It's 7:00 p.m. on Friday night and Ira and I are hanging out in the dining room. I'm moving back and forth between the kitchen and the table, but quietly, because he's about to tell me a story. I try to get together with Ira every Friday, but lately I've been busy and I haven't been able to make it. I'm glad I caught him this week though, because tonight is special: it's the 500th episode of This American Life and Ira Glass, its host, has invited me and the rest of his 1.8 million listeners to join past producers as they discuss their favorite moments on the show.

Some of the stories are favorites of mine, and hearing them is like hearing from one of those smart, kind, creative people you knew when you were younger and admired from a distance. Then, Glass and one of the show's long-time producers, Sarah Koenig, do something that surprises me. They start talking about . . . research methods, of all things. ¹

Koenig's point-that Glass's tendency to go "off-topic" is what creates the most revealing and affecting moments in his interviews-makes my ears perk up. As a social scientist trying to merge ethnography and audio documentary, I spend a lot of time trying to reconcile what I learned in ethnographic methods classes with what I've learned in audio production courses. I would have thought I'd be happy about Glass and Koenig's conversation. But instead it's making me uneasy, and I can't figure out why.

As an example of a "surprising" moment typical of Glass's interviews, Koenig plays a clip of one he conducted with comedian Tami Sagher. Sagher is telling Glass about the time she told a joke at a new job. The way she delivered it made it seem like it was a casual, off-the-cuff remark. But the truth is, she'd written the joke years beforehand and hadn't had an opportunity to use it until then. She'd been storing it up, waiting for just the right moment. When that moment arrived, in front of all those other comics, she nailed the joke, the room exploded in laughter, and Sagher was on top of the world.

At this point in the interview, Glass asks Sagher to step back and reflect on what that moment meant to her. Is she "in" now, he wants to know? Has launching that joke moved her from being the "new girl" to being part of the in crowd? A little, Sagher, says. But, then she asks:

What stands out about this exchange is that it does not sound like a conventional interview by a radio show host; it sounds instead like an intimate conversation between friends: Glass calls Sagher "dude," and Sagher, in turn, feels free to tell Glass what is "sucky" about his situation. It is Sagher who is asking questions of Glass, so that the distinction between interviewer and interviewee has blurred.

Koenig points out yet another significant feature of their exchange: Glass is surprisingly open in his responses, especially given that this is an interview that he will eventually make public. What he says about himself in that moment is something that she knows about him, because they've worked together for ten years. But for him to reveal it to a stranger, someone he's interviewing, and then put it on air, "that's ballsy," Koenig says. Glass however, isn't convinced. Here's his response:

Glass asserts that he's doing what any producer would do to create a strong story. But Koenig counters that his level of intimacy and openness is unusual and even a bit scary: she insists that although the principle is one most radio producers agree with, Glass employs it to a degree that many others find too difficult to do.

It takes me a while to figure out why the unexpected methods lesson has made me so nervous. Then it hits me: Glass and Koenig's conversation has highlighted how little I understand about how to make good tape.² Ethnographers and documentary makers often appear to be doing the same thing: we both ask people to tell us their story, we both review what they've said, and we both present an account of the world that illustrates something we understand to be true. In order to do that, we both ask our interviewees to share intimate information with us and we use multiple strategies to help them feel comfortable doing so. But Glass and Koenig's conversation seems to suggest that audio documentary requires a level of emotional (as opposed to an informational) reveal, from both interviewer and interviewee, that I'm not sure I want to engage in and whose implications I don't yet understand. As a new producer, I know good tape is a necessary part of documentary work and I want desperately to learn how to create it. But I'm also afraid of it: afraid of what it takes to get it and afraid of what it does to my scholarship.

These kinds of concerns are likely familiar ones for any traditionally trained scholar who is learning to use creative practice to collect, analyze, and represent data. Ethnographers have long acknowledged the literary nature of ethnographic accounts (Behar and Gordon 1995; Goodall 2008; Narayan 2012; Waterston and Vesperi 2011). And they are increasingly representing their research in formats other than written scholarly texts, including fiction, performance, film, and most recently, audio documentary (Goldstein 2001; Makagon and Neumann 2009; Pink 2009; Vannini 2012; Boebel and Walley 2013). Yet any researcher who is new to cross-disciplinary, participatory, or creative work is faced with her own version of the question that Glass and Koenig have raised for me: Is it my job to get good tape?

