How to Write Suspense: What Theatre Horror's Jaws Effect Teaches Student Writers
If you are learning how to write suspense without showing the threat directly, signal it only through disturbances and let the reader's imagination supply the monster. Chris Fisher, illusions designer for the Paranormal Activity stage show, told the Guardian it is 'the Jaws effect': scarier to see the fin and hear the music than the mechanical shark's head. Sam Beech's 1am car scare, where a sensor kept triggering while every look found nothing, later fed his show Blackout. The four stages below turn that principle into a repeatable drafting method you can apply to any suspense scene in your own fiction.
Stage 1: Establish Presence Through Disturbance Alone
Theatre horror builds dread by refusing to show the threat. Chris Fisher, the illusions designer behind the Paranormal Activity stage show, told the Guardian it is 'the Jaws effect': scarier to see the fin and hear the music than the mechanical head of the shark. Sam Beech, writer and director of Blackout, described the same logic from inside the fear. Locking up a wedding venue at 1am, his car sensor kept triggering while something moved from the front bumper round the back, snuffling, and every time he looked there was nothing. The presence was real to him entirely through its disturbances: sound, motion, a sensor, and an absence that answered each check. That is the transferable observation for fiction: a threat can be established on the page purely through the changes it leaves behind.
Our suggested exercise is the disturbance ledger. Pick a scene where your character senses something nearby but never sees it, and write a 200-word ledger listing five observable changes the presence leaves: a triggered sensor, a sound shifting from one side to another, an object slightly displaced. Every entry must be perceivable without identifying the source. Then draft the scene using only ledger items, keeping the threat entirely off the page: no name, no glimpse, no explanation. Evaluate the result by rereading each line with two questions: could the character perceive this without knowing its cause, and does the threat ever appear? If any entry names or shows the source, replace it with a subtler trace. The goal is what Beech's night teaches: the thing around his car was only ever snuffling sounds that stopped whenever he looked, and that was enough.
Stage 2: Convert Failed Checks into Rising Dread
The second stage concerns what happens after a character looks for the threat and finds nothing. In Sam Beech's account, the disturbance around his car went on for half an hour, moving from the front bumper round the back and snuffling, yet every look he made found nothing. The dread did not come from any single empty check but from the sequence: each failed verification was followed by the sound continuing, which made the next look feel more necessary and more dangerous. The pressure climbed until he stopped looking altogether, deciding that if he did not look behind him, it would not get him. The transferable observation is that a failed check is a rhythm unit, not a stall: dread rises when every empty result is answered by a disturbance that has grown along some axis.
Our suggested exercise is the three-check ladder. Write a 300-word scene built around three moments where your character checks for the threat, and let every check find nothing. Between checks, insert a disturbance that escalates along one axis: louder, closer, longer, or more personal. Keep the threat off the page throughout, and push the cost of checking upward until the character nearly abandons the search, the point Beech reached when he decided not looking was safer. To evaluate the result, reread and mark three things: every check must end empty, each disturbance must be greater than the one before it on its single axis, and the scene must close on the character's reluctance to look again rather than on a reveal. If any check finds something, or two disturbances stay level, the ladder has broken and the scene deflates.
Stage 3: Let the Reader Finish the Picture
The third stage rests on Fisher's Jaws effect: the fin and the music outscare the mechanical head of the shark, because a partial signal makes each viewer finish the picture with whatever frightens them most. Beech's night shows the same mechanism from the inside. When his sensor triggered at 1am and every look found nothing, his mind did not leave the darkness empty: he had seen Andy Nyman's Ghost Stories four times, and the gnarly end the character met on stage filled the gap for him. The threat on the page was nothing; the threat in his head was a story he knew. That is the transferable observation: an unnamed gap is not blank space but an invitation, and the reader's stock of fears will fill it faster and more precisely than any description you supply.
Our suggested exercise is the two-detail completion test. Write a 150-word passage containing exactly two mismatched sensory cues, for instance a scrape and a smell that plausibly belong to different creatures, and no portrait of the threat: no name, no shape, no reveal. The mismatch matters, because two unrelated cues force the reader to invent the connecting body, which is the completion work the Jaws effect depends on. Then give the passage to two readers separately and ask each to describe what they imagined. Evaluate the result by their answers: if their inventions differ, the gap is working and imagination is doing the frightening; if either asks what the thing looks like, or both land on the same image, the cues are over-specifying, so remove one detail or add one ambiguity and run the test again.
Stage 4: Mine a Real Uncanny Moment for Scene Shape
The fourth stage concerns where suspense material comes from. Sam Beech's account shows that a real unexplained moment already carries the structure a scene needs: a fixed time and place, one in the morning while locking up a wedding venue; a sequence of observable events, a car sensor triggering while something snuffled around the car, moving from the front bumper round the back; a duration, half an hour; failed checks, since every time he looked there was nothing; a decision point, when he stopped looking because not looking felt safer; and a lasting aftermath, his admission that he does not like driving in the dark. That experience later fed into his show Blackout, which uses the quiet, the absence, and the idea of something lurking in the shadows. The transferable observation: the gap at the centre needs preserving, not filling.
Our suggested exercise is the incident-report method. Take one real moment that unsettled you and was never explained, and write it first as a neutral report containing only what you actually observed: a fixed time and place, a sequence of events, a duration, the checks you made, and the aftermath. Then convert the report into fiction, keeping the unexplained gap at the centre and forbidding yourself from filling it with an invented monster or explicit gore, so the scene inherits the true shape of the experience rather than a decorated version of it. Evaluate the result by checking three things: every detail in the scene must trace back to the report, the gap must remain unexplained by the ending, and the close should land on aftermath rather than reveal, the way Beech's account ends on his lasting dislike of driving in the dark.