The Bandwidth of Podcasting
5. Distribution and Circulation
Following a period in which rhetoric and writing scholarship deamplified questions of delivery (Trimbur, 2000; Yancey, 2006), scholars in the field have written voluminously about distribution and circulation (see Ridolfo & DeVoss, 2009). For example, building on Margaret Syverson's work that "places the ‘scene' of writing into a field that is distributed and socially situated," Jenny Rice offered a "model of public rhetoric that sets its sights across a wider social field of distribution" (Rice, 2005/2018, p. 172). Rice argued that "theorizing public rhetorics … as a circulating ecology of effects, enactments, and events" offers a more robust conception of rhetoric than models that see rhetoric as "the totality of its discrete elements" (p. 167). More recently, Dustin Edwards (2017) has noted that while many scholars consider circulation in connection with ancient rhetoric's canon of delivery, with "a delivery-futurity focus allow[ing] writers to theorize how their texts may come to matter via a process of circulation," circulation is also relevant to the canons of invention and memory: "circulation can … be theorized as already-in-motion fields from which rhetors can invent anew." Ultimately, he argued that tactical rhetorical approaches informed by circulation would emphasize "a savvy understanding of redistribution mechanisms, practices, and economies, where rhetors make use of multiple channels of circulation depending on a desired outcome" as well as "an appreciation of unpredictability, where rhetors understand that, once redistributed, circulating discourse can amass a rhetorical life of its own."
Perhaps the most influential recent work on circulation's relevance to rhetoric and writing is Laurie Gries's (2015) Still Life With Rhetoric: A New Materialist Approach for Visual Rhetorics. In that book, Gries tracked the digital circulation, distribution, and manipulation of Shepard Fairey's iconic Hope poster (Gross, 2009), the original version of which depicted a red, white, and blue rendering of then-presidential candidate Barack Obama's head and shoulders above the word "HOPE" (Gries, 2015, p. 3). Drawing on what she called "circulation studies—an interdisciplinary approach to studying rhetoric and writing in motion" (p. xix), Gries used an array of digital tools and methods to track the image's circulation, including the original iteration as well as subsequent parodies and homages. Gries's methods "attend to seven distinct yet co-implicating material processes: composition, production, distribution, circulation, transformation, collectivity, and consequentiality" (p. 14). Given this complex set of intertwined terms, before I put the sources referenced above in conversation with podcasting, let me take a moment to tease out what I mean by "distribution" and "circulation."
When scholars like Rice (2005/2018) and Gries (2015) offer "distributed" models of rhetoric and writing, they are often arguing that rhetorical actions and effects are not simply attributable to a single person or agent (e.g., a politician giving a speech) nor to a discrete, separable collection of human participants and nonhuman elements (e.g., a rhetor, an audience, a microphone, a stage). Rather, rhetoric is a process that is distributed through and among the fluid, fluctuating relations between various human and nonhuman actors. In this sense, "distributed" means roughly the opposite of "centralized," a conception of rhetoric that challenges but does not erase the role of human agency in rhetorical causes and effects. As Rice (2005/2018) put it, such models "permanently trouble sender-receiver models" of rhetorical situations (p. 167). But in the case of podcasts, "distribution" also has a meaning more tied to the vocabulary of supply chains than that of rhetorical theory. In this second sense, we might think of "distribution" in terms of the distribution centers operated by food banks. As explored in an episode of the podcast Planet Money, these centers do rely on a somewhat centralized logic, functioning as starting points from which food is distributed to communities (Smith & Goldstein, 2015). Of course, even in this sense, distribution is not a strictly one-to-many phenomenon. A food bank might have a number of localized distribution centers, and the goods it gathers for the sake of subsequent distribution come from a variety of sources: food donated directly by individual people, canned goods collected in bins at grocery stores, products purchased by food bank employees using monetary donations, and so on. Moreover, the distributed goods might well be subject to further exchange after they leave the food bank, shared among or across communities before or after being prepared.
If we stick with a simple conception of this second sense of "distribution," circulation is often presented as a subsequent step. Edwards (2017) argued that circulation "is … about navigating and inventing from what's already there," and, as we've seen, linked circulation with "redistribution." If distribution typically names the processes by which something—a meal, a podcast, a rhetorical concept—moves from its ostensible point of origin or moment of creation out into broader communities or ecologies, circulation typically names the processes by which that thing is exchanged, shared, delivered, and reinvented thereafter. Meanwhile, Gries's and Rice's theoretical senses of rhetoric as distributed can remind us that circulation and distribution constantly bleed into each other (Rice, 2005/2018, p. 178), defying the expectations and intentions of early-stage composers, producers, and distributors.
