Stream-lining Collaboration: Participatory Composition and Twitch

Zack Shaw

Video Transcript

In this video essay, I make an argument about the changing landscape of twenty-first century composition as collaborative, especially as media compositions and media literacy become integral components in a modern literacy education.

Many scholars, like Sarah Arroyo, have used platforms like YouTube to discuss new forms of rhetorical circulation and participatory rhetoric. In this video, I will discuss how another platform, Twitch, similarly engages audience participation. Twitch is an online live streaming video platform, which differs from YouTube traditionally—as Twitch was largely made for live streaming videogame content and YouTube is made for published videos of varying kinds of content like music, videogames, comedy, and so on. However, as they compete with one another, both platforms now overlap in many ways, as live Twitch broadcasts can be archived as Video On Demand—or VODs—and YouTube has since added live streaming with interactive chat functionalities.

Twitch is at the forefront of innovation with interactivity, constantly providing new infrastructure for audience participation, often—though certainly not always—through monetary transaction. Twitch viewers currently not only have the ability to chat with streamers and other viewers live, but also donate money to directly support the stream, donate bits (i.e., another form of currency on Twitch that are paid for or earned by watching advertisements), and even subscribe to a channel or directly pay the channel for a tiered subscription with benefits such as removing interruptive advertisements and gaining channel specific emotes. More recently, Twitch has added new participatory narratives like “sub trains” and “bit cups”—ways to track and participate in larger trends of support for channels. There’s also Watch Parties—where viewers with subscriptions to Amazon Prime can participate in a movie night with streamers and content from Amazon Prime Video and Crowd Control, a new way for viewers to control a videogame narrative that I’ll discuss more later.

Content on Twitch has expanded beyond videogames as well, to ASMR (autonomous sensory meridian response), Just Chatting, professional sports and e-sports broadcasts, online teaching, and announcements or live conventions. While other platforms have added live media and immediate rhetorical transformations, only Twitch has myriad avenues for immediate collaboration. Currently, these avenues remain primarily relevant to the financial support of live streams—this is because Twitch is constantly seeking new ways to not only generate interactive environments, but also encourage viewers to donate their time and money. Yet, as these interactive environments thrive, so too do the media literacy practices of the viewers, who learn and engage with live digital communities, and who digitally circulate meme ideologies.

These collaborative compositions on Twitch have flown under the scholarly radar, unlike discussions of YouTube and videocy, which arise from decades of work about digital rhetoric and participation from figures like Gregory Ulmer and Sarah Arroyo.

YouTube, Twitch, and Rhetorical Transformation

Arroyo (2013) herself wrote that “the commands of our online world relentlessly prompt participation” and “encourage collaboration,” and she used the word “commands” to signify that online media creators, as well as the platforms themselves, specifically use language commands to ask for audience participation (p.1). While her focus is on YouTube, this commanding language also exists on Twitch, where streamers, as well as language directly on the platform, ask viewers to subscribe to and follow channels, to click a bell to be notified about streams, to share the stream with others, to donate to support a stream, and to chat with other viewers.

Arroyo (2013) continued, considering YouTube a “complex network” and an “electrate apparatus” of communities in digital spaces that encourages what Gregory Ulmer called electracy (p. 8). And furthermore, she wrote that videocy, which is her evolution of Ulmer’s terminology specific to videos, has created an interconnected network of digital platforms that reproduce media in different spaces for different audiences, and many of these platforms encourage viewer participation—like news sites encouraging the public to upload videos of live events (p. 11).

Arroyo’s notions lead to Laurie Gries's (2015) understanding of circulatory rhetoric. Gries wrote that new materialists understand images becoming “rhetorical in divergent ways circulat[ing] with time, enter[ing] new associations” and begetting “a multiplicity of consequences,” and she called this process “rhetorical transformation” (p. 14) The images that she presented in Still Life with Rhetoric of the Obama Hope original photograph, stencil portrait, and parody versions evidence the potential for rhetorical transformation that images and their formatting can undergo.

