Case Study 1: Looking You in the Eye — Errol Morris, the Interrotron, and The Fog of War

An interview subject who looks just past the lens is having a conversation you get to watch. A subject who looks into the lens is having it with you. Errol Morris built a machine so the same person could do both at once — and then aimed it at a man who had helped run a war. — on the method behind Morris's Interrotron

This is an analysis of a real filmmaker's well-documented technique, rendered in words as attributed Described Shots — an homage, never a reproduction. Every shot box below is labeled [after The Fog of War, 2003] and describes the film at the level of its widely-discussed craft and structure. Watch the real film with this analysis beside you; it is one of the most instructive interview-driven documentaries ever made. This is also this chapter's "Watch This" shelf entry — the deliberate exception to the off-axis rule §19.1 taught you as the default.

Why this technique matters

In §19.1 you learned the default eyeline for a nonfiction interview: off-axis — the subject looks at an interviewer just beside the lens, so the viewer becomes an observer of a real conversation. It is the right choice the overwhelming majority of the time. This case study is about a filmmaker who looked at that default, understood exactly why it works, and then broke it on purpose — and in doing so produced some of the most unsettling, intimate, and revealing interviews in the history of documentary.

Errol Morris is worth studying here for three reasons, and each maps to something this chapter argued:

  • He proves that the eyeline is a decision with consequences, not a convention to obey. By moving the subject's gaze from just off the lens to dead into it, Morris changes the entire relationship between the person on screen and the person watching. Understanding why his choice works teaches you the default more deeply than any amount of following it.
  • He is the patron saint of "the interview is the film." Morris's films are, structurally, people in chairs talking — lit, framed, scored, and cut with obsessive care, but at bottom, interviews. If a filmmaker can build Academy Award-winning cinema out of a person answering questions, then the excuse that "an interview isn't visual enough" collapses. The interview, done at this level, is the cinema.
  • He is a master of the thing §19.4 and §19.5 are about: getting people to reveal themselves. Morris's genius is not only the machine. It is a method of listening, silence, and patience that gets powerful, guarded people to say things they arguably should not. The technology is famous; the interviewing is the real lesson.

The Fog of War (2003) — subtitled Eleven Lessons from the Life of Robert S. McNamara — is our primary text, because it is Morris's technique at its purest and highest stakes: a single subject, one chair, direct into the lens, for roughly two hours. It won the Academy Award for Best Documentary Feature. Let us take apart how a movie that is, on paper, "an old man talking" became unforgettable.

Context: the maker, the machine, and the subject

The maker. Errol Morris began making documentaries in the late 1970s (Gates of Heaven, 1978) and became widely known for The Thin Blue Line (1988), an investigation into a Texas murder conviction so rigorous that it is credited with helping to free the wrongly-convicted man — a documentary that literally changed reality, built substantially on interviews. Across his career (A Brief History of Time, Fast, Cheap & Out of Control, Standard Operating Procedure, The Unknown Known, and more), Morris developed an instantly recognizable house style: stylized, richly lit interviews, often scored by Philip Glass, in which subjects speak directly to camera and are given enough rope, enough silence, and enough time to reveal far more than they intend. He is, in the most literal sense, a professional interviewer who happens to make films.

The machine. Early in his career Morris confronted the exact tension §19.1 describes. The off-axis eyeline feels honest and observational, but it also keeps the subject at a slight remove — they are talking to someone else, and the viewer watches from the side. Morris wanted the opposite: he wanted the subject to look the audience in the eye. But if you simply put the camera where the interviewer sits, the subject looks into the lens and loses the interviewer — there's no one to actually talk to, and the human connection that produces good answers evaporates. People need a face to speak to.

His solution, developed with his team and nicknamed the Interrotron (the name is usually credited to his wife, Julia Sheehan, blending "interview" and "terror"), is a beautifully simple piece of misdirection built from teleprompter parts. A teleprompter works by reflecting a screen off a two-way mirror mounted in front of the lens, so a presenter can read text while appearing to look straight at the camera. Morris's insight was to replace the scrolling text with a live video image of his own face. Now the subject sits in front of a camera whose lens is hidden behind a two-way mirror — and on that mirror they see Morris, talking to them, in real time. They look at his face; his face happens to be superimposed exactly over the lens; so they are looking directly down the barrel of the camera and into the eyes of a real person they are in conversation with, at the same instant. A matching rig lets Morris see them. Two people have a genuine, eye-to-eye conversation, and the camera records each of them apparently staring straight into the audience's soul.

