Getting People to Tell You the Truth in Interviews
There is a moment in interviews — not every interview, but the good ones — when something shifts. The subject stops performing and starts talking. That careful distance they maintained at the beginning dissolves without fanfare. They say something they didn't plan to say, or they say something they've said a hundred times before but this time mean it, and you can hear the difference in their voice. If you've experienced this as an interviewer, you know exactly what I'm describing. If you haven't yet, that moment is what we're chasing here.
I've been interviewing people for the better part of two decades — documentary subjects and death row inmates, Nobel laureates and neighborhood witnesses, politicians who wanted to use me and grieving parents who didn't want to talk to anyone at all. I've done interviews that ran three minutes on deadline and oral histories that stretched across years. I've watched interviews collapse in front of cameras and come alive in the back corners of diners. And if there's one thing I've learned with any real confidence, it's this: the interviewers who consistently get people to say true things are almost never the ones you'd expect. They're not the most aggressive. They're not always the most prepared. They're not usually the most famous. They are, almost without exception, the ones who have learned — really learned, through practice and failure and deliberate work on themselves — how to be genuinely present with another human being under unusual conditions. That sounds simple. It is profoundly not simple.
This course is built around one core argument: the interview is not an extraction. It's a relationship compressed into time. But that's easy to say. Let me show you what it actually means in practice before we go any further.
The Extraction Model, and Why It Fails
A few years ago I watched a journalist — talented, well-prepared, genuinely committed to his story — interview a woman whose son had been killed by police. He had his questions printed on a sheet of paper. He looked at them more than he looked at her. When she answered, he was already reading the next question, nodding at a rhythm that had nothing to do with what she was saying. He got answers. He filled his recorder. He did not get what she could have given him, which was considerably more.
What he was practicing is the extraction model, and it is the dominant mode of interviewing in almost every professional context. In the extraction model, the interview is a delivery mechanism. The subject has information inside them; your job is to get it out efficiently. You prepare your questions, you work through them, you capture the answers, you leave. The interview is a transaction, and like most transactions, it produces exactly what was agreed to in advance: the expected, the prepared for, the already-known-to-be-available.
The problem is not efficiency — it's epistemology. The extraction model assumes you already know what you're looking for. It treats the interview as a process of confirmation rather than discovery. And subjects can feel this. Not consciously, usually, but somewhere in the body, in the quality of their attention, they register that the person across from them is not actually listening — they're retrieving. And when people feel retrieved from rather than listened to, they give you retrieval-level answers: careful, edited, pre-packaged, safe.
The woman whose son was killed — she had spent two years being interviewed. She knew how it worked. She had a version of her grief that she kept ready, the version that translated into usable quotes. The journalist got that version. Afterward, when he'd left and I was still there, she told me things that would have broken his story open. She'd withheld them not out of malice but out of a well-founded sense that he wasn't there to receive them.
That's the extraction model's fundamental cost: it gets you what people have decided to give, not what they actually have. The relationship model gets you the rest.
What the Relationship Model Actually Means
Calling it a "relationship model" risks sounding like a soft, feelings-first approach — the kind of thing that makes hard-nosed journalists roll their eyes. It isn't. The relationship model is more demanding than the extraction model, not less. It requires more preparation, more presence, and more genuine intellectual engagement with your subject as a human being rather than a container of quotable material.
What it means in practice is this: you treat every decision you make — how you enter the room, how you set up your recorder, how you phrase the first question, when you go quiet — as a relational decision that your subject will register and respond to. You are not just gathering information. You are building, in real time, the conditions under which information can actually move between two people. That process is relational, it has an emotional logic, and it cannot be fully scripted in advance.
The fundamental job of the interviewer, in this model, is not to ask the right questions. It is to create conditions where truth becomes easier to say than to withhold. Those are different jobs, and they produce different results.
The Paradox of Control
Here is something that took me an embarrassingly long time to understand: the more you try to control an interview, the less you get from it.
This runs against every instinct a rigorous person brings to a professional task. You're trained, in most work contexts, to prepare thoroughly, execute your plan, and maintain command of the process. In interviewing, this impulse — if you follow it all the way — will kill your interview. Not immediately and not dramatically, but you will notice at the end that the subject never quite settled, never quite opened up, that you got the information you came for and nothing more, and that the "nothing more" was, in fact, where the real story was.
The paradox works like this. A subject in an interview is always reading you — reading your level of interest, your flexibility, your willingness to be surprised. When you're locked to your question list, they can feel the lock. When you're visibly waiting for them to finish answering so you can ask the next prepared question, they stop going deep in their answers because going deep doesn't feel like it's going anywhere. But when you're genuinely responsive — when your next question comes from what they actually said rather than what you planned to ask — they experience the interview as a real conversation, and conversations invite a different register of honesty than performances do.
Control in interviewing is not a grip. It's a current. The interviewers who consistently navigate their subjects toward the things they need — the specifics, the admissions, the unguarded moments — are the ones who hold it lightly enough that subjects don't feel held at all. We'll spend an entire section on this later. For now, the point is simply that the instinct to control is the instinct you most need to monitor in yourself.
A Taxonomy of Interview Types
Not all interviews are the same, and one of the most common mistakes even experienced interviewers make is applying the same posture to every kind of conversation. There are three fundamental modes, and each demands something different from you.