In academic literature, this question usually appears as a debate about validity:³ critics assert that the work is art, not science-in other words, that it is an edited, constructed product that reflects the creative and intellectual bias of its maker rather than general patterns that would be observed by another researcher interested in the same themes. Proponents, in turn, insist that all knowledge is constructed: that creative-research work is merely more explicit about that construction process and that this form of scholarship is superior at capturing and expressing the emotional aspects of social life that traditional methods cannot (Barone and Eisner 2011). While it's certainly useful to explore the relationship between science and art, this debate misses a more pressing and practical question raised by ethnographers' forays into audio documentary. To clearly see that question, it's useful to consider what's so great about good tape.

 

If I'm Lucky, I Have Tape

As Glass and Koenig suggest, good tape is partly dependent on interview style-or more accurately, on the relationship that producers are able to establish with their interviewees. The strategy that Glass adopts is what researcher refer to as a "dialogic interview," in which either party may be asking questions, answering them, commenting, and interjecting, so that "both the interviewer and the interview partner generate meaning together" as they speak (Marshall and Rossman, 2011, 144). The assumption with this technique is that neither the researcher nor the subject/partner is the expert, but that they both gain understanding about a topic through discussion with one another. Scholars argue that because this kind of interview involves the exchange of personal information (rather than its extraction by the interviewer), it helps mitigate the fundamentally unequal power relationship between researchers and interview participants.

But good tape requires more than the use of a conversational, confessional interview style. When Glass and other audio producers refer to good tape, they are talking, at base, about broadcast-quality sound. Unlike the tinny, muffled recordings typically produced by ethnographers when we slip our recorders onto the kitchen table, good tape is clear, audible, and undistorted-akin to what a listener hears on an NPR newscast. Sound of this quality helps creates a sense of intimacy, of "being there," partly because it has what audio producers call a large "signal-to-noise" ratio. The signal

is the sound you're trying to record, like a person's voice. The noise is all the other sounds in the environment, like the refrigerator and the traffic and sound waves echoing off walls and floors and ceilings. Generally, good audio recordings have a lot of signal and very little noise (Carrier 2013).

Good tape also refers to a recording that reveals something the producer has discovered about the world, a broad principle that many people can identify with. While an audio documentary can make these points through narration that is written and read by the host or the producer, good tape makes that point more powerfully, typically by capturing the sound of a place, reflection, experience, or interaction that feels human and authentic and relatable.

One reason pieces of good tape are so powerful is that they humanize social analysis. Compare, for example, the difference between these two analyses, both of which describe how the loss of culture and social networks impacted Hurricane Katrina survivors who were permanently displaced from New Orleans in 2005. The first is written by Lee Miller, contributor to the edited volume Displaced: Life in the Katrina Diaspora. She argues that

two losses universally experienced by the displaced were less tangible and harder to measure and yet made rebuilding their lives difficult: the loss of culture and the loss of support networks. . . . New Orleans was home to a unique culture, and the comfort of its traditions, cultural events, familiar accents, and favorite foods was impossible to find in new locations. Also missing were the social ties to family, friends, and neighbors who helped each other meet the demands of everyday life-from providing information and advice to sharing transportation and taking turns with childcare. As evacuees were scattered across the United States, many lost these networks, making their lives increasingly difficult (Miller 2012, 26).

Miller's analysis is clear, concise, and accurate. Despite the fact that the book is primarily targeted toward academics and decision makers, it is written in a style that is straightforward and easy to understand. She gives specific examples that illustrate what she means when she refers to "the demands of everyday life" and the reader has a clear sense of the obstacles that stood in the way of people who fled New Orleans after the hurricane. It's also obvious that Miller feels a great deal of compassion for the struggles of the people she writes about. In other words, hers is a strong piece of writing, not the impenetrable, alienating, tedious kind that bores even scholars themselves (Richardson 1994).