That is the whole trick, and it dissolves the compromise §19.1 forces most of us to accept. You normally choose: off-axis (a real person to talk to, but the viewer is an observer) or into-the-lens (direct address to the viewer, but no one real to talk to). The Interrotron refuses to choose. It is worth understanding deeply, because even without the device, it teaches you what the eyeline is for: connection with a real listener on one hand, connection with the audience on the other. Every interview you shoot is trading between those two, and Morris found a way to have both.

The subject. Robert S. McNamara was U.S. Secretary of Defense under Presidents Kennedy and Johnson, a principal architect and manager of American involvement in the Vietnam War, a figure in the Cuban Missile Crisis, and — before all of that — an analyst involved in the WWII firebombing campaigns against Japan. He was, in other words, a man near the center of some of the twentieth century's most consequential and controversial decisions about mass death. In The Fog of War, then in his eighties, he sits in Morris's chair and looks straight at us for two hours, structured around eleven "lessons" drawn from his life — among them, in the film's widely-cited framing, ideas like empathize with your enemy and rationality will not save us. The subject is not a craftsman or a neighbor. It is a man reckoning, or partly refusing to reckon, with historic responsibility — and the interview technique is what makes the reckoning visible.

The technique, beat by beat

Below are representative beats of Morris's method, rendered as Described Shots. The real film has far more; these carry the craft this chapter is about. Read them as a demonstration of what the direct-to-lens interview does that the off-axis one cannot.

FIGURE CS1.1 — "The direct gaze"        [after The Fog of War, 2003]
  THE FRAME    A tight medium close-up of an elderly man against a dark, controlled background. He is roughly
               centered — not on a third — and he looks straight into the lens. There is nowhere for the
               viewer's eye to hide; his eyes are locked onto ours.
  THE MOVE     Essentially locked off. Stillness forces all attention onto the face and the words; any camera
               move would break the spell of the direct address.
  THE LIGHT    Sculpted, controlled, a touch theatrical — soft key shaping the face, a darker surround, the
               background dropped well below the subject. This is not run-and-gun; it is lit like a portrait.
  THE SOUND    His voice, close and clear; underneath, a spare, cyclical Philip Glass score that lends dread
               and gravity without ever telling us what to think in words.
  THE CUT      Long holds. Morris lets answers run, lets pauses sit, and cuts to archival footage and
               reenactment over the voice — but returns, again and again, to this face looking at us.
  THE EFFECT   Because he looks *into* us, we cannot be neutral observers. We are implicated, addressed,
               almost interrogated in reverse — a witness he is speaking to directly.
  THE LESSON   The into-the-lens eyeline collapses the distance between subject and viewer. Use it when you
               *want* the audience implicated and addressed — and know it is a different contract than off-axis.

Compare that, deliberately, to the default you built in §19.1 (FIGURE 19.3). The off-axis frame makes you feel you are overhearing something true. Morris's direct frame makes you feel you are being told something, personally, and cannot look away. Neither is better; they are different tools for different jobs. Morris chose the one that fit his subject — a man addressing history, and us — and the choice is the meaning.

FIGURE CS1.2 — "The silence he doesn't fill"        [after The Fog of War, 2003]
  THE FRAME    The same tight, direct frame. The subject finishes a careful, lawyerly answer — and then the
               shot holds on him, not cutting away, as the question does not immediately come.
  THE MOVE     Static. The refusal to move or cut is itself the technique.
  THE LIGHT    Unchanged — but as he sits in the held silence, the modeling seems to deepen; the face works.
  THE SOUND    A beat of near-silence (just room tone and a breath of Glass). Then, unprompted, he adds
               something — a qualification, an admission, a flicker of doubt the prepared answer had sealed over.
  THE CUT      Morris keeps the *addition*, not just the answer. The held silence is where the film's truest
               material is manufactured.
  THE EFFECT   We watch a guarded, brilliant man reveal a crack he did not mean to reveal, precisely because
               no one rushed to fill the gap. The silence did the interviewing.
  THE LESSON   This is §19.5, at the highest level: don't fill the silence. The subject will — and the thing
               after the pause is the thing you came for. The interviewer's hardest, most productive act is to say nothing.

