On Radiolab: the Sound of Science

Jad Abumrad was talking about “tension” — the tension between the certainty of science and mysteries that inspire wonder; between authenticity and artifice; between a sound that feels carefully constructed and one that feels anarchic and alive. “You want to seduce people,” he said. “But you also want to disturb them.” Abumrad, who is 37, is the co-host and producer of “Radiolab,” a public radio show that breaks from public radio sensibilities, not least in its striking sound. “I put in a lot of jaggedy sounds, little plurps and things, strange staccato, percussive things — people don’t like that so much,” he said. “Some people don’t, anyway.”

We were in Abumrad’s small and ambivalently decorated office at WNYC. I had spent a few days hanging around as he and his co-host, Robert Krulwich, and the show’s half-dozen or so staff members worked long days to finish an episode for the latest “Radiolab” season, currently being broadcast on several hundred public radio stations around the country. Jaggedy plurps and all, “Radiolab” has since 2005 developed a devoted following for its unconventional approach to both the medium and the message of radio. Its five-episode seasons now attract about a million listeners who hear them over the air, and 1.8 million who, significantly, listen via podcast. Abumrad, who planned on a career as a composer, not a broadcaster, has become a star among producers for his creative sound design. “Radiolab” fans, who tend to be younger than typical public radio listeners, are rabid, selling out most live events in hours. Ira Glass, the “This American Life” creator and progressive-radio hero, has become an unabashed booster, declaring that the show has “invented a new aesthetic for the medium.”

A relevant question to ask at this moment is: Why would anyone bother to invent a new aesthetic for such a retrograde form? This is an exciting time for innovation in new media: interactive forms for active consumers. Radio, in contrast, just washes over you or drifts by in the background. It seems ill suited to an audience that multitasks, demands to react or contradict in real time, insists on controlling information rather than receiving it. Yet “Radiolab” — which just won a 2010 Peabody Award — has responded to all this by designing a show for sustained and undivided attention. It wrestles with big, serious ideas like stochasticity, time and deception. It ignores the news cycle completely. And it expects you to stop checking your inbox, updating your status or playing Angry Birds and spend a solid hour listening.

An episode from the current “Radiolab” season titled “Lost and Found” begins with a woman named Sharon Roseman describing her childhood discovery that she was afflicted with a bizarre condition that scrambled her sense of direction to the point of nightmarish dysfunction. We hear from the doctor who gave the diagnosis, an expert on the brain (who illuminates the mechanics of location cognition) and finally from Roseman, again, on how she found she was not so alone as she had believed. The segment lasts 19 minutes, a long time to keep listeners interested in a condition we’ve never heard of, and likely never will again.

On first listen, the bantering tone struck by Abumrad and Krulwich, and their palpable chemistry, stands out. They joke, challenge, openly admit to not fully understanding their expert guests and give the general impression that they’re having a ball. They don’t sound like an obvious match, and when I met them in a WNYC conference room during the production of the “Lost and Found” episode, they didn’t look it either. Abumrad is trim, bespectacled; his parents, a surgeon and a biologist, emigrated from Lebanon to Syracuse, where he was born, and then moved to Nashville. That’s where Abumrad grew up and discovered the solitary pleasures of music and composition. His earnest, thoughtful demeanor in conversation is a quieter and more polite version of his on-air style.

Krulwich, who is 63, reminds you of your favorite college professor, full-blown and voluble, unable to go more than four minutes without making everyone in the room laugh. Probably best known as a network television correspondent, he is also a storied figure in public radio history and dreamed of being a broadcaster since childhood, when he would listen to a transistor radio under his pillow. (“Really?” Abumrad asks. “No, you would not.” Yes, Krulwich insists; he’d pretend to sleep when his parents peeked in on him — a scene he punctuates with a theatrical snore. “That’s insane,” Abumrad says, laughing. “I never had that relationship with radio.”) What they seem to share is a blend of curiosity and skepticism, a willingness to be convinced — and a delight in convincing.

Of course, from “Car Talk” to any number of drive-time zoo crews, informal banter isn’t exactly a new formula for successful radio. But to focus on the “unscripted” elements of “Radiolab” misses the point. This is an elaborately constructed show. The initial interviews for the Roseman piece took place weeks earlier and were edited into a draft that included mostly improvised voiceovers and commentary. When I was in the studio, the pair sat down with Soren Wheeler, a producer, and spent an hour and a half debating various phrasings and cuts, mostly focusing on three or four passages. That afternoon they assembled again to rerecord their chatter, improvising some bits a dozen times in a row. It took another hour to agree on a minute or two of usable sound — some of it later rejected and recorded again.

