Category III: Identity Under Pressure · Pilot 12 of 15
A witness identifies a suspect with absolute certainty. The security footage shows that she was alone when she made the identification.
Mina, a court interpreter, is assigned to a witness interview after a violent incident in a transit station. The witness, Eli, insists another person stood beside him and told him what to say. The footage shows Eli alone. Mina notices that Eli’s account shifts in rhythm and vocabulary, as though he is quoting someone else. She must determine whether he is lying, experiencing a trauma response, or accurately describing a person the cameras cannot record. The answer is not a supernatural diagnosis. The story tests the difference between faithful transcription and deciding what a person means.
Mina interprets Eli’s account. He names a suspect and says a second witness coached him. The interviewer dismisses the detail as confusion, but Mina records it accurately.
Security footage shows Eli alone in the station corridor. Mina watches the clip several times and notices Eli turns his head toward empty space at the same point in each version of his account.
During a second interview, Eli’s word choice and cadence change when he describes the unseen witness. Mina asks him to repeat the statement in his own words. He cannot.
Mina asks the interviewer to pose questions in a different order. Eli gives consistent answers to factual details but changes the identity of the person who coached him. The inconsistency is not random.
Mina obtains the raw audio. A faint second voice can be heard beneath the room tone, but it may be a radio bleed from the adjacent interview room. The evidence supports neither certainty nor dismissal.
Mina refuses to translate Eli’s uncertainty into a definitive identification. The report preserves what he said and what the footage shows. Outside, Eli thanks her for not making him sound more certain than he was.
The case remains unresolved, but the record is honest. Mina’s professional integrity is the emotional resolution.
Stable speaker identity; nuanced vocal shifts without arbitrary voice swapping; accurate dialogue attribution; evidence-first continuity; no visual confirmation that destroys ambiguity.