You are in the Sounds Research maze.

Here’s a story.

Long before I even conceived of doing a podcast on our research, I was in a truck with a pest control professional for a ride-along. It was our first or second meeting, we were still getting to know each other. We chatted about LA traffic, we joked about whether squirrels are the dumbest animals. I was learning about his day-to-day routine. He was asking me questions about the overuse of poison, and about the behavior of angry, animal-loving humans towards him, but also about things like artificial intelligence. I started to talk, as if I knew something, which is an occupational hazard of having a Dr. in front of your name.

But before I got very far he stopped me; “Wait,” he said, “Can we record this for my podcast?”

Of course we can. This was one of those classic anthropological reversals—who can tell the anthropologist from the anthropology? Your collaborators always know more than you do. Their theories have often been better tested than yours, and you are, if you are any good at being an anthropologist, always a bit depressed that you’ve condemned yourself to writing partial blurred accounts of a richly lived social reality, which they increasingly document better and more richly themselves. But this is also why many anthropologists turn to other forms for inspiration and guidance, why the straight-jacket of the scholarly article seems to bind you ever tighter the more you struggle against it.

The pandemic ended fieldwork for most anthropologists, of course. For about 9 months I waited for things to settle, to be able to go out into the world again, and it never came. In Jan. of 2021 I invited the members of my lab to think about doing something other than field research. As a result, these podcasts were created over the course of 8 months in a zoom-based collaborative research setting that involved faculty, graduate students, undergrads, and artists, one of whom also served as the audio engineer. Even before we were finished, people wanted to know how we did it, and questions began to emerge about how it might be repeated, how to assess it, how to document it etc.

.form.

Podcast, the word, for those of you who enjoy philosophical etymology, comes from the iPod, introduced in 2001, which was not the first, but nearly the first device, to allow a kind of digital version of radio. Like the “Walkman” before it, the sudden ubiquity of this consumer object worked its way into our language before we had a chance to blink–and thus “podcasting”, rhymes with “broadcasting”, was a no-brainer. The pod in iPod, incidentally, and more significantly, refers, (according to legend) to the pod bay doors which the murderous Artificial Intelligence HAL in Kubrik’s 2001: A Space Odyssey would not open for Dave. HAL, of course, was simply a successor iteration, letter by letter, of the Apple of Kubrik’s day, IBM. We are all Dave in this story.

What is a podcast really? It’s pretty hard to define it. And maybe not worth it. At a first cut, podcasts are just containers. Any performance can be recorded—a reading of a short story, a small musical performance, a Ted Talk, an academic lecture. A podcast then would be little more than an audio file and an “RSS feed” (the “really simple syndication” system that allows computers to know what’s in the list of podcasts). The content doesn’t matter in this definition and that is the way software engineers like it—ever since Claude Shannon.

There is obviously more to it than this.

Podcasts today tend to be interviews or discussions, or freewheeling conversations. For the most part, they privilege regular periodicity, much like social media, over the creation of a permanent work, like a film or a novel. Even the platforms are designed to nudge this constant production, this regular engagement and demand to feed the maw of new content. A generous way to characterize such a form is to suggest that a podcast is a bit like a musical score, and each episode is a performance of that score. The form it creates, based in the style, knowledge, and personality of the hosts, directs the content. So although there is an implicit imaginary around podcasts—the idea that they contain new content each day, week or season—it might actually be more illuminating to think of them as having the form of a musical performance. You can listen to a song over and over again; you can also become obsessed with different covers of a song, or virtuoso performances of a more classical piece. Contemporary composers, and most jazz artists, it is clear, often push the limits of this relation, creating scores that maximize these variations. Conversational podcasts direct their subjects–performers–to interact according to certain limits, with certain techniques and styles.