The banter, then, is not really what defines the “Radiolab” experience: it’s the aggressively edited context in which it resides. People clearly interviewed separately seem to answer one another’s questions or speak in unison. Elements pop up that are sliced away in most radio productions, like interviewees spelling their names during a microphone volume-level check. In the initial interview with Roseman’s doctor, it emerged that he had never met her in person. “What if you, like, meet her and fall in love with her?” Krulwich abruptly asked. “The science is ruined, but you get married!” This hammy interlude would be, in any classical form of information gathering and dissemination, a waste of time. The doctor (“I am already married”) sounds embarrassed for everyone involved. But the passage makes it into the final cut.

“I guess it’s there, it’s there because — ” Abumrad began when I asked him what the point of including such digressions might be. He paused. “You’re trying to capture the rhythms and the movements, the messiness of the actual experience.” This raw bit of tape captures a real moment, he continued, “and I’m always trying to figure out where we can place those artifacts that create a sense of transparency.”

“Normally a reporter goes out and learns something and writes it down and speaks from knowledge,” Krulwich added. Jokes and glitches puncture the illusion of the all-knowing authority, who no longer commands much respect these days anyway. It’s more honest to “let the audience hear and know that you are manufacturing a version of events,” he noted later.

“It’s consciously letting people see outside the frame,” Abumrad suggested. “I think those moments are really powerful. What it’s saying to the listener is: ‘Look, we all know what’s happening here. I’m telling you a story, I’m trying to sort of dupe you in some cosmic way.’ We all know it’s happening — and in a sense we all want it to happen.”

This is how “Radiolab” addresses the tension between authenticity and artifice: capturing raw, off-the-cuff moments (or trying many times to get them right) and editing them in a gripping pastiche. The hope, Abumrad said, is to preserve the sense of excitement and discovery that often drains away in the authoritative accounts of traditional journalism. “It’s a funny thing,” he concluded, “when you find yourselves laboring for weeks to create what you felt at that first moment.”

I heard several episodes of “Radiolab” before I figured out that it was supposed to be about science. I thought the “lab” part of the title referred to experimentation with the medium. A lot of listeners have had the same experience — and technically, we’re right. The original “Radiolab” was a three-hour weekly show on WNYC’s AM station, which played documentary radio work, with Abumrad as host. It was conceived, back in 2002, as a space for experimentation, and also as a way to fill a “blank space” on the station’s Sunday-night schedule, says Dean Cappello, chief content officer and senior vice president of programming for WNYC Radio.

Abumrad was cast as a kind of free-form D.J., spinning documentaries. But instead of the vamping-for-the-mic style that this description invokes, his interstitial segments were intricately produced, with unusual sound beds and monologues that drew surprising connections. “What Jad was doing was actually more interesting than the pieces we were trying to showcase,” Cappello says. “What if we have an hour of Jad doing that?”

It may surprise regular listeners that Abumrad never studied science — and in fact he has said that he doesn’t particularly like it; he’s just curious about the world. He studied music at Oberlin, though he found composing for actual musicians invariably disappointing. What he loved was the iterative process of making sound by editing tape samples, or coaxing it out of early synthesizers, in a music-and-technology class. Radio seems to have occurred to Abumrad as an afterthought when a few years out of school he realized that his dreams of film composing weren’t working out, and he hated his late-1990s Internet job. He volunteered at WBAI, quickly ended up on the air and started contributing freelance pieces to WNYC and NPR programs.

Then he met Krulwich, with whom he had WBAI and Oberlin in common. They hit it off, and lively discussion over breakfast became a regular ritual. Krulwich was part of NPR back in the 1970s, when its informed-yet-conversational style represented an earlier “new aesthetic” for radio, and he was praised for his creative coverage of business and economics, including, famously, a radio opera about interest rates (“Rato Interesso”). He left in the mid-’80s to work in television, specializing in complex topics. At first, he gave his young friend a lot of mentorish advice. Until, he says, “I realized that he knew things that I really didn’t know. There were beats in him I didn’t have and had never heard before. And that’s pretty intoxicating.” They collaborated on a piece for “This American Life” — which Ira Glass cheerfully describes as one of the worst things he has ever heard, and which did not air. But by 2004, “Radiolab” had evolved into a one-hour show with a tighter focus, and Krulwich started joining in as a regular guest. This eventually led to a five-episode season with the pair as co-hosts.

While working with Krulwich loosened up Abumrad’s on-air presence, and he learned the ropes of professional journalism, Abumrad never really stopped thinking like a composer. That is, he thinks like someone interested in how sound makes a listener feel. When he talks about influences, the most prominent name is Walter Murch, a legendary film editor and sound designer.

