated with command in public, and policing each other on what’s appropriate to disclose on a public stream.
I also was not sure what to make of the actual gaming in the videos. Most of them are from Magic: The Gathering, and for the most part the play is incidental to the conversations on screen. In some cases, the soldiers talk about a recent competition (the one where the USAE soldier-players are complaining about losing a Counterstrike match was particularly funny), or they will explain a gambit they’re undertaking with a card or something. But the Game Talk was very limited. Similarly, the Stream Talk, about things specific to the stream or to Twitch, was just not particularly interesting. There’s useful stuff in there for academics, but I doubt there is a lot of significance beyond that. The Army Talk, though, really seemed to be the interesting core of this study. It’s interesting that they’re using the exhibition of gameplay as a backdrop to then talk about other topics in an effort to persuade young people of the virtues of military service, which is famously not oriented around playing video games.
It also, taking a few steps back, really highlighted the kinds of limitations the US military faces when it engages more directly with the public. It is really rare for soliders outside of a recruiting booth to have casual, social chat with random members of the public. It is unusual for soldiers and civilians to bicker like they did on Twitch, and it is especially rare for that encounter to rise to a legally precarious state that forces the Army to back off. And the Army clearly thinks this as well, since they have abandoned live streaming in favor of more traditional (and more distant) methods of engagement online.
That doesn’t mean militaries are out of the games-for-influence game. Not by a long shot. I’ll be covering more of that over the next few months as more work is ready to discuss. Let me know what you think!