Informational interviews are what most people picture when they picture interviewing. Someone knows something you need to know — a process, a history, a technical detail — and you're there to get it. The primary challenge here is precision: getting the thing accurately rather than approximately, and knowing enough before you walk in to recognize when something's wrong or incomplete. The relationship still matters — people give more to people they trust — but the primary skill is knowledge organization and the ability to ask with specificity.
Emotional interviews are a different animal. You're talking to someone about their experience of something: a loss, a triumph, a transformation, a trauma. Here, the primary challenge is not informational but psychological. You're asking someone to revisit something that cost them something, and the interview can only go as deep as your subject's sense of safety allows. Precision still matters — you want the specific memory, not the general impression — but you get to precision through a different route, one that requires patience, genuine warmth, and a willingness to follow where the subject leads rather than where your questions point.
Accountability interviews are the ones most people find hardest. You're asking someone to answer for something — a decision, an action, a failure, a discrepancy between what they've said publicly and what the evidence shows. The subject has reason to resist you. The interview is, in some fundamental sense, adversarial, even if you conduct it with courtesy. Here, preparation is not just respectful — it's tactical. And the primary skill is holding the subject accountable without giving them something to be aggrieved about, which requires a particular kind of precise, non-theatrical persistence.
Most interviews contain elements of all three. An oral history of a public official involves information, emotion, and at least some accountability for decisions made. A profile of a CEO who's navigating a scandal is partly emotional, mostly accountability, and necessarily informational. The skill is knowing which mode you're primarily in at any given moment and shifting your posture accordingly — because the tool that opens an emotional subject will sometimes shut down an accountable one, and the directness that works in an informational context can feel like an attack in an emotional one.
Conversations and Interviews: Why the Difference Matters
Here is a thing people say about great interviews: "It felt just like a conversation." This is almost always meant as a compliment to the interviewer. But it's worth unpacking what it actually means, because conversations and interviews are not the same thing, and understanding the difference is crucial.
In a conversation, both parties have interests in roughly equal measure. You can change the subject when you're bored. You can go quiet without it being a technique. You can choose not to follow up on something uncomfortable because you don't want to create friction in a relationship that has to continue after the conversation ends. Conversations have the freedom of mutual indifference to outcome.
Interviews have none of this freedom. The interviewer has a purpose — a story to tell, a record to build, an accountability to execute. That purpose is not shared symmetrically by the subject, who may have come with entirely different intentions: to promote something, to control a narrative, to honor a memory, to survive without saying something that can be used against them. The interview is, structurally, a negotiation — not always a hostile one, often a warm one, but always one in which both parties are doing something more deliberate than simply talking.
What the best interviewers do is manage the gap between these two things. They create the conditions of a conversation — the ease, the mutuality, the sense that ideas are being genuinely explored rather than extracted — while maintaining the internal discipline of an interview: knowing what they need, tracking what they've gotten, noticing what hasn't been said. The subject gets the experience of conversation. The interviewer keeps the structure of an interview. That asymmetry is not deceptive — it's professional. And getting it right is one of the hardest things in this craft.
The Traditions This Course Draws On
Everything in this course draws on real, practitioner-developed traditions, not theory for its own sake.
Investigative journalism has developed the most rigorous approaches to adversarial and accountability interviewing — how to confront people with evidence, how to interview hostile sources, how to get something from someone who has decided to give you nothing. It is also, at its best, deeply ethical about the obligations those confrontations create.
Oral history is the oldest form of systematic interviewing and arguably the most thoughtful about the relationship between interviewer and subject. It has developed sophisticated practices around informed consent, the reliability of memory, multi-session interviewing, and what happens after the interview is over — who owns the record, who can access it, and what the interviewer owes to someone who has trusted them with their life story.
Documentary filmmaking has had to solve the problem of the camera — how you get people to stop performing for a lens and actually talk — and in solving it has developed practical knowledge about setup, environment, and how technical choices affect human openness in ways that print journalists rarely have to think about.
Audio journalism — the Terry Gross and Michael Barbaro tradition — has developed particular fluency with listening as a visible, active act, with the specific demands of a format where voice is everything and there is nowhere to hide.
These traditions have different vocabularies, different ethical frameworks, different ideas about what an interview is for. They also share more than they know, and the synthesis is, I think, more useful than any one of them alone.
Who This Course Is For
If you conduct interviews of any kind in your professional life — journalist, documentary maker, oral historian, researcher, podcaster, or someone in a field adjacent to any of these — this course is for you. You don't need to be a beginner. Some of the most entrenched bad habits I've encountered belong to people with twenty years of practice who just never examined what they were doing carefully enough to notice.
What you do need is willingness to look at your own practice honestly, including its uncomfortable corners. Some of what follows will name things you do that you'd rather not see named. That's not an accusation — it's the only way I know how to make this useful.
We're going to build a philosophy, section by section. Not a checklist, not a collection of tricks, but a unified way of thinking about what you're doing when you sit down across from another human being and ask them to tell you the truth. By the end, you'll have that philosophy, you'll have the specific skills it requires, and you'll understand — I hope — why those skills are not optional extras but the very foundation of the whole enterprise.
Let's begin.
Only visible to you
Sign in to take notes.