Listen more closely to a “Radiolab” episode, and once you get past the cutting up and the jumpiness, you’ll notice just how intricate the underlying soundscapes are. Consider the second segment of the “Lost and Found” episode, about pigeons’ freakish sense of direction. In its most sciencey passage, Tim Howard, the reporter, presents an explanatory theory from an expert described as “a heavy hitter in the pigeon world.” It goes like this: Churning iron in the center of the earth throws off lines of magnetic force; a pigeon flies through the lines and can perhaps “feel them” by way of particles of magnetite in its beak. During my visit, Abumrad listened to a minute-long edit of this passage with Howard and Wheeler. “Do you know Alvin Lucier’s ‘Music on a Long Thin Wire’?” he asked Howard when it ended. “I’ll play it for you.” He had an idea for the sound — not a sound effect, and not music, but a “musical gesture” — to play against the dialogue. “The sound’s going to be going bruup bruup bruup,” he told Howard, advising him to take the pigeon’s point of view. “It’s moving — fhewm, fhewm — through bands, some are thick, some are thin. You know? That’s the part where it’s gonna feel very visual.” Ultimately, it took him and Howard hours to construct the track, using the audio-editing program Pro Tools, and a bunch of audio filters.

I asked Abumrad what a traditional radio producer would make of his meticulously constructed bruup bruup fhewm fhewm. “They would say it’s insane,” he said. Early on, he had to deal with “radio people” who thought he was wasting time on “artsy-fartsy namby-pampy” technical distractions. “But do you want to know why ‘Radiolab’ has worked beyond public radio?” he asked. “Because it sounds like life. You watch TV, and someone has labored over the feel. Look at Mad Men or The Sopranos: the mood, the pacing, the richness of it, comes from those fine, quote-unquote technical choices.”

The final segment of “Lost and Found” revolved around a young woman hit by an 18-wheeler, and it largely concerns her boyfriend’s stubborn refusal to believe that she will never return from the semicomatose state into which she has disappeared. The 20-minute piece is “sort of a love story,” as Abumrad says in the final cut. Emilie, the woman, could not see or hear, but her boyfriend, Alan, was determined to “figure out how much of her was actually there, and maybe even prove it to the doctors.” At one point he uses his phone to make an audio recording, narrating as he tries to communicate with Emilie by finger-spelling words onto her hand. For more than 10 minutes, we’ve heard only Alan and Emilie’s mother talking about her. So when we finally hear Emilie’s voice answering his questions — “I don’t know where I am,” she says — it’s electrifying.

The piece is about the boundaries of the conscious self, and how medical science grapples with that mystery. But the question of what science can tell us and what we can only wonder at with awe is not so much settled as dramatized. The piece practically works as a radio play. You could listen to it again and again, now or a year from now. Actually, that’s the goal: Abumrad has often said that he edits with the fourth or fifth listen in mind.

This approach — a smaller number of shows, painstakingly assembled and treated more like small movies than like regularly scheduled programs — addresses a different tension, around new habits of media consumption. That is the tension between relevance and disposability. Discussions of technology and media tend to focus on speed — what’s the fastest way to break the story, consume the story, influence the story? After all, media consumers today seem like info-rats chewing through heaps of micro-facts and instant-expiration data points.

But the other interesting thing about media these days is that it can stand perfectly still. In fact it loiters: shows don’t simply spill over the airwaves and evaporate; they linger on DVRs, DVDs, various online services. Newspaper articles pile up in Web “archives.” And clearly we still accept, still crave, some deeper media experience too. In experimenting with a show that produces (at most) 10 episodes a year, WNYC was specifically thinking of HBO’s success in building powerful cultural franchises that ignore the mores of traditional broadcasting.

Yes, radio drifts by or washes over you when it comes out of a box on the other side of the room — but remember, a majority of “Radiolab” listeners actually take in the show via podcast, and there’s something different going on when it enters your head through earbuds at the exact moment you have chosen to hear it, while you’re commuting with nothing else to think about, or cleaning the kitchen, or lying down for the night. “Artistically, these are the glory days for producers, if they really have the time to put into production,” says Julie Shapiro, artistic director of Third Coast International Audio Festival, a nonprofit that promotes independently produced audio documentary work, “because you can have your listeners’ attention focused so carefully on every nuance.”

In that situation, the value of a media product does not come from being fast. It comes from being timeless. Abumrad made this point, indirectly, in that first long conversation I had with him and Krulwich. It wouldn’t make sense, he said, to devote the effort to seduce, disturb and engage the listener if “Radiolab” episodes were merely broadcast once and disappeared.

“But until 10 years ago, it always disappeared,” Krulwich pointed out. A show was something that was created, was hopefully heard by a huge number of people and then vanished.

“Why would you spend this kind of time on it?” Abumrad asked.

Krulwich answered, “Nobody would.”

This essay was originally published in The New York Times Magazine, April 7, 2011.