The podcast-as-composition would be to simply interview people who are used to talking: celebrities, academics, politicians. They have things to say, and often all a host needs to do is ask them to say it. But more complex versions are more popular. Marc Maron’s podcats WTF is a great example. Over the course of an hour or two, he talks to primarily comedians, but also actors and other people, even President Obama at one point. But he brings his comedian’s “score” to the conversation, which includes listening carefully, bringing a bit of light sparring or risky ridicule into the conversation, and then changing the key dramatically into something deeply personal. Everyone he interviews responds to this score, and as a result, even though every episode really is different, they are also strangely the same. A podcast like the overwhelmingly successful Joe Rogan podcast does something similar. Listeners come to hear Rogan roast a guest, say provocative things, but also to listen and to try to provoke a debate. Again, different every day, but also always the same.

Other forms clearly compete with this composer-performance style. The most obvious is the tradition of radio documentary. National Public Radio in the US, and CBC in Canada have strong traditions in this respect, ranging from one-off works like Glenn Gould’s Solitude Trilogy to the well known show “This American Life” hosted by Ira Glass (and running from 1994 to the present), or interviewers like Terry Gross and Fresh Aire. Often associated with local radio stations, these programs have set the bar high for both aural presentation and investigative journalism. Indeed, Podcasting itself saw a significant transition with the appearance of the extremely popular podcast Serial. What Serial did was to radicalize what previous radio documentary did well… rather than one good episode about a topic, it turned the documentary into a 13 episode epic which, crucially, unfolded in real time as the producers did their researching and interviewing. In retrospect, the name was an ingenious choice. The story at the center of it, of a possibly wrongly convicted high-school murder that played on American anxieties about Muslims and teenage freedom, was in a lot of ways undistinguished. But the serialization of it struck people as new—even though this is a form that goes back hundreds of years and gave us some of our most beloved Victorian novels. Today one can find literally thousands of serialized true crime podcasts.

Other traditions also lurk in the background here: experimental aural forms that stretch back to radio’s early days. The jazz poetry of Ken Nordine, or the experimental radio storytelling of Joe Frank, a Los Angeles subcultural hero, is the precursor to the success of a podcast like Welcome to Night Vale—a long-running alternate reality public radio station, of sorts, and boasting its own scholarly study (Weinstock 2018). Or one might look to the more underground traditions of Samizdat cassette recordings in the Cold War, or the Islamic public sphere of cassette sermons that Charles Hirschkind documents (Hirschkind 2006). Indeed, sound anthropology is a rich tradition here that may or may not get its due recognition in the world of podcasting (Feld and Brenneis 2004; Helmreich 2015).

By 2022, there were over a million different podcasts, and some 50 million episodes. The vast majority of these were not professionally produced, and a large majority of them exist for the same reasons a vibrant public sphere once occupied cafes, or later zines: to be heard, to debate and discuss, and to hear it in a form that emphasizes the emotion associated with the human voice and with the soundscape of our lives—something social media is particularly bad at doing, often with devastating consequences.

But like the public sphere that Habermas once wrote about the podcast sphere has transformed, through monetization, into something driven by money. The very idea of monetizing a podcast was, not that long ago, something very hard to imagine. After all the form was built on the model of radio—easy to access, regularly updated and free (as in gratis) to listen. But many committed to the form wanted to find a way to make it sustainable— such as NPR’s Alex Blumberg, who as host of the news show Planet Money, decided to both create a podcast startup in 2014 and to document it in a podcast, of course. Gimlet media, the outcome of this exercise is now one of the largest of the podcast producing corporations, alongside others like Roman Mars’ Radiotopia (Love and Radio, 99% Invisible), Crooked Media (Pod Save America), Nerdist, and Malcolm Gladwell’s “Pushkin Industries.” The definition of a podcast corporation is blurred by other companies like iHeartRadio, Public Radio Exchange, American Public Media, WNYC Studios, all of which have roots in radio and now function as a mix of radio, streaming, and podcasting. Alongside all this is the rise of the hosting companies that provide both technical distribution to your devices, companies like Anchor.fm and Buzzsprout, and who in turn connect sponsors, data gathering, statistics, and content into a glorious revenue-generating machine. But that is another story.

