The lessons I can explain follow. I have no reason to believe they apply only to those conducting weekly 55-minute-ish conversations on public radio. You might well adapt them profitably to other journalistic situations, documentary filmmaking (another interest of mine), interviews of the to-get-a-real-job variety, or even regular old casual interpersonal exchanges. In fact, the more alien the domain to which you bring them, the happier I'll be. Go on, leave a comment about it.
Naturally, I could draw the exact opposite conclusions from all these by interview number 200, but here goes anyway:
Don't write questions. If conventional journalism has taught me anything — and it hasn't — it's to lead counterintuitive, so here's your counterintuition. I, as did every super-novice interviewer except maybe Michael Silverblatt, started out by scripting elaborate sheets of questions ahead of time. This is death. Work from a question sheet, and you kill any chance of organic exchange, of real, vital intellectual back-and-forth. You just tick off the boxes and grind awkwardly forth. Which leads me right into my next point, which should probably be a subpoint of this one...
Interviews are not conversations — conversations are objectively better. As Jack Paar told Dick Cavett, "Kid, don’t do interviews. That’s clipboards, and David Frost, and what’s your pet peeve and favorite movie. Make it a conversation." I once thought of this as a dichotomy between two equal and opposite hosting strategies. In the "facilitated speech," the host's goal is simply to elicit maximally interesting and detailed responses from the guest, minimizing their own presence. In the "conversation," the host both contributes and seeks contribution, potentially even mirroring the guest's role.
I still regard those strategies as opposite, but they sure ain't equal. Perhaps this is an undiplomatic claim to make, but conversations are just better, always and absolutely, than formal, traditionally-conceived interviews. Those strike me as nothing more than hokey by-products of such journalistic rigidities as twitch time limits and miniscule word counts. Planned, borderline-rehearsed simulacra of conversations aren't conversations at all. They can't wander into the unexpected, exciting places genuine conversations do, nor can they hope to arrive at the the surprising, fascinating conclusions genuine conversations do.
We know from everyday experience that doing conversation means sometimes leading, and sometimes being led. Hence the need, even in broadcast situations, to be prepared to...
Follow your guest. Because some broadcasters don't quite grok this, I'm forced to enshrine it as a separate point. This isn't to say that you should follow your guest all the time — certainly you'll want to show the way into promising intellectual territory now and then — but if he seems interested in and enthusiastic about discussing a certain subject, fall in line. It doesn't matter if you hadn't intended to talk about it. It doesn't matter if it doesn't keep with the interview's "theme." It doesn't matter if you don't know anything about it. (After all, you can learn from the guest, then and there.) Scrap your plan and follow.
Everything is interesting, properly approached, properly thought about, properly framed, properly discussed. I don't care if your guest wants to talk about their puppy. If they're that consumed with the puppy, there's something interesting about the puppy. It's not your job to steer him away from the puppy; it's your job to find the best angle from which to ask about the puppy, to...
Follow up. I don't remember where I first heard this, but it's become an old interviewer's saw: if a guest says, "And then I killed my entire family and buried them in the back yard," your next question should not be, "So where did you go to high school?" Ideally, everything you ask should flow smoothly from what the guest just said. I realize this isn't always easy; sometimes you've just got to jump the tracks. But if the subject has to change, far better to run with something your guest happened to bring up than to throw the switch yourself.
As elementary-school as this sounds, don't forget to ask your guest for real-world examples. I myself could use some improvement here; I've been complacent about letting abstractions and generalizations slide. It can feel a tad harsh to prod someone with the kind of responses that draw out examples — "Like what?", "Such as?" — and it's especially difficult for someone like me, who hates putting people on the spot, but the concrete provides much richer conversational material.
There's one follow-up technique I like very much, which I've come to call "The Old Distill-and-Expand." When a guest finishes answering a question, I sometimes try to summarize and clarify their answer back to them, asking if I've got it right. The trick to make this more interesting is to take this summary a step further, to extrapolate from it, to make a guess about its implications and see how the guest responds. This routinely gets me pretty compelling material.
