Hari names Robert Caro's iconic interview question. Caro always asks, "What would I see if I was there? What would I have seen if I was there?" And he just asks it over and over and over again. He says that there's something about that question that brings out really good answers.1 Hari uses the question literally. He uses it because it produces a specific kind of material — sensory-scene material that abstract description cannot reach.
He extends the technique. Often I'll say, "What do you see in your mind when you picture that moment?" He has a related question of his own for narratively-resistant subjects. If we were making a movie about your life, what would the key scenes be?1
The two questions do similar work. They convert the subject's abstract description into concrete scene. The conversion produces material the writer can actually use — sense-data, dialogue, image, gesture. The reader of the eventual book gets to see the moment instead of being told it happened.
These are image-elicitation questions. Most interview questions invite the subject to summarize or analyze (tell me about that period, what was that like, how did you feel). Summarize-and-analyze questions produce cooked material — material the subject has already digested into a self-narrative. Hari and Caro use questions that bypass the digestion by asking for the unprocessed sensory and spatial content of specific moments.
What would you see if you were there requires the subject to reconstruct a scene from inside it. The reconstruction produces details — the temperature of the room, the placement of furniture, who was where, what was being said, what the subject's body was doing. The details are what writing actually rests on.
If we were making a movie about your life, what would the key scenes be? requires the subject to identify the moments that carry meaning and to describe them as visual-spatial events. The frame around the question (movie-making) gives the subject permission to skip toward the operationally significant scenes rather than walking the writer through every detail of biography.
This page is operational technique within Look for a Reason to Love Them (Hari) — the stance produces the openness; the questions produce the material. The pages work together.
It connects to Texture Over Abstraction (Wang) — both are about scene-level detail as the actual carrying-medium for ideas. Wang's discipline is in what you write; Hari's question is in how you produce material to write from.
It sits adjacent to Six-Minute X-Ray (Hughes) — Hughes's elicitation methodology includes similar scene-reconstruction techniques used for forensic and intelligence purposes. The same questions produce the same material across journalistic and forensic settings. The architecture is morally indeterminate; the use determines the ethics.
Imagine an interview with someone whose father died when they were a child. The summary question — what was that like for you? — produces a summary answer: it was really hard. I struggled for a long time. It shaped a lot of who I became. The summary is honest. It's also unusable for writing. The reader can't see anything.
Now imagine the Caro question. What would I have seen if I was there at the moment you found out? The subject pauses. They reconstruct. We were in the kitchen. I was eating cereal. The phone rang. My mother answered. She didn't say anything for a long time. Then she sat down on the floor. She didn't make a sound. She just sat down. I remember thinking she was sitting in a strange place. The same event. The reader can now see the kitchen, the cereal, the mother on the floor, the child noticing the strange placement of her body.
The second material is what the book needs. The Caro question produced it. The summary question never would have. The technique is operationally specific in producing the difference.
You're interviewing someone for a story that requires scene-material. Three operational moves:
One. Have the Caro question ready. When the subject gives you a summary answer, follow with what would you have seen if you were there? The follow-up signals that you want scene-level detail. The subject will recalibrate.
Two. Have the movie question ready for life-history. When the subject is walking you through their biography in chronological summary, interrupt with if we were making a movie about that period, what would the scenes be? The subject will shift from chronology to significance.
Three. Don't interrupt the scene-reconstruction. When the subject is in the scene-reconstruction mode, they need silence and time. Don't ask clarifying questions until they've reached a natural stopping-point. The interruption forces them out of the reconstruction back into the summary-mode.
The discipline is a small set of well-deployed questions. The deployment matters more than the question-list size.
You'll know your interview produced summary-mode material when your notes are full of general statements with few specific scenes. The reader of any book you write from these notes will get analysis without texture. The summary is the diagnostic. The interview missed the level the writing needs.
You'll know the questions worked when your notes contain specific physical details you can later use to render scenes. The kitchen, the cereal, the placement of the mother's body — these are the texture. The detail is what the technique was after.
Caro's question is widely cited in the longform-journalism craft literature. The technique has substantial cross-practitioner endorsement. The principle is robust.
The tension: subjects with poor episodic-memory or significant trauma may not be able to reconstruct scenes accurately. The questions assume retrievable scene-memory. Some subjects don't have it; some shouldn't be pushed to access it. The discipline requires judgment about which subjects can produce scene-material safely.
Psychology: Episodic Memory Reconstruction — the cognitive-psychology literature on memory documents that episodic memories are reconstructed in the moment of recall, not retrieved from storage. The scene-question is leveraging this reconstruction process. The handshake reveals that the writer's question is doing the same operation as the clinician's "tell me what you remember" prompt — both produce reconstruction events that surface material the summary-mode hides.
Behavioral Mechanics: Cognitive Interview Technique — the forensic-interview literature has developed cognitive-interview techniques for eyewitness work that include scene-reconstruction prompts identical to Caro's question. The convergence across journalism and forensic work confirms the technique's operational robustness.
The Sharpest Implication The difference between summary-material and scene-material in nonfiction is largely a difference of the questions the writer asked. Most journalism asks summary questions. Most journalism produces summary material. The shift to Caro-style questions is small in technical-difficulty and large in operational-effect. Writers who don't make the shift are producing thinner work than they could be, often without knowing they're doing it.
Generative Questions