Now listen to the same idea, as expressed by Cedric Johnson, editor of The Neoliberal Deluge: Hurricane Katrina, Late Capitalism, and the Remaking of New Orleans, when I interviewed him for an audio documentary on the same topic. Johnson grew up in Opelousas, LA, a few hours west of New Orleans, then left home to attend college and graduate school (Johnson 2013). He explains:⁴

Both Johnson and Miller are scholars who have researched and written about the causes and effects of Hurricane Katrina. And Johnson's verbal analysis shares Miller's clarity, accuracy, and empathy. But the audio clip contains two elements that the writing does not. First, Johnson's verbal explanation combines personal experience and professional analysis. He did not live in New Orleans when Hurricane Katrina hit the Gulf Coast, but he draws on his own experience leaving Opelousas to explain what scholarship has taught him about the nature and significance of social networks and local southern culture. Second, the audio clip humanizes the analysis, partly because it features a person's voice, but also because it contains a degree of expression and emotion that shows up in Johnson's word choice, volume changes, inflections, emphasis, and pacing.⁵

This ability to humanize both the interviewee and the researcher is particularly important for those who do research on marginalized communities, since dehumanization is a fundamental tool of subordination. And it is one of the primary strengths of all interview-based, recorded research. But because scholars primarily communicate through writing, the power of sound is usually lost, whether through the transcriptions of oral historians, the vignettes of ethnographers, or the excerpted quotes of sociologists. In contrast, audio documentary maintains that power, because its final format is aural. When interview, narration, ambience, and music are woven together in an artful portrayal of social scientific knowledge, the listener "can grasp something at a sensuous level that is considerably more abstract and difficult to convey in a written ethnography" (Feld and Brenneis 2004, 465).⁶

A second reason that sound is so powerful is that it activates listeners' imaginations so that they are co-creating their experience with the producer.⁷ This aspect of sound is the reason independent radio producer Scott Carrier describes it as the "show" part of "show don't tell," that well-worn admonition that guides creative writers and audio producers alike. Because good tape captures what someone experienced in the moment, it "shows" the listener what happened. Writing, on the other hand, is the "tell" part of the equation, because it's what producers use to contextualize and give meaning to good tape. He explains that tape is much more valuable because when you tell people something they forget it, but when you show it to them, make them imagine it in their own minds, they remember it. This is how it becomes real. So I look for the tape that makes this happen. If I'm lucky, I have a lot of good tape. If I'm not, then I have to write. (Carrier 2010, 29)⁸

Carrier's assertion that audio work has great capacity for impacting learning and retention is supported by social psychology research on visualization and simulation. Because good tape helps the listener visualize the world being described to them, it can also stimulate the imaginative and cognitive processes that help people understand, absorb, and retain what they hear (Heath and Heath 2007).

These two strengths—the capacity to humanize and to foster engagement—clarify the limitations of the debate over audio documentary's validity. While it raises important questions, this debate assumes that researchers' only concern is knowledge production. Yet, the growing institutionalization of "public" social sciences and humanities shows that scholars are equally concerned with knowledge impact and accessibility. Even scholars who are not particularly interested in art or public engagement have a desire to share their work and findings in ways that convey their passion for their research and its significance for the world. Good tape is worth making because it allows researchers to create memorable, understandable, emotionally compelling work that explains broad social patterns to a public audience.

Yet in order to produce that kind of work, to expand past the inner circle of academic conversation, ethnographer-producers must develop new skills in capturing and editing sound. We must also think through what these new techniques mean for how we understand and practice scholarship. Very little work exists though, to help academic audio producers tackle these issues. Producers often argue that anyone can make good audio. As a result, they've written extensively about the technique required to get good tape (Abel and Glass 1999; Glass 2010; Hardcastle Jones 2013; Makagon and Neumann 2009). Yet none of these sources explicitly considers how the process of getting good tape reshapes our traditional process of knowledge creation. Thus, for scholars-turned-producers like me, the more pressing question isn't whether making good tape is our job; the question is how making good tape changes our job.