Research indirectly by immersing yourself in the subject's world. Reading the guest's book, watching their movie or listening to their album is only step one. (And you'd be astonished how many interviewers don't reach step one.) Use other books, other movies, other music and the awesome power of the internet to get into and absorb the world around the subject, the culture they call home. Check out their other interviews. Look into their influences. Research people around the guest — geographically, intellectually, spiritually — as if you were going to interview them as well.
Only ask questions you're genuinely curious to hear answered. It so comes through when an interviewer asks a question for no better reason than that they assume it "should" be asked. If you don't actually care about the answer to a question, if you're not actually excited to know how the guest will respond, then for the love of YHWH, do not ask it.
(Were I given to dopier names, I would possibly have called this point "Curiosity, curiosity, curiosity," "Keep it curious" or "It's the curiosity, stupid!")
This speaks to a much greater divide that bears on every art form, every craft — probably every pursuit. Are you creating your work because you honestly find it cool, or because you're laboring under the notion that "people" will like it? This fumbly guesswork about the taste of some abstract audience has produced humanity's stupidest, least valuable art. Ditch the "people"-pleasing and make what you like.
The ideal question is askable of no one else, by no one else. And I emphasize the word ideal; most questions won't fit these criteria. While it's theoretically a trivial matter to ask your guest a question only he can answer — get into the specifics of his book, say, or an unusual event in his life — you'll find it's still done with shocking rarity, at least in the mainstream media. Pose a question that could be asked of anyone, "How did you get this idea?" being the most dishearteningly common, and you might as well wave a red flag signaling to the guest that he's in the hands of an idiot.
The "by no one else" part is more neglected, and thus more important. Though some will no doubt challenge me, I think it separates the great interviewers from the mediocre. Nearly every time I watch my favorite interviewers in action — your Roses, your Thorns, your Silverblatts — I think, "Man, only he would have asked that!" I sometimes even think, "Man, only he could have asked that!", which is truly nxtlvl.
Upshot: if you want to interview and you don't have a personality, get one. If you want to interview and you're unwilling to expose your personality, get over yourself. You'll ultimately find you have to...
Reveal your ignorance. Hey, it beats the alternative. There's no shame in not knowing what your guest is talking about; if you don't, chances are a lot of your audience doesn't either. I rarely cringe so sourly as when I'm listening to an interview and it becomes obvious that the host is bluffing his was through something for which he neglected to prepare. I'm not about non-preparation; by all means, prepare as much as possible. But you can't know about everything in heaven and Earth.
Besides, you're in this game to develop, aren't you? Nothing impedes learning, improvement and evolution as much as pretending you already know, have fully improved and are perfectly evolved. Let's not even get into the embarrassment of realizing that you're a thousand steps up your wobbly Jenga tower of half-truths, guesses and elisions with no safe way down in sight.
Get to the human. And get past the boilerplate. The boilerplate is your enemy. If you're talking to a media veteran, someone who's been interviewed countless times before, they will have boilerplate. Behind this boilerplate is a fascinating human being, so you'll need to get around it or punch straight through it to produce so much as a reasonably interesting conversation. Your only tools are your questions.
The most effective weapon is surprise. Some interviewers interpret this as an endorsement of "gotcha" questions, but man, few things are more tiresome than "gotcha" questions. (Need I invoke the wan spectacle of Terry Gross going on the attack when presented with a guest too far outside her own ideological realm?) I mean more the sort of surprise one feels when presented with an entirely different way to think about something familiar. It's the wonderful surprise produced by framing and approach, not the deadening surprise produced by the j'accuse.
Whenever possible, interface with your guest as a human being, with all the complications that entails, rather than as a mere vehicle for ideas. The danger here, of course, is that you might lose sight of the value of ideas and turn into StoryCorps. The trick, to the extent that I've figured it out so far, is to conduct ideas-oriented conversations on a human level. And not, you know, to make Eastern Europeans weep in a bus.
You can't have a worthwhile conversation in seven minutes. If some suit insists you have to, take your skills elsewhere.