Furthermore, in 2010, Chris Anderson, the custodian of TED Talks, delivered a TED Talk of his own about YouTube driving innovation because of the ways that audiences and content creators become inspired by each other. Here’s a clip from the talk:

Recently, I’ve become intrigued by a different way of thinking of large human crowds, because there are circumstances where they can do something really cool. It’s a phenomenon that I think any organization or individual can tap into. It certainly impacted the way we think about TED’s future, and perhaps the world’s future overall. So, let’s explore. The story starts with just a single person, a child, behaving a little strangely. This kid is known online as Lil Demon. He’s doing tricks here, dance tricks, that probably no six-year-old in history ever managed before.

YouTube dancers were challenging each other to get better at dancing and inventing new dance moves in the process; even the six year-olds were joining in. An interested filmmaker, Jon Chu, set up an online challenge for these dancers, and he recruited the best of them to dance for his new Web series he called "The Legion of Extraordinary Dancers" - The LXD. The dancers were web-taught, but they were so excellent that they were invited to perform at the Oscars in 2010. Anderson explained the implications of this online, community-driven phenomenon:

So, in both of these cases, you’ve got these cycles of improvement, apparently driven by people watching Web video. What is going on here? Well, I think it’s the latest iteration of a phenomenon we can call "crowd-accelerated innovation.” And there are just three things you need for this thing to kick into gear. You can think of them as three dials on a giant wheel. You turn up the dials, the wheel starts to turn. And the first thing you need is... a crowd, a group of people who share a common interest.

Anderson’s ideas should sound quite similar to what Gries and Arroyo consider the rhetorical transformation and participatory networks that images and videos can exist within, yet this fluidity transcends YouTube as a platform.

Twitch Streamers and Disruptions

Throughout the rest of this video, I will discuss Twitch as a platform that is innovating with its infrastructure to create entirely new and rapid forms of participatory composition. I’ll begin with a clip from a popular video game streamer, named chocoTaco.

“Ah, I was looking at chat. Don’t run me over. Oh what the heck? Oh come on. Whatever man. That sucks dude. I blame—I blame you chat, for this.”

chocoTaco looks at his Twitch chat instead of the game and crashes his vehicle into a rock; as a result, when he wins the game and his editor makes a video for YouTube, the video is titled “Clearly Chat’s Fault, Hello?” Not only is the narrative of the game affected by his interaction with his Twitch channel’s viewers, but the eventual video title, which is used to attract potential viewers on YouTube, reflects the narrative of Twitch chat interactivity. In other situations streamers will ask their chat for help when they are stuck in a videogame and cannot figure out the answer to a puzzle or where they need to go. Occasionally, viewers will donate to Twitch channels to make their favorite streamers do something in the stream, usually within the game they are playing. These are examples of the narrative changing in real-time due to a participatory audience.

Here’s another example of a streamer, Macie Jay, who survives an in-game explosion, and jokingly claims that his skill in the game caused him to avoid the explosion.

“I am a God. I simply cannot be killed, chat. I simply cannot be killed, all right? Like, I mean, come on. I’m—there’s just—the game understands who the better player is. I don’t want to hear it. I don’t want to hear a word out of any of you.”

His character’s death, moments after his proclamations of invincibility, makes for a humorous clip and results in another conversation that he has with his chat, as he laughs and asks them not to speak a word about the event to anyone.

These first two examples are surface-level interactivity and narrative collaboration between viewers and streamers. However, they do occur instantaneously; Macie Jay’s invincibility being thrown back at him by his chat, and chocoTaco’s humorous attempts to blame his chat for his distracted driving both exemplify the circulatory rhetoric Laurie Gries and Chris Anderson attribute to online video—yet they do so immediately. The major difference between the live interaction between streamer and chat and the participatory rhetoric of Sarah Arroyo’s videocy with YouTube is this immediacy; in fact, content from both chocoTaco and Macie Jay has been parodied or spliced into compilation videos that fans make, functioning much like Arroyo’s videocy. However, considering only video-to-video remixes and adaptations misses the fluent participatory compositions that happen on live streaming platforms and the rhetorical transformations that happen within each composition before video publications on YouTube. This quote from Arroyo (2013) suggests something critical about participatory culture on Twitch; she described the video publications of YouTube, which parody, adapt, or reference one another, as media that “welcom[es] disruptions instead of systematically excluding them” (p. 60). chocoTaco’s YouTube video title suggests that he is welcoming and profiting from the live disruptions, even if he jokingly blames them for his driving mishaps.

