# 16. Performance, Eyeline, and Voice The word “angry” does not create a performance on its own. It may produce furrowed brows and dramatic lighting, but it does not explain **why** a person stops, where they look, or what changes after another person speaks. The viewer reads behavior. For example: a finger stops on a line, an answer is delayed, the gaze goes to the door and only then returns to the partner. In this lesson, you will build that behavior for one shot: cause, action, reaction, voice, and an exit state. ```learner-callout {"stable_id":"c170-key-behavior-not-label","role":"key-point","title":"Start with what can be observed","body":"An emotion label can help the author think, but a shot needs visible and audible evidence: action, eyeline, pause, reaction, voice, and a change of state."} ``` A performance record does not promise a first-try result. It makes the intention testable: after you watch the frame or clip, you can see what worked and what needs to be simplified or separated. ## Minimum Input Start with one shot record, not a long character description. Complete it before you describe the behavior. ```template Line or signal that triggers the reaction: Shot size: Current eyeline direction: Entry state: Allowed action: Exit state: Exact line, if any: Pace and volume: Voice timbre or range: low, medium, or high: Delivery and forbidden improvisation: ``` Here is how the form works in a simple scene: ```text Signal: the man says, “I wasn't there.” Shot size: medium close-up of the woman. Eyeline: down, on a line in the document. Entry: her finger moves along the line; shoulders are calm. Allowed action: finger stops; short pause; one slow blink; gaze rises. Exit: gaze holds on the man; body remains calm. Exact line: “Then why is your signature here?” Pace and volume: slow and quiet, with no rise in volume. Timbre: medium, slightly lower than usual at the end of the question. Delivery: restrained; do not add words or nervous laughter. ``` If the trigger, eyeline, or entry state is unknown, do not fill the gap with the word “emotion.” Return to the script, scene geography, or the previous shot’s exit and recover one missing field. ## 16.1. Intention First, Muscles Second As the author, you can use this short chain: ```text what the character wants -> what blocks them -> how they try to get it -> what the viewer sees and hears -> what has changed by the end of the shot ``` The first three points are working thought, not mandatory shot text. In the shot description, keep the consequence a viewer can notice. ```learner-callout {"stable_id":"c170-technique-intention-to-action","role":"technique","title":"Turn intention into action","body":"Write the objective and obstacle in one line, then choose one behavior tactic and record only its observable signs in the current shot."} ``` For example, instead of “she suspects that he is lying,” write: “she stops turning pages and keeps one finger on a line; after his answer, she blinks slowly once and raises her gaze without moving her head.” ## 16.2. Build Living Behavior Choose only details that will actually be visible in this shot: - posture and weight shift; - pace or depth of breathing; - tension or release in the jaw; - eyeline direction and the moment it changes; - one blink when it carries meaning; - an action with a hand or prop; - a pause before the line; - reaction to another person’s words; - change in distance; - a residual state from the previous shot. Do not give every second five micro-actions. A small action matters not because it is small, but because it reveals a thought, a decision, or a change. ## 16.3. Listening Is Half the Dialogue While one character speaks, the other remains alive in the shot. For the listener, define who or what they look at, when the information reaches them, which physical reaction appears first, whether they try to hide it, and the state from which their answer begins. ```learner-callout {"stable_id":"c170-observation-listener","role":"observation","title":"The listener’s reaction","body":"A good reaction is often shorter than the line. A held breath, stopped hand, or changed eyeline can set up the next meaning more precisely than an extra sentence."} ``` Tie the reaction not to the scene’s general mood, but to a specific word, line, or visible action from the partner. ## 16.4. Breakdown: Restrained Suspicion Situation: a man says, “I wasn't there.” A woman does not believe him but does not want to reveal that directly yet. A weak record names a state: ```text the woman looks at the man suspiciously and angrily ``` A working record shows the reaction unfolding: ```text while the man speaks, the woman keeps looking at the document; after the words “wasn't there,” her finger stops on the line; she holds a short pause, blinks slowly once, and only then raises her eyes to the man; her shoulders and voice remain calm ``` ```learner-callout {"stable_id":"c170-inference-causal-performance","role":"inference","title":"Why this reads","body":"The reaction has an exact trigger, develops over time, and reaches an exit state. Calm shoulders do not cancel suspicion; they make the stopped hand and changed eyeline more meaningful."