 

How Not to Say Hello

"Ring left, record right," I mutter under my breath, as I look for the name on the buzzer. I'm standing outside the home of Linda Grady, trying to remember in which order and with what hand I'm supposed to do all the things I need to do. I'm wearing a pair of fat, black, round headphones, and holding a seven-inch microphone in my right hand. In my left is my portable tape recorder. It's not big, about the size of four iPhones stacked on top of one another. But there are also two cables drooping from the recorder—one to the mic, and the other to the earphones—and every time I change my mind about what object to hold in which hand, they become more and more entangled.

With the hand that's holding the recorder, I press the buzzer, remembering to hold the microphone up to it so I can record the sound of its bleating. I'm so focused on readying myself to record Linda as she opens the door, that I don't even notice the sound of a screen door sliding open, or Linda emerging from the wall to my left. Soon enough I hear a faint "Hi" and turn to see Linda standing half in and half out of her patio doorway, a small smile on her face. "Mrs. Grady?" I ask, sticking out my mic hand, realizing too late that it's not available for shaking. I shift the mic to the hand that's holding the recorder, trying not to bang the two together, and pump Linda's hand in spastic syncopation. As she invites me inside, I quickly recover my mic. "I'm already recording," I say in an overly bright tone. "I always have my recorder going in case anybody has anything important to say right away!" As if this happens all the time. As if I have the faintest idea what I'm doing.

Linda and I have talked on the phone, but this is the first time we've met in person. I'm taking my third audio production class, doing a piece on extreme weather and racial displacement, and Linda has agreed to let me interview her. She used to live in New Orleans, but she is now part of the "Katrina Diaspora," the group of people who left the city after the 2005 hurricane and did not return. Contrary to my insistence that this is what I do "all the time," I am behaving in ways that go against every ingrained habit I have about how to start and conduct myself in an interview. Part of my awkwardness is just a lack of experience: I've done lots of ethnographic interviews, but only a handful of audio ones. But part of my awkwardness reflects the fact that audio documentary interviewing is a technically different interaction than its academic counterpart.

Audio interviewing requires not only that we record the story of what happened to someone, but that we capture and manage sound in ways that aid the telling of that story. One way we do that is by capturing ambient sound, the adjectives and adverbs you would typically read in a written document. You hear the sound of the doorbell, the creak of the door opening, the pneumatic hiss of a screen door swinging back into place, and you can easily imagine that you are right beside me, walking into someone's home. Because it helps listeners create an internal image of what they are listening to, ambient sound is crucial to audio documentary's emotional and intellectual impact. As independent audio producer Robin Wright (2013) says, "If radio is like a magazine, ambient sounds are the photographs."⁹

A second way audio producers manage sound is by eliminating the sounds we don't want, what I call intrusive sound. This form of sound management is even more important than the first: If you're recording in a pasture and the cows are silent, there are online sites from which you can borrow or buy a wide variety of moo.¹⁰ But say the cows are mooing in a field next to the interstate, and trucks keep rumbling by; not only can the trucks make it difficult to hear the cows, but the intermittent background noise will make it difficult to create a piece that sounds like it was recorded in the same place at the same time.

Having to manage sound in these ways changes a fundamental aspect of qualitative research interviewing: it deprives us of at least some of the self presentation strategies we typically use during the initial moments of establishing rapport with the person we are talking with. Ethnographers in particular take a special pride in our ability to make our presence in a new community as unremarkable as possible. We spend weeks, sometimes months, making silent observations of the communities we want to be a part of, gaining access to key informants, and building a sense of trust with the people who we will eventually ask for interviews. Especially when we're interviewing someone whose community we have not immersed ourselves in, we rely heavily on the skillful and strategic deployment of interpersonal skills to make our interviewee feel comfortable talking with us.¹¹

But as I step into Linda's house, very few of my old strategies feel available to me. In my effort to capture ambient sound, I walked from the car in a tangle of equipment that easily made me the oddest-looking person around. And in my effort to eliminate intrusive sound, I'm on high alert the minute I enter her house, my eyes and ears already roving over her belongings, seeking out the worst aural offenders and trying to figure out how to tell her to turn them off. Instead of making Linda feel at ease—beginning with chit chat to promote the idea that we are going to have a casual, friendly conversation—I am creating more awkwardness by announcing that I'm recording within 5 seconds of entering her home. This is me just seconds after I arrive:¹²