This second beat is the one to tattoo on your brain, because it is available to you with no machine at all. Morris's most-copied skill is not the Interrotron — it is the patience. He asks, he listens, and then he waits, and the waiting is unbearable, and the subject fills it, and the film lives in what they say when they crack. You practiced this exact move in Exercise 16. Morris does it to a former Secretary of Defense, and it works, because it is a fact about human beings, not about equipment.

FIGURE CS1.3 — "Voice over history"        [after The Fog of War, 2003]
  THE FRAME    Cuts away from the face to archival material — grainy war footage, maps, falling ordnance,
               numbers rendered as motion graphics — while the interview voice continues underneath.
  THE MOVE     The archival and graphic material moves; the interview stays the anchor we return to.
  THE LIGHT    N/A on the archival; the contrast between the sculpted interview and the raw historical
               footage is itself expressive — the polished present remembering the brutal past.
  THE SOUND    The interview audio (an L-cut, in the grammar of §19.6 and Ch.28) carries across the pictures;
               Glass's score binds interview and archive into one continuous emotional line.
  THE CUT      The film cuts *away* from the talking head constantly — to illustrate, to implicate, to let
               the words land over evidence — and back to the face for the moments that must be watched.
  THE EFFECT   The interview stops being "a man talking" and becomes the narration of a visual argument. We
               hear the reflection and see the consequence at the same time.
  THE LESSON   Even the most compelling talking head needs somewhere to cut. The voice-over-archive L-cut is
               how an interview becomes a film — exactly the B-roll logic Chapter 20 is about to give you.

What Morris decided — and risked

Every strong technique is a set of brave, arguable decisions. Naming Morris's teaches you to make your own.

He decided to break the off-axis default — and risked feeling gimmicky. Direct-to-lens interviewing was, and still is, unusual for a reason: done carelessly it feels like a hostage statement or a spokesperson ad, the wrong register for observed truth. Morris bet that with the right subject, the right lighting, and the right patience, the direct gaze would feel not like address but like confession — and that the discomfort of being looked at was exactly the feeling his material about war and responsibility demanded. The risk was real: aimed at the wrong subject, the same technique would be unbearable in the bad way. He motivated the choice (Theme 4) — the eyeline serves the story of a man addressing history — and that is why it lands.

He decided the interview was the movie — and risked "boring." A conventional producer might have said a two-hour film of one elderly man talking is not cinema. Morris's whole career is the counter-argument: that a face, lit and framed and listened to with total attention, and cut against history, is as gripping as any action. He invested everything in the quality of the interview — the light, the score, the silence, the structure — rather than fleeing from it. The lesson for you, with your one subject and your Project 2: the answer to "is an interview visual enough?" is not to abandon the interview but to shoot it better.

He decided to let a monstrous-and-human subject speak for himself — and risked being accused of platforming him. Morris does not narrate over McNamara telling him he's wrong; he lets the man talk, at length, and trusts the audience — and the silences, and the archival juxtapositions — to do the moral work. This is the documentary-ethics tightrope (Chapter 21): giving a controversial subject room to reveal themselves, without editorializing the reveal for them and without laundering them either. It is the same tension §19.6 raised about the honest use of reaction shots, scaled up to a Secretary of Defense. Reasonable people argue about whether Morris got the balance right — which is itself proof of how much the interviewer's choices matter.

He decided the machine should serve connection, not replace it. It would be easy to read the Interrotron as a cold gadget. In fact it exists to increase human connection — to let the subject talk to a real face while addressing the audience. The technology is in service of the oldest thing in interviewing: one person, genuinely listening, making another person feel safe enough to be honest. Never let the lesson of this case study become "buy a machine." The lesson is "the eyeline is a relationship, and you are responsible for engineering it."