Twitch and GDQ Collaborations

The next few examples, which similarly show a systematic welcoming of disruptions to create new narratives collaboratively, are from a bi-annual videogame speedrunning marathon called Games Done Quick (GDQ). Speedrunning is a technique in which gamers use glitches, exploits, and the fastest possible movements to finish a videogame as quickly as possible, and GDQ marathons have viewers donate to charity to meet incentives. If the incentives are met, viewers will get to see the games, glitches, and races they want to watch.

These marathons portray Gries's rhetorical transformation at faster rates than video-to-video remixes, as the gamers, audiences, commentators, donors, and viewers collaborate to create a single narrative and videogame composition. This example is from the final game of the Summer Games Done Quick 2019 marathon. In this first clip, puwexil (the person playing) stalls for time while viewers are donating so that more donations can come through to meet a specific incentive.

SporadicErratic: “Yeah it’s less than ten grand away.”
Crowd: *cheering*
Vulajin: “Well, you know you might need to do some menuing before you go to the black omen. Just double-check like you know do you have all the right stuff equipped.
StingerPA: “All the right characters.”
Vulajin: “You should have, uh, dash ring on-on Ayaya.”
SporadicErratic: “All right guys, just a little over five grand. Come on!”
Crowd: *cheering loudly*
Vulajin: "Okay. Okay it’s good to double-check.”
StingerPA: “Good to double check that.”
Vulajin: “Yeah. Yeah okay.”
SporadicErratic: “We’re down to three. Three grand.”
StingerPA: “What was the battle speed? Was it three?”
Vulajin: “Hold on, hold on, hold on, hold on, hold on.”
StingerPA: “Do you have the notes? What was the battle speed again? Do you—can you read this?”
SporadicErratic: “Fifteen-hundred.”
Vulajin: “No. Um, uh… I think two—are you sure it’s three?”
puwexil: “Yeah, yeah it’s three.”
SporadicErradic: “Did we make it? Did we make it? Did we make it? Did we make it? Yeah we made it!”
Crowd: *cheering*
Vulajin: “Let’s go do the thing we all wanted to do.”

After reaching the incentive, everyone starts to realize that they are inching closer to the three million dollar mark, and the crowd starts to chant to encourage the Twitch chat to donate even more.

Crowd: *chanting “Let’s go Twitch chat!”* throughout
puwexil: “Sounds like we have the approval from management to, uh, to finish off SGDQ the same way.”
Vulajin: “The best part is nobody in the room can hear anything we’re saying.”
Puwexil: “Well, hopefully the home viewers can. I can’t hear it either”
Vulajin: “Catch it on the VOD, it’s fine. Should we join the count? We’re at fourteen”

The donation total, which is visible at the bottom right corner of the screen rises toward three million dollars for the first time in GDQ history, and the crowd begins to cheer on the viewers to encourage them to donate more for a new marathon record. The narrative of this speedrun changes entirely from a display of skill and quick gameplay to a collaborative composition where the audience helps motivate the viewers to donate for a greater cause, which was, in this case, Doctors Without Borders. In this next clip, they reach the three million dollar mark, and the crowd and commentators erupt in joy.

Crowd: *cheering loudly*
SporadicErratic: “Can’t handle it. I can’t handle this. Come on!”
Everyone: “One!”
Vulajin: “That’s what you guys are. You’re number one!”
SporadicErratic: “Here we go!”
Vulajin: “Here it comes!”
SporadicErratic: “Come on! Come on!” *Screaming until the donation tracker reaches three million dollars*

Each of these clips display interactivity between the various groups of people involved in streaming. However, this final example from Awesome Games Done Quick in 2020 is the most recent example of a new technological apparatus supported by the infrastructure of the Twitch platform. The new technology is called: Crowd Control, which reminds us of Chris Anderson’s notion of “crowd accelerated innovation.” I draw the parallel between these two ideas because Anderson speaks of crowd-sourcing as a process that drives innovation—new ideas and new collaborative compositions—and in every game played with Crowd Control, the viewers have the power to create an entirely new composition, working with the conventions of the game. Each Crowd Control playthrough is a unique composition. With Crowd Control, viewers can donate to cause something to happen directly within the videogame. By donating bits, the chat can cause the streamer, Andy, to gain or lose health, to be swarmed by flying chickens that inflict damage to the character, to deal with ice physics that make his character slide on the ground, to be killed in one hit by anything, and even to be killed instantly, with a large enough donation. In this sense, making it more difficult for the streamer to continue the game is seen as a humorous power that the viewers have over the streamer. Not only are they collaboratively composing this specific experience of the videogame and making a video online, they are essentially making their own version of the videogame together, an entirely new composition.