With that vocabulary in place, let's return to podcasts. If we understand distribution in the sense of a step that occurs between composition and circulation, a typical podcast's distribution process arguably begins when someone uploads an audio file of an episode to the internet. In most cases, this now looks a little different than the process implemented by Christopher Lydon and Dave Winer. While RSS feeds still play a role in how podcasts get distributed, they usually aren't set up on a case-by-case basis, and they're much less obvious to those doing the uploading and downloading. Instead, podcast hosting services like Liberated Syndication and Anchor automatically generate an RSS feed for podcast producers. The person responsible for distributing a podcast can then submit that RSS feed to podcast clients—sometimes referred to as "podcatchers"—like Stitcher and Overcast, as well as programs and apps like Audible and Spotify, which offer podcasts alongside music and other media. Once the RSS feed is synced with a podcast client, any episodes uploaded to the podcast's hosting service are automatically shared with and posted to that podcast client. This is the point at which fans and other interested listeners can stream or download the episode—manually or automatically—using their own devices. Many podcast hosting services also function as podcast clients in their own right. For example, while my podcast Rhetoricity is available via Apple Podcasts and Stitcher, Liberated Syndication also provides a dedicated website where episodes are automatically posted after being uploaded.
There are, of course, hundreds of variations on this process. For example, This American Life's and Radiolab's distribution processes begin when they're played on NPR stations, with online distribution in the form of a podcast only occurring later. Some podcasts don't use RSS feeds to distribute episodes to multiple digital locations, simply uploading episodes to a single platform like SoundCloud. Some put a much greater premium on their own websites as a distribution hub, while others maintain no centralized web presence and rely almost entirely on podcast clients. Moreover, depending on the podcast, an array of other entities might be involved in distribution, both before and after the moment of an episode's release. In recent years, podcast networks like Gimlet Media, Earwolf, Maximum Fun, Night Vale Presents, and Radiotopia have become major players in podcasting. These networks function a little like NPR, offering financial support and access to higher-end audio technology that many unaffiliated podcasts lack. Meanwhile, other podcasts might receive support or be spun off from established media companies, as in the case of podcasts like Vox Media's The Weeds, or universities and foundations.
Of course, most teachers interested in making or assigning podcasts aren't seeking out affiliations with established podcast networks or devoting the time it takes to manage and keep up with a multidestination RSS feed via a paid hosting service. In many cases, if a teacher- or student-created podcast has any sort of public digital face (which is arguably the point at which it becomes a podcast rather than, say, an audio essay assignment), it might just be a course website or blog where the audio files are uploaded. And that's just fine—a lo-fi throwback to the early days of Radio Open Source, and an approach befitting podcasting's "GuerrillaMedia" origins.
But even if scholars, teachers, and students aren't interested in pursuing robust distribution channels for a podcast they're creating, it's important to consider these factors as both hypothetical variables and material realities that play a major role in how podcasts sound and get around. My point here resonates with the pedagogical implications with which Rice's (2005/2018) article concluded. She wrote, "Whereas research is often considered by students (and even some teachers) as a process leading to public production and circulation (a means to an end, so to speak), we can look to the logics of a generative research method that takes the circulation of effects as an aim" (p. 180). Rice's claim emphasizes circulation, which I'll return to in a moment, but I would argue it applies to distribution as well. The "logics of a generative research method" are relevant to both analytically oriented and production-oriented podcasting assignments. For instance, students writing a rhetorical analysis of a podcast shouldn't just be considering things like how a podcaster's tone establishes their ethos or how the use of sound effects and music generate pathos. Podcast analysis should not be limited to the rote, pedantic echoes of rhetorical appeals and concepts represented in some contemporary pedagogies and textbooks. Analyzing podcasts as podcasts requires attention to distribution, not just composition and production. Though some aspects of a podcast's distribution process might happen out of the public ear, students and scholars analyzing podcasts might consider questions like the following: Is the podcast affiliated with and/or distributed by any particular radio stations, podcasting networks, or other entities? What hosting services does the podcast seem to use? Is the podcast available via podcast clients and, if so, which ones? Is it distributed via any sort of standalone website? Pursuing these infrastructure-oriented distribution questions might suggest just as much about a podcast's unintended and intended audiences, effects, and possible futures as questions about its composition. Moreover, in cases where students are encouraged to circulate their own audio work rather than simply study others', a deep dive into matters of distribution can help students think practically and theoretically about the trajectory of their own work as part of a "generative research method" (Rice, 2005/2018, p. 180).