Linda follows me around, gamely silencing, unhooking, and dismounting whatever I point out. "Be nice but firm," my teacher says when we're in class. "You're the professional. Be in charge." I get what she's saying, but I don't feel like the professional. I feel like the Sound Police, and this feels like an aural stop and frisk.¹³

In these circumstances, researcher-producers need different ways of establishing and maintaining rapport with the person they interview. Happily, I stumble upon one only a few minutes after my aural takedown. It happens when I notice the wall across from Linda's patio door, which is covered with family photos from top to bottom. It's the first thing you see when you walk in, at least if you're not a newbie at hunting sound. When I finally notice it, here's what happens:

When Linda begins describing her pictures, I immediately perceive the change in our interaction. Suddenly, I'm following her around, and she's the one talking, not me. When I do talk, I'm asking to hear more about the pictures instead of laying out a set of rules for her to follow. There are no Ira Glass moments; no big reveals that produce really good tape. But it's the first moment in the interview that I'm able to do what every interviewer aims for, whether they're creating ethnography or documentary: show the person I'm talking to that I can be trusted to care about what she has to say.

That happened because I found and asked about something that, had I been thinking, I would have realized earlier was incredibly important to Linda. I had no immediate sense of how the pictures might figure into the piece I was trying to create. But Linda fits into the piece I was trying to create. And as someone who lost even the concrete slab upon which her New Orleans home was built, the family photos and framed articles about the storm that lined her wall were a touchstone for her, a pleasure and a balm to consider. In some ways this technique is not so different from something I might do in an ethnographic interview: ask someone about something that matters to them. The difference is that, because I'm distracted by the technical aspects of recording and controlling the aural environment, I'm distracted from the equally important task of connecting with Linda. The trick, I discover, is to look for what she cares about before listening for what I'm worried about. Yes, it's important to capture high-quality, undistorted sound that gives the listener a sense of place; and doing so requires that I give a fair amount of attention to the technical aspects of recording. But the heart of any audio documentary is the relationship on which it is based—it's this relationship that ultimately allows producers to make such odd demands about the quality of their sound environment.

Making such simple, seemingly obvious discoveries is both humbling and strangely delightful, and it reminds me that the fun of academia isn't what you know but what you learn. That feeling, and the insight it produces, is one of the greatest benefits of being a researcher-turned-producer. It's a good thing I feel this way, because I've got a lot more to learn. Even as I celebrate the victory of getting Linda to talk, I encounter yet another problem I've never before faced in an ethnographic interview.

 

Dance With Me

Linda's description of her family photos has put me back on familiar ground: instead of demanding that she modify her home for our interview, I am asking questions, prompting her recollections, serving as the receptacle for her story. With my attention on her, rather than my own incompetence, things are starting to look up. But because Linda is standing before a wall, walking its length, and moving from picture to picture, my mic and I have to move as well, and we have to do so in a way that maintains the volume of her voice. Several times I have to reach out to keep the mic in range of her mouth, then fall back suddenly when she turns to face me or reaches to point out a picture. I walk behind and to the side of her, bending awkwardly, almost hitting her in the mouth several times. In my headphones, I hear her voice dropping and rising in volume as she moves away, then back again, my mic lurching back and forth as I try desperately to track her movements.

As she describes her family members, my mind drifts, just a second, as I remember the first time I realized that making good tape might require something other than plunking a recorder down on a coffee table. It was during my first audio course at the Center for Documentary Studies, when I asked the teacher what to do if the perwson . . . moved.

"Are you just supposed to follow them around with the mic in their face?" I asked.

It was the first day of the class, and we were all sitting in a big circle, having just introduced ourselves. To show me how it worked, the teacher made me walk around the room introducing the people I'd just met. I was nervous, the way everyone is when asked questions with a fat mic four inches from their face.

"Don't worry about the mic," he said, as I craned toward him. "Worrying about the mic is my job."