Listening to a podcast and making a podcast are related. For some, listening is active and engaged, for others it is passive entertainment. A colleague of mine complained that listening to podcasts is difficult because you can’t do anything else while you are listening. By this he meant, I think, that a good podcast captures you, immobilizes your attention. I think this is the right way to approach both music and aural storytelling, and maybe also scholarly presentation as well. But for others, a podcast is not supposed to be challenging or difficult. For most people, podcasts are for listening to at double speed while exercising, cooking, or simultaneously consuming other media. Such distracted listening is simultaneously a diminishment of the form, and an expression of the anxieties of solitude or FOMO (fear of missing out) that the pandemic only exacerbated. But it is a reality, and I am not judging anyone. I do it myself.

Podcasts tend to be very human. The voice is central, language and storytelling are at the center. But humans are not the only things that make noise. In many ways, the podcast we set out to make originally wanted to push this envelope. We wanted field recordings of the sounds of a multispecies city; we talked about what it is like to hear the city as different kinds of animals—what a cougar hears vs. what rat or a bird, or an insect hears. What the sounds of cars or honking sound like to different animals; what the city feels like to a bat (the classic problem raised by Nagel (Nagel 1974)). Many other people have been here long before me, especially in anthropology (Stefan Helmreich has long studied sound and its relation to non-humanity (Helmreich 2015)), so we claim no originality here. But there are nonetheless traditions and approaches that merit linking up.

I learned for instance, from composer Liza Lim, about “acoustic ecology” in the work of people like Hildegard Westerkamp and R. Murray Shafer, a real attempt to listen to our sounds and words as material, acoustic phenomena not just as listeners, but also as noise, and not only as representations, but as participations (Kelty 2019). Acoustic ecology listens from the inside out; if you listen to Westerkamp’s work, there are no distinctions between the sounds of cities and the sounds of barnacles or fish or the wind; rather they coalesce in the ways she herself listens to and records, and they emerge in a form that makes sense, sometimes with her narration, sometimes without.

In what we created, there was much attention to the mix of substantive material, sounds and musical elements, the environment of Los Angeles, and something like a path between simply presenting research, and attempting to use the form to theorize or discovers something. Can we learn by listening? Is it a method, or perhaps better understood simply as a practice like breathing or holding a pose? Can you argue about listening? Can you validate it or falsify it or test it? I am all for experiment and creativity, but I am also committed to the truth of words and worlds.

Part of the willingness to explore this form, speaking for myself, is that I have always been fascinated by the material and formal aspects of what people create—whether that is a quilt or a piece of pottery, or software, writing, music, a scientific fact, or a podcast. These formal aspects include the specialized tools that people use, the techniques that are both explicit and tacit, and the more abstract relation between the vague idea, the more or less concrete image, and the thing produced. I think it is always worth the effort to learn to use these tools, and to avoid treating the technical as a magical domain, or a domain of outsourced labor inferior to the work of thought and writing and narration. The tools create constraints, and the constraints are often sources of creativity and insight—the more deeply one knows those tools, the more elegantly one can respond to them. Rather than just a downstream phase of technical something-or-other, it is a zone of expertise and experts—which is why we were lucky to find an audio engineer-slash-artist who was willing to think with us about these podcasts.

In the worlds of academic production we have watched digitization transform everything we do over the last 50 years. For me, some of this is very central to the substance of my work (Kelty 2012), not just a question of how it is produced (Kelty 2008). Databases and structured programming languages have utterly transformed most scientific fields, as well as regions of the humanities and social sciences. One need not be a positivist about these tools to rely on the for your work or to experiment with them. Even the basic tools of the hermenaut have changed: armchairs are more ergonomic of course, but also the basic tools of writing and thinking and note-taking, to the mechanics of publishing, the ability to use images, video, and audio.