Adding circulation into the mix raises even more questions and considerations. In the conclusion to Networked Media, Networked Rhetorics, Damien Smith Pfister (2014) argued that the rhetorical canon of arrangement should be reframed so that it "might pay more careful attention to the interaction between different modes of communication and take seriously the possibilities of rearrangement" (p. 191). Like all networked media, podcasts are circulated via more than one communication channel and communicative mode. That is, they don't just move around as isolated digital audio files. Podcasters rely on metadata tags, social media, websites, fan communities, discussion threads, and an array of other platforms and media to gain attention and help their episodes circulate (Hilmes, 2013). They circulate via ads on local NPR affiliates, plugs in other shows distributed by the same podcast network, links posted to social media, and word-of-mouth recommendations shared between college students. These circulatory factors generate copious rhetorical and ethical questions. For example, in addition to its podcast form, Welcome to Night Vale is well-known for its Twitter handle, @NightValeRadio, which has over 350,000 followers and tweets things like, "This post brought to you by Uber. Burn down all cities. Salt all fields. Uber. The end of all things" (Night Vale podcast, 2017). While the tweet resembles satirical ads that occur in most Night Vale episodes, Twitter renders the ad in a more discrete and readily searchable form. Students might consider the relationship between podcasts themselves and complementary platforms like social media, exploring how, for example, Twitter attracts new fans to the Welcome to Night Vale podcast and/or offers an additional form of engagement with existing fans. With true crime podcasts, which have become ubiquitous, students might track hashtags, conduct web searches, and search online discussion forums to see how theories about a podcast's focal crime are circulating, and whether or how such theories are driving more attention to a podcast and vice versa. Given the troubling tendency of online communities to assign both crowdsourced shame and criminal blame in very short order, students might consider podcasters' ethical responsibilities in a digital climate where any stray thread might be taken up and circulated among a crowd of amateur investigators who lack the ethical training and compunctions of some journalists.8 8 For more on these issues in the context of the podcast S-Town, see Goudeau, 2017; Kilkenny, 2017. Again, these need not be isolated analytic exercises, but questions that recursively shape how students and teachers approach matters of circulation and reconsider matters of composition, storytelling, production, and distribution in light of circulatory possibilities.
These possibilities are intertwined with significant questions about podcasts' audiences. In terms of circulation, it's not just a matter of who is listening to a given podcast, but how they are listening and what they are doing with the material to which they're listening. As I touched on in section 3, podcasts can attract and speak to particularly narrow, particularly engaged audiences or broad casual ones. Whether creating or analyzing podcasts, scholars and students might thus ask what kinds of audiences are being targeted or likely to coalesce based on factors like a podcast's format, genre, content, mode(s) of distribution, and other media with which it is imbricated. For example, are audience members likely to be listening while they cook, commute, or do other tasks that might divide their attention? Or are they likely to listen in a careful and concentrated manner, perhaps becoming contributors to one of the detailed subreddits or wikis that have sprung up around many podcasts? The ways audiences listen, whether anticipated or unanticipated, might inform whether a podcast privileges quantity (e.g., long, frequent, relatively unedited episodes that allow listeners to zone in and out without losing the thread) or quality (e.g., shorter, less frequent episodes with a degree of production value and narrative coherence that rewards attentive listening). While these may sound like somewhat benign matters, there are serious related questions to be asked about the social, political, and cultural consequences of the ways of listening that podcasts afford. For example, numerous news stories about the radicalization of white supremacists position the ubiquitous hum of far-right podcasts as a key part of their subjects' media ecologies (McCoy, 2018; Roose, 2019). In short, the potential for podcasts' distribution and circulation mechanisms to target fringe audiences, in tandem with the ways they can be listened to (e.g., casually but constantly), can serve alarming ends, not just the frivolous proliferation of niche content and audiences.
With all this in mind, scholars and advanced students might take up more ambitious projects with regard to the circulation and distribution of podcasts. For example, tracking the circulation and distribution of podcasts might help us develop new models of digital rhetoric and communication. While Rice (2005/2018) argued that decades of scholarship on the rhetorical situation "permanently trouble sender-receiver models" (p. 167) of communication like the well-known and oft-maligned Shannon–Weaver model (see Losh, 2009, pp. 91–93), the MP3—an audio format that, as I mentioned earlier, was central to the rise of podcasts—was in many ways developed based on Claude Shannon's research (Sterne, 2012, pp. 19–21). While podcasts are not coextensive with formats, considering the history of and theoretical possibilities for podcasts' circulation and distribution might help us develop new rhetorical models that challenge not only scholarship and pedagogical practices in rhetoric and writing, but work in fields like communication studies and sound studies, opening new angles for thinking about media and formats alike.
We might also consider novel ways of tracking podcasts. While transcripts are available for many podcasts, they are not available for all, which can make the medium inaccessible as well as difficult to track using many of the resources digital humanities scholars have developed for studying and indexing digitized print texts. In recent years, rhetoric scholars have been involved in developing programs that can automatically aggregate and analyze nonprint media (Kuhn et al., 2015; consider also Gries's discussion of PikTrack in Detweiler, 2017). While I imagine a similar program for analyzing audio as a purely speculative possibility, rhetoric scholars with a background in the circulation and distribution of podcasts could be well-positioned to contribute to conceptualizing and applying programs that could better help scholars and students analyze audio archives, particularly distribution-centric media like podcasts (consider Clement, 2018).
All told, these theoretical considerations and pedagogical suggestions—while far from exhaustive—demonstrate how much distribution and circulation stand to offer any robust engagement with podcasts as both objects for analysis and a kind of media students, teachers, and scholars can create. In many ways, these are the factors that distinguish podcasts from the many other remarkable sonic genres and media covered in this collection. That said, there is one more piece of podcasts' distribution model that I have yet to touch on: as Bowie (2012b) pointed out, "Podcasts are episodic digital media files" (emphasis added). This episodic structure—in a word, seriality—is one last key part of what makes a podcast a podcast, and it merits sustained attention in its own right. It also brings us back to questions of genre, organization, and collaboration that are right in rhetoric and writing studies' wheelhouse.
More on seriality—after the break.