TraMa: An annotation scheme for detecting ideological manipulation in translation
TraMa is an annotation scheme for detecting, classifying and quantifying ideological manipulation in translation. It annotates aligned source–target segment pairs, on sentence or paragraph