As such, the podcast we produced ought not be viewed as an attempt to “make a podcast” as if that form were settled and obvious and just something to aim at. Rather, it is the result of years of thinking and experimenting in scholarship, learning how to make tools yield insight—failing at this much of the time, of course, and succeeding only rarely. It is the fermentation of years of experience and reading, observation and thinking, collaboration and, for a while now, isolation. In short, as Dolly Parton famously said: “It takes a lot of time and money to look this cheap, honey.”

Are scholarly podcasts serious? This is the most important question for the scholarly use of them. While high quality and creative podcasts are likely to be taken seriously (Serial, Welcome to Night Vale), this is unlikely to be the case for scholarly podcasts. This has little to do with the quality of any given podcast and much more to do with the very limited social imagination of how scholarly work is produced and legitimated. Scholars are their own worst enemies in this respect. A scholarly podcast could never itself—at least not in the existing scholarly system—really be rewarded for being the vehicle of research, that is reserved for the journal article, the book, or the conference presentation. I would love to see this change, but the forces that restrict it also include a general sense within and outside of the scholarly system, that scholars do not officially speak to each other through podcasts, but can only use them to speak to the public, by “engaging” that imaginary beast. The existing economy of science journalism also restricts the form. Scientists might speak to each other through podcasts, but only obliquely; and scientists might take inspiration from a podcast, but only one from a different domain than their own, probably.

All this is less true in the domains of the humanities and arts that privilege and reward experimentation; but different constraints can operate here as well. “Podcast” for an audience of artists, is probably like to seem utterly corrupt—a monetized platform run by feckless millenials, full of pointless, repetitive engagement, mechanically reproduced attention-sucking, distracted listening of the worst sort. Customers who liked Horkeheimer and Adorno might also like… Besides, there are a lot of other aesthetic tradistions of sonic production that contain and precede the podcast. Avante-garde sound art is a thing, and it is not a thing you will find in a podcast. Ubuweb is a glorious documentation of the limits and capacities of aesthetic sound (and visual and linguistic) production, all of which participates in an economy of artistic distinction and scholarly criticism, and which is not likely to be confused with podcasting

But it is possible to imagine the podcast as a component of a research process—which is exactly what it has been in our case.

.content.

We we started, I first suggested not a podcast but an experiment. One of the grad students I work with (Chase Niesner) invited another graduate student who helps to run the The Laboratory for Experimental Narrative Strategies (LENS, started by my UCLA colleagues Ursula Heise and Allison Carruth). We already had an large and untameable archive of material collected in the previous two years of fieldwork, before the pandemic interrupted us: interviews, field recordings, media clips, papers and documents, fieldnotes. The experiment would start with our anthropological archive, if archive is not too overwrought a term here, since the organizing principle is not preservation or documentation. What we might normally use to craft a scholarly paper, we wanted to use instead to craft an aural experiment.

Our archive had over a hundred interviews with all sorts of people, from LA residents who responded to a flyer, to experts on various topics, to recordings from ride-alongs and observations at various sites. Such material, along with scholarly literature, news reports, social media posts and discussions, was mostly made of the sounds that humans make. It was only later, and perhaps still in the future for us, to approach it as a world of sound that other-than-humans make, to say nothing of the stories they tell. But for the time being that means our approach was primarily in the tradition of radio documentary, and less in the tradition, for instance, of audio ecology. The archive consists of the work of close to 25 different students and myself on 5 or 6 different projects related to nature in Los Angeles: Mountain lions, rat poison, wildlife management, feral cats, coyotes, wetlands restoration, green new deals, landscape architecture, lawns and grasses, droughts, sustainability and resilience, to give a few keywords.

Our organizaing principle, such as it was and is, is that there is no such thing as a nature/culture split (i.e. one nature everyone resides in differently accordingly to their culture or history or subjectivity, the classic space of cultural relativism) but rather a nature/nature split: individuals and groups make different, and conflicting natures around them, live in these natures differently, protect, exploit, or fear them differently, and ultimately have to live with the consequences of such choices, and especially, the consequence that other people live in another nature than they do. Such consequences are multiple, but a central one for us is absurdity.

