Hypno-questioning technique: A questioning strategy to uncover witness honesty in Indonesian courtroom discourse
Description
This dataset contains analytically reconstructed courtroom interaction excerpts illustrating hypno-questioning techniques employed by attorneys when examining suspect witnesses and fact witnesses in Indonesian criminal trials. The dataset is derived from systematic qualitative analysis grounded in Relevance Theory and hypnotic language patterns, focusing on how questioning strategies guide, constrain, and shape witness responses during courtroom interaction. The data comprise reconstructed excerpts from criminal cases adjudicated in three Indonesian district courts: Pengadilan Negeri Bantul (PN-BTL), Pengadilan Negeri Yogyakarta (PN-YK), and Pengadilan Negeri Sleman (PN-SLM). To ensure ethical compliance and confidentiality, all excerpts are analytically reconstructed rather than verbatim transcripts, while preserving interactional logic, pragmatic force, and institutional norms characteristic of Indonesian courtroom discourse. The dataset includes two witness categories: suspect witnesses and fact witnesses. For suspect witnesses, the dataset documents nine types of hypno-questioning techniques, including Presupposition, Embedded Command, Mental Imagery, Soft Suggestion, State Elicitation, Commitment Framing, Temporal Framing, Perspective Shift, and Double Bind. For fact witnesses, the same set of techniques is represented, reflecting differences in distribution and interactional function across witness roles. Each data entry consists of bilingual representations (Indonesian as the source language and English as the target language), accompanied by metadata specifying court location, case type, time stamp (minutes and seconds), and technique code (HQ1–HQ9). Case types are also presented bilingually to facilitate interpretability for international audiences. This dataset serves as the empirical foundation for the associated journal article and is intended to support transparency, replicability, and comparative discourse analysis in legal linguistics, forensic pragmatics, and courtroom interaction research.
Files
Steps to reproduce
The dataset was derived from anonymized courtroom interaction data focusing on attorney–lay witness and fact witness exchanges in Indonesian criminal trials. The data consist of question–response sequences extracted from examination phases in which attorneys addressed non-expert witnesses. Identifying information, including names, case numbers, and institutional references, was removed prior to analysis. Data selection followed a qualitative purposive sampling approach. Only interactional sequences where an attorney’s question was immediately followed by a witness response were included. The dataset retains the original labels lay witness and fact witness as presented in the source materials; analytically, both refer to non-expert witnesses providing factual, experience-based testimony rather than expert opinion. The transcripts were cleaned and standardized by harmonizing transcription conventions, turn-taking structures, and time-stamping. Original Indonesian utterances were aligned with English translations, prioritizing pragmatic equivalence, particularly in relation to modality, temporal reference, presupposition, and pragmatic force. Analysis was conducted using a qualitative discourse-analytic framework informed by Relevance Theory and hypnotic language pattern analysis. Operational definitions were established for each hypno-questioning strategy, including presupposition, embedded command, mental imagery, soft suggestion, state elicitation, commitment framing, temporal framing, perspective shift, and double bind. Attorney questions were coded at the utterance level and assigned to a dominant strategy based on primary pragmatic function. Coding was performed manually to allow fine-grained interpretation. Ambiguous cases were reviewed iteratively to ensure consistency. Coded instances were aggregated to generate frequency distributions of hypno-questioning strategies. To reproduce the analysis, researchers may apply the same coding framework to the dataset by identifying attorney–witness exchanges, coding each question using the operational definitions provided, and calculating strategy frequencies. Reproducibility is grounded in the transparency of the analytical workflow rather than in the replication of courtroom events.