EmoSaru: “I’m really excited that chat is already making the combo effects happen. We’ve got the one-hit KO, we have the ice physics. I think that the ice physics, one-hit KO, inverted and swapped D-pad, is great.”
Andy: “Yeah, I’m actually currently moving with the face buttons if you can see, uh.”
Everyone: *laughing*
Jaku: “What a crowd favorite. I think Xwater really helped him out with that.”
Andy: “Oh. I’m a bunny. That’s a new effect that I didn’t know about.”
Everyone: *laughing*
Andy: “I forgot we added that.”
Everyone: *laughing*
Patty: “So how you doing?”
Andy: “Um, doing great!”
Patty: “We got a little bit of downtime while Andy cosplays.”
Everyone: *laughing*

In this clip, the commentators discuss the joys of the combination effects that the donors on Twitch can donate for. In the middle of the screen at the bottom, there is a list of effects that viewers are donating for, updating in real-time, and at the bottom right of the screen another list shows the current effects that Andy is dealing with and the time each effect has remaining--in this case he is dealing with five different effects: ice physics, a bunny costume, one-hit KO, swapping the buttons on his controller, and swarming enemy chickens. Here’s another tough moment for Andy.

Patty: “I would disagree.”
EmoSaru: “Those are Cuccos—those are the chickens.”
Patty: “I thought they were Deadrocks.”
EmoSaru: “Okay—they are flying Deadrocks.”
Jaku: “Oh the crowd.”
Patty: “So you were talking about stronger effects. How do you make dying stronger?”
EmoSaru: “Uh, I mean that is a pretty strong effect. I—I actually think that, uh, there are permanent upgrades and downgrades, and while I think—so Andy is used to dying, but when you just take away his stuff, that’s pretty great.”
Jaku: “Oh—they gave him a sword.”
Patty: “Oh… who did that? Aw.”
Jaku: “It was the crowd. Show yourself crowd.”
EmoSaru: “They can take it away. That’s fine. Swords can be taken away.”
Patty: “You gave him the sword? I gave him the sword.”

In this clip, the crowd donated enough to instantly kill Andy’s character and force him to respawn, losing time. Yet, a little later on, they donate to give him a weapon, and in the effect list, one can see a number of viewer donations from those who felt bad for Andy, deciding to give him essential supplies like bombs and rupees.

Twitch Plays Pokémon and Media Literacy Engagements

“Crowd Control” technology opens entirely new possibilities for playing videogames with streamers online collaboratively. It is an evolution of the narrative created during the now famous first instance of Twitch Plays Pokémon, where the game was played exclusively by viewers voting to make the in-game character take specific actions. The original Twitch Plays Pokémon set a record for the most participants of a single-player videogame online at 1,165,140 and was completed over 16 days, 7 hours, 50 minutes, and 19 seconds. During the original Twitch Plays Pokémon, the two “hive-mind” modes of viewer participation were called “democracy,” wherein viewer inputs would be collected as votes and the actions with the highest votes, over twenty-second intervals, would be executed in the game, and “anarchy,” wherein all inputs are recorded and executed in sequence, often leading to chaos. While anarchy was responsible for rapid progression in the game at times, it was also responsible for disastrous failures. Furthermore, viewers could vote in the chat window for which political mode should be active.

From the introduction of The Rhetoric of Participation: Interrogating Commonplaces In and Beyond the Classroom, Banaji et al. (2019) wrote that “composition scholars take these ideas of participation and community a step further by arguing that participation in the community of the classroom prepares students to participate in a democratic community” (“Community”). While the Twitch Plays Pokémon channel may not explicitly represent traditional classrooms, participants quickly observed that democracy led to slower, yet much steadier progression. Even so, disruptions stemming from differing plans and motivations were a welcome part of the Twitch Plays Pokémon narrative and an inevitability of millions of participants for a single player videogame. Furthermore, learning how viewer participation modes were coded into the flow of the game became part of the community’s media literacy engagements; participants communicated directly with media and with one another to enact overarching narratives in the game. Moreover, related participatory fandoms with memes and lore emerged from Twitch Plays Pokémon, spilling over to other online platforms, wikis, and even musical ensembles.