The absurd, incidentally is a nice word for this project, since it originally comes from the latin word for deaf or mute surdus. The idea that what makes sense, logically or rationally, is identified with what is audible is an old philosophical idea, dispute even. But what makes the idea of the absurd valuable here is that it concerns not just that which can be heard, but that which is noisy or too audible—too much meaning, not a lack of it.

A larger political point here is that this principle of a nature/nature split is also part of project to provincialize the Anthropocene, since it is clear that the current configuration of that concept is one that simply re-inscribes the exceptional and distinctive role of humans (that humans are special now in a bad way, not a good way, as the only creature capable of changing or destroying a climate). Rather, we wanted to take seriously the urban and regional responses to the climate crisis that we could identify in our home: the Los Angelocene, the Encinocene, the Glendalecene or the Lakewoodocene. Not only do I insist theoretically on the priority of difference, which manifests itself everywhere, but we observe, objectively, that the real and lasting responses to, for instance, climate change and mass extinction, are first and foremost parochial—which is to say, made according to a nature/nature difference close to the local, and not by the global urge to address climate change as a 1.5 degree problem, or by trhe IUCN or by homogenizing the anthropocene. Climate Change is global, to be sure. But it is not universal. It creates difference and inequality everywhere. The reconfiguration of the planetary responses at the local level are one of the reasons why our current practices of governing animals in Los Angeles seems absurd.

The thing about large data sets in anthropology is that they are never structured avant le lettre, and they are never used in their entirety. Often an hour long interview contains only one or two snippets that really matter to an anthropologist. Much in the same way that an archaeologist can unearth several tons of shards, or an entire midden of shells, but only a few of them are exemplary, only a few demonstrate the meaning and context of the rest with clarity. This lapidary approach to the words, images, actions and gestures of people—and in our case, of non-humans as well—is not restricted to anthropology; it is part of a hermeneutic and interpretative tradition that includes most of the humanities and arts as well. If there is a difference it is that anthropologists are more likely to put themselves at a certain kind of social risk by involving themselves in social life directly—making friends and kin, interfering sometimes, recording what might not otherwise be recorded (although, footnote, this is its own problem entirely in the age of massive digital surveillance), and these days, for some of us at least, doing our best to do so in ways that are respectful, participatory, inclusive, and not extractive, violent, or overly normative— a result of an enormous and guilty legacy of being the handmaidens of the colonial enterprise. The risk of being an anthropologist today is ironically greater than it was in the past, given this awareness, certainly by those historically on the extracting end of the anthropological endeavor, but now also by younger generations bent on fixing that past, and the future, of the discipline.

Interestingly, it is only just beginning to become a question for the observation of animals. Over the last several years, questions about animals right to privacy have appeared in scholarly literature. The massive expansion of animal tracking and monitoring— whether that is PIT and RFID chips in pets, or radio collars and glue-on transmitters on everything from elephants to butterflies—would seem to have a resonance with the practices anthropologists worry over in the recording of human voices and actions. Do we have a right to record a coyote; do the rat and the mountain lion get to shape the narrative—what, and how, do they mean. These are serious questions not only for anthropologists, but also for conservation biologists, wildlife managers, and pet owners.

Once collected, how do we use this material in anthropology? How do we turn it into science, or scholarship? This “data” is ambivalent. Although the collection of it is systematic, and the rules for establishing its validity are something anthropologists debate and discuss, it cannot really be presented in the way that other kinds of data, like statistical data, can be presented, with pretty graphs and clear rules about sampling and bias. For one, it is often sensitive data: it is hard to get people to talk, and even harder to get them to talk comfortably, openly, and more importantly, to start to reveal their expertise and understanding. We also anonymize these discussions; we try to protect people because we depend on their collaboration for our own ability analyze and understand, and the data don’t really “belong” to us anyway— we don’t have the rights to do just as we please with it. A mistake that anthropologists accuse others of making—colleagues in other social sciences, and journalism—is to ignore the need to “learn how to ask” as Charles Briggs put it. To pay attention not just to the knowledge inside brains, but the pragmatic context of talk and the metapragmatic rules governing those contexts (when a recorder is on the table, or lights and cameras on a face, or when certain other kin or relations are present or absent, or the presentation of genders or races).