} ``` This exit also gives the next shot an anchor: the woman is already looking at the man and can begin her question without an unexplained new turn. ## 16.5. Stable Performance Profile and Current Shot For a recurring character, keep a separate profile: usual body rhythm, preferred distance, breathing, typical hand behavior, behavior under pressure, stable voice qualities, and any residual state that can pass into the next shot. ```template Character performance profile Usual body rhythm: Preferred distance: Breathing pattern: Typical hand action: Behavior under pressure: Stable voice qualities: What may pass into the next shot: ``` Do not paste the whole profile into a specific shot record. Keep only the objective, obstacle, one action, eyeline, reaction, voice, and changed state. --- ## 16.6. Voice Is Separate from Text A line answers “what is said.” A voice profile answers “how this person usually sounds.” For a recurring character, record timbre and range, pace, a relevant accent, ordinary delivery, a voice shift under pressure, forbidden improvisations, and pronunciation of names or special words. ```learner-callout {"stable_id":"c170-caution-voice-boundary","role":"caution","title":"Text and voice are separate decisions","body":"In the current shot, keep the exact line, required delivery, and visible action. A silent character does not acquire an accidental line, and a stable voice profile does not replace specific dialogue."} ``` For a connected scene, preserve the voice descriptor and, when timing needs it, a pause, sound tail, or ambience. This does not promise that the chosen route produces speech: if it does not, prepare and review the voice separately for the edit. ## 16.7. Cause, Line, Reaction A dialogue needs a short causal line: ```text speaker’s exact line -> visible action during the words -> listener’s reaction -> state that begins the next shot ``` Between shots, retain the residue that is genuinely needed: posture, eyeline, breathing, an object in hand, unfinished movement, or a sound tail. Do not carry details by inertia; carry only what lets the scene continue without a jump. ## 16.8. When to Simplify Simplify the moment when several internal changes must happen almost at once, an important reaction is unreadable at the selected shot size, gaze, hands, and dialogue demand different focus, the camera makes a complex move over complex performance, or the speaker and listener roles become mixed. Choose one repair: put the reaction in another shot, move closer, remove one movement, or move part of the meaning into the edit. ```learner-callout {"stable_id":"c170-practice-simplify-one-variable","role":"practice","title":"One clear repair","body":"Do not add more instructions on top of unreadable performance. Remove one equally important task and check again whether the main action, eyeline, or reaction is now distinct."} ``` ## 16.9. Practice: Line and Response without the Word “Emotion” Take a short line and the other character’s reaction. 1. Write each character’s objective and obstacle in one line. 2. Choose one action for the speaker during the line. 3. Give the listener an exact eyeline direction. 4. Add one causal reaction to a specified word or action. 5. Record the visible change by the end of the shot. 6. For the speaker, state timbre, pace, and delivery; for the silent character, forbid an accidental line. 7. Remove the words `angry`, `sad`, `afraid`, and `emotional`, then check whether the meaning remains. ### Check before Moving On ```learner-callout {"stable_id":"c170-checkpoint-card-ready","role":"checkpoint","title":"The record is ready to check","body":"Every movement has a cause; the listener answers an exact signal; voice does not conflict with body and breath; the line remains exact; the moment has no several equal changes; and the exit state can pass to the next shot."} ``` If action, eyeline, and voice conflict, remove one equally important task, preserve the exact trigger, and build the record again. The record is prepared when behavior is causal and physically feasible. Execution is confirmed separately: the actual frame or clip is watched, the voice is heard, and one controlled retry shows what changed. The lesson result is one performance record for a line and one reaction record for the response. You can apply them to a shot record without rewriting the entire character profile every time.