The craft moves, named

Strip away the fame and the stakes, and Morris's method is a small set of nameable techniques you can study and steal — most of them the exact skills of this chapter.

Eyeline as authorship. Morris treats the direction of the subject's gaze as a primary creative decision, on the level of framing or score. Off-axis for observation; into-the-lens for address and implication. You now own both settings on that dial (§19.1). The move is to choose consciously, every time, based on the contract you want with the viewer.

Silence as an instrument. The held pause (FIGURE CS1.2) is Morris's most transferable tool and the heart of §19.5. He asks, listens, and waits past the point of comfort, and the subject reveals themselves in the gap. This requires no budget and enormous discipline. It is the single most valuable thing in this case study.

The interview as visual argument. By cutting the voice over archival and reenactment (FIGURE CS1.3, the L-cut of §19.6), Morris turns a talking head into narration for a constructed argument. The interview supplies the thread; the images supply the evidence. This is precisely why Chapter 20's B-roll matters — an interview without pictures to cut to is trapped.

Control the room, control the reveal. Morris's interviews are lit like portraits, scored, and staged with total control — the opposite of run-and-gun. That control is not vanity; a subject in a calm, considered, well-lit space, talking to a patient listener, gives more than a subject perched in chaos. The setup (§19.1–19.2) is not decoration; it is what makes the honesty possible.

Structure the harvest. The Fog of War is organized into eleven lessons — a spine imposed in the edit on hours of talk. Morris interviews expansively and then finds the structure afterward (the paper edit of Chapter 30). The interviewer's job on the day is to harvest enough, and enough kinds, of self-contained material that a structure can be built from it later.

Trust the face. Ultimately Morris trusts that a human face, looked at long enough, tells the truth the words are dodging. He holds on it. He returns to it. He believes the audience will read the flicker, the swallow, the glance away. That trust — in the subject's face and the viewer's intelligence — is the deepest craft decision of all, and it is free.

Keep this list. When you plan your Project 2 interview, these six moves — eyeline as authorship, silence as instrument, interview-as-argument, control the room, structure the harvest, trust the face — are the questions Morris would ask.

The moment that teaches the most

There is a recurring quality in The Fog of War that first-time viewers remember: the sense of watching an extraordinarily intelligent, guarded man almost — but not quite — break through his own defenses, live, because the interviewer will not rescue him from the silence. It is worth naming exactly why it lands, because the technique is entirely learnable and requires nothing Morris had that you don't:

  • The direct eyeline removes the subject's hiding place. Looking off-axis, a subject can address a sympathetic interviewer. Looking into the lens, they must address everyone — the whole audience, history, themselves. There is nowhere to look but at us, and the discomfort produces candor.
  • The silence transfers the pressure. When Morris doesn't fill the gap, the social burden to speak falls back on the subject, and a lifetime of media training is no match for three seconds of a patient, waiting face. The prepared answer ends; the real one begins.
  • The control makes it safe enough to be honest. Paradoxically, the theatrical, careful setup relaxes the subject. A calm, considered room signals "you are being taken seriously," and people who feel taken seriously reveal more.
  • The cutting trusts us. Morris holds the shot when a lesser edit would cut away from the discomfort. The refusal to cut is the refusal to let us — or the subject — off the hook.

For your own work the lesson is exact and usable, and it scales all the way down to a friend in a kitchen chair: engineer the eyeline for the contract you want, then listen, then wait. The most powerful interview moments are almost never the answer to the question. They are what the subject says next, into a silence you were brave enough to leave open.

Why it holds up

A test of great craft is whether it survives study — whether knowing the trick makes it more impressive, not less. Morris's technique passes. Once you know the Interrotron is a teleprompter with his face on it, the direct gaze becomes more remarkable, because you see the deliberate engineering behind an effect that feels like raw honesty. Once you know the silences are a method, the reveals feel more earned, because you understand the discipline it took to leave them open. This is the mark of technique built on a true fact about human beings rather than on a gimmick: it deepens under examination.