Colin Lankshear and Michelle Knobel (2011) provided a particularly useful definition of literacy engagements in the 21st century. They define literacies as “socially recognized ways of generating, communicating and negotiating meaningful content through the medium of encoded texts within contexts of participating Discourses” (p. 72). They offer a range of different activities that engage with these kinds of coded texts, which explicitly include gaming, fanfic writing, and meme-ing: all part of the Twitch Plays Pokémon phenomenon. Moreover, Twitch, as well as other live-media platforms, can be a useful tool for literacy pedagogues, largely due to the potential for group media discussion, even if not for the financial transactions. Regarding the classroom, Jason Palmeri and Abby Dubisar (2019) asserted that “digital media can support a participatory pedagogy by offering more options for collaborative student composing and inquiry both within and outside the walls of the classroom” (“Introduction”). Offering more opportunities for collaborative composition through digital media leads to the aforementioned productive democratic communities who simultaneously welcome disruption as a meaning-making opportunity.

Furthermore, the National Council for the Teachers of English (NCTE) (2019) ambitiously requires 21st century literate people to “explore and engage critically, thoughtfully, and across a wide variety of inclusive texts and tools/modalities,” to “advocate for equitable access to... texts, tools, and information,” and to “build and sustain intentional global and cross-cultural connections and relationships with others so as to pose and solve problems collaboratively and strengthen independent thought” (Introduction). Moreover, The Council of Writing Program Administrators (WPA) listed the desired learning outcomes for first-year composition students, heavily emphasizing “rhetorical situations,” “technologies” for composition and texts, and “electronic environments” (2014). I argue that instructors should encourage students to engage in collaborative digital environments like Twitch to work toward these daunting standards.

Twitch certainly isn’t the only live streaming media platform, yet its infrastructure and features allow for a number of avenues for viewer participation. If Crowd Control is the latest feature in a series of collaborative composition features that increase the participatory element of online video, one wonders where technologies like this will take Twitch, and other streaming platforms in the future, and what incredible applications these collaborations will have. As media broadcasting continues to trend toward point-to-point structures, I expect that we will see media composition, rhetoric, and digital rhetoric become far more collaborative.


Anderson, Chris. (2010). Chris Anderson: How YouTube is driving innovation. Youtube.

Arroyo, Sarah. (2013). Participatory composition: Video culture, writing, and electracy. Carbondale, IL: Southern Illinois University Press.

Banaji, Paige V.; Blankenship, Lisa; DeLuca, Katherine; Obermark, Lauren; & Omizo, Ryan. (Eds.). (2019). The rhetoric of participation: Interrogating commonplaces in and beyond the classroom. Computers and Composition Digital Press:

Council of Writing Program Administrators. (2014). WPA outcomes statement for first-year composition (3.0).

Gries, Laurie E. (2015). Still life with rhetoric: A new materialist approach for visual rhetorics. Logan, UT: Utah State University Press.

Lankshear, Colin & Knobel, Michelle. (2011). New literacies: Everyday practices and classroom learning (3rd ed.). Open University Press.

NCTE 21st Century Literacies Definition and Framework Revision Committee. (2019). Definition of literacy in a digital age. National Council of Teachers of English:

Palmeri, Jason & Dubisar, Abby. (2019). Participation as reflective practice: Digital composing and feminist pedagogy. In Paige V. Banaji, Lisa Blankenship, Katherine DeLuca, Lauren Obermark, & Ryan Omizo (Eds.), The rhetoric of participation: Interrogating commonplaces in and beyond the classroom. Computers and Composition Digital Press:

Fair Use Disclaimer

This site is for educational and research purposes only.

This site contains copyrighted material, and the use of which has not always been specifically authorized by the copyright owner. I am making such material available in the efforts to advance scholarly conversations surrounding media, pedagogy, literacy, participatory composition, and so on. I believe this constitutes a fair use of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17, U. S. C. Section 107, the material on this site is distributed without profit to those who have expressed a prior interest in receiving the included information for research and education purposes. For more information, please visit:

If you wish to use copyrighted material from this site for purposes of your own that go beyond fair use, you must obtain permission from the copyright owner.