He followed me around the circle, somehow keeping his mic close enough to me when I looked at the person I was introducing, then swinging it away gracefully when I looked back at him. He even asked a few quiet questions so that I'd have to turn his way, each time catching the easy rhythm of my movement as if I'd told him what I was going to do. It had all seemed so natural when he did it, like a lovely, lilting two-step whose logic would settle into my body the minute I stepped on the dance floor.¹⁴

"Oh I get it," I remember saying, as we finished our mock moving interview. "I think I can do that."

As Linda's voice fades yet again, then seconds later, comes blaring back into my ears, I feel both exasperation and sympathy for the fool I was when I spoke those words.

My botched attempt to capture Linda's narration of her family photos points out yet another dilemma that audio recording poses for ethnographers. It contains an unfamiliar physical dimension, one that—like ambient and intrusive sound—changes the way we interact with our interviewee. Ethnographers typically record so that we can pay close attention to our respondent instead of being worried about capturing everything we've said in notes. Because we are the only ones who ever hear those recordings, we try to make our equipment disappear. It is the transcript that matters, that makes possible the coding, categorizing, and theorizing of data that makes its way on to the page. Making broadcast quality recordings, on the other hand, requires that we intrude on people's personal space by sticking a mic in their face, and keeping it there for the duration of the interview. Recommended distances vary with the mic, but with the starter kit I own it's about the width of your hand. In order to get the mic that close to Linda, I sit on one side of her square, glass dining room table and ask her to sit on the adjacent side. We fold ourselves awkwardly around our shared corner, shifting uncomfortably and occasionally knocking against one another's knees.

This situation is uncomfortable for everyone involved, physically and interpersonally. I know from previous interviews what will happen. Linda will look uncertainly at the microphone, trying at first to stay within range, then slowly and unconsciously, she'll back farther and farther away from me until she is on the other side of her chair. I, on the hand, will try as hard as possible not to move my fingers or my arm, lest I make intrusive noise through handling the mic. Muscles will cramp. Limbs will go numb. Long dormant running injuries will flare up. The only movement that will seem possible is the slight dip of my neck as I constantly check the volume level on my recorder. As my instructor has pointed out, it's not enough to hear the sound through your headphones. You also have to check the readout on your recorder to make sure that the volume of the recording is neither too high nor too low, but that it hits an electronic sweet spot.

Ten minutes earlier, my obsession with ambient and intrusive sound had disrupted my strategies for establishing rapport. Now, my concerns about capturing Linda's voice are disrupting my strategies for maintaining it. One of the simplest ways I have of maintaining rapport is to show her—with eye contact, facial expressions, and our verbal back-and-forth—that I am interested in what she has to say (Marshall and Rossman 2011). But it's difficult to give someone the sense that you are listening when your body is tensed in silent agony and you're constantly glancing down at your recorder.

Instead of clinging to what I've learned as an ethnographer, I surrender to what seems most counterintuitive. I cease pretending that there is anything normal about this interaction, and instead bring attention to how weird it is. Before I ask Linda a single question, I tell her why I'm wearing the "geeky" earphones. I explain why she'll see me look down at my recorder so often, but I assure her that I'll be listening to everything she's saying. I apologize for how close the mic is going to be to her face, but explain that she has a beautiful voice, and I want her to sound as good as possible in my piece. What happens then is what always happens to me when I sit down with another human being who has agreed to tell me their story. She somehow accepts everything that's weird and uncomfortable about this situation, as well as the clumsy way I've handled it.

I've always been amazed by that, by the casual generosity with which people share themselves when asked. And I've been equally amazed by the substance of what they've shared: the depth of their knowledge, the frankness of their disclosure, the passion of their beliefs. What good tape helps researchers do is to better convey those elements, and thereby render a fuller, more complex picture of the people we talk to and the social patterns that shape and constrain their lives. But getting good tape, while it looks a lot like ethnographic interviewing, requires that we strengthen our lesser-used sense of hearing, and modify our familiar strategies for making human connection. Fortunately for me, there are people like Linda, who, even now, is doing all she can to help. Just as I'm about to ask my first question, she turtles out her neck, trying to improve our recording by sticking close to the mic. "Don't worry about the mic," I say to her, with all the courage I can muster. "Worrying about the mic is my job."

 

Works Cited

 

Notes