To make a podcast with such voices is necessarily transformative of those voices, but it need not be extractive. This is why storytelling, contextualization, explanation, and analysis are not subsequent to but constitutive of the anthropological, or more generally, hermeneutic endeavor. This is also why a podcast of this sort can be research itself, and not only the communication of research.

An example: before beginning the podcast project, one of the graduate students had written a paper making use of conversations from our archive of social media posts. These conversations helped to articulate an understanding of certain relations of domination and domestication which were then tied to some current theoretical work. A second graduate student, interested in the same archive, took a different approach to the material, organizing and arranging the social media conversations into categories and disputes that had a certain regularity. Based on this work, a podcast narrative emerged in dialogue among all of us which resulted in the coyotes in the cloud podcast. That podcast, in turn provided both written material and a new analytical frame for revising and expanding the original paper. As such the podcast itself was part of an ongoing research practice.

In a written context, we often process these materials into stories and explanations, arguments and assertions. There is a dance in writing that pairs the material gathered through the experiences of participation and observation with the scholarly literature that precedes it, as well as the concerns and expectations of the people who collaborate with us. It’s more than a “theory and data” relationship. It’s not only about representing the world accurately or objectively, but also about feeling out these other forms of expertise and relation-making which are not part of our scholarly practice. The best anthropology does its theorizing by thinking through the material; not by “applying” a theory to a set of data, or by testing a theoretical claim against some statement or another, but by taking seriously the words and actions and practices of our collaborators as ways of apprehending, explaining, and experiencing the world. Good anthropology channels or connects, it doesn’t extract or purify. If we are lucky to make relations with collaborators, we might learn from them, we might argue with them, and we might help them communicate to others, but it is always a relation of listening carefully, and speaking about what we have heard even more carefullly.

Even though they are about listening, podcasts present a different challenge. These days the word podcast tends to mean “A recording of a conversation between two people with some music at each end.” This is the lowest common denominator of a podcast and although it still takes a lot of work (scheduling, reading, asking good questions, thinking on your feet), it is different from fieldwork. A lot of the material we collect as “interviews” is not actually an interview in any conventional sense, but discussions or debates, or even sometimes just background recordings of things going on in a setting. For instance, recordings of people working or dealing with a problem, recordings of meetings or public events, recordings of a debate between two neighbors.

To make this podcast, we started with just such small units of an interview. Each of us sat with these small snippets as well as a bunch of readings and context and materials related to the snippet. We wrote “mini-episodes” trying to explain what something meant, why it was interesting or surprising or confusing. Some of us used the same clip, but wrote different things about it, which meant that a network of clips and explanations started to developed. We used an online whiteboard to connect them up with concepts and other ideas, and ended up with a network complicated enough to produce anxiety. In part, we were thinking with ecological diagrams: food webs and nutrient cycling. As if an anthropological podcast could be like the life cycle of a lake or a forest, metabolizing ideas that grow and decay.

Ultimately however, the contemporary form of the podcast took over. Telling a story is hard work in any context. One has to be sensitive to the listenter’s limitations. It’s hard to remember from one minute to the next who are the main characters, what they have done, what the argument or plot is. Repetition is essential. Writing out a script, and thinking about how it will have a structure, like a theatre play, where characters come in, action takes place, and then other characters come in, and now you have to connect them up somehow. All this alongside the need to communicate something “theoretical” or scholarly: ideas about, e.g. political life and governance, about the current science of zoonotic disease, about the biological aspects of secondary poisoning, or the historical relations between the countercultural left and satanism. In short, a mess of ideas and concepts and stories that need to be tamed and channeled into less than an hour, with no images and no questions or clarifications.

This is not a failing—the process was very much the product in this case, and the future episodes of the podcast will sound different and will be new forms of research and analysis because of this process. There is always more to say, but saying things is never simple.

References