And there is a democratizing lesson buried in the fame. Yes, Morris had a custom machine, a Philip Glass score, and a historic subject. But the transferable power — the conscious eyeline, the patient silence, the trust in the face, the interview-as-argument — costs nothing and works on anyone. Trade the Interrotron for a phone with a note taped beside the lens, the Glass score for a quiet room, and McNamara for your own subject, keep the method, and you would still get answers most interviewers never reach. That is the promise of this chapter, proved by one of its most rigorous practitioners.

Discussion questions

  1. Morris breaks §19.1's off-axis default and shoots direct-to-lens. Make the case for why that eyeline fits his subject and material — and name a Project 2 subject for whom it would be exactly the wrong choice.
  2. The Interrotron exists to give the subject a real face to talk to while they address the camera. Without the machine, what are two low-tech ways you could give a subject "somewhere to look" for a direct-to-lens interview, and what does each cost you?
  3. Morris's most copied skill is silence, not technology. Why is leaving a three-second pause so hard to actually do, and what specifically does it produce?
  4. The Fog of War lets a controversial subject speak at length without narrated rebuttal. When is giving a subject that much room a strength, and when does it risk "laundering" them? How would you decide for your own piece?
  5. The film cuts constantly from the face to archival footage over the interview audio. Why can even a riveting talking head not sustain a whole film alone — and how does this preview Chapter 20?
  6. Morris interviews expansively and finds his eleven-lesson structure in the edit. What are the risks of the opposite approach — going in with a rigid script and only asking for the soundbites you already planned?

Your turn: the direct-address interview

You now try Morris's core move — scaled to something you can shoot this week — precisely so you understand the default by breaking it.

The brief: shoot a two-minute direct-to-lens interview with one willing subject, and get one genuinely revealing answer using silence.

Constraints and guidance:

  • Give them somewhere to look. Rig a phone showing your face (a video call), or simply tape a small photo or your own face right beside/over the lens, and ask them to talk to that. Get the eyeline as close to the lens as you can.
  • Light it like a portrait, not a snapshot. One soft key on a controlled background, darker behind them (§19.2). Calm, considered, "you are being taken seriously."
  • Ask one real, open question — then wait. After their first answer, hold a full three seconds of silence, eyes on them, and record what comes next. Do not rescue them.
  • Compare to off-axis. Shoot the same question off-axis (Exercise 10's setup) and watch both. What does the direct gaze add? What does it cost? When would you actually choose it?
  • Be ethical about it. The direct-address technique is powerful at the viewer; use it honestly, get consent, and never use silence or the gaze to manufacture a "reveal" that misrepresents your subject (Chapter 21; releases in Chapter 38).

Do not worry about matching Morris's polish; worry about the method. If your subject says something after the silence that they didn't say in the answer, you have done the one thing this case study is really teaching — and it had nothing to do with a machine.

Key takeaways

  • The eyeline is an authored decision, not a convention. Off-axis makes the viewer an observer; into-the-lens makes them the addressee. Morris chose the second, on purpose, for a subject addressing history — and the choice is the meaning.
  • The Interrotron dissolves the eyeline trade-off — it lets a subject talk to a real face while looking down the lens — but its real lesson is that the eyeline is a relationship you are responsible for engineering, machine or no machine.
  • Silence is the transferable superpower. Morris's most-copied skill is patience: ask, listen, and wait past comfort, and the subject reveals what the prepared answer hid. This costs nothing and works on anyone (§19.5).
  • A great interview is still not a film by itself. Morris cuts the voice over archive and reenactment (the L-cut) — the interview supplies the thread, the pictures supply the evidence. This is why Chapter 20's B-roll exists.
  • Control the room to earn the honesty. A calm, portrait-lit, considered setup relaxes a subject into candor; the setup (§19.1–19.2) is not decoration, it is what makes the reveal possible.
  • Interviewing a powerful subject is an ethical act. Giving someone room to reveal themselves, without editorializing for them and without laundering them, is a real and contested responsibility — proof that the interviewer's choices carry moral weight.
  • Every transferable lesson here — conscious eyeline, patient silence, trust in the face, interview-as-argument — is available to you with a phone and a quiet room. The machine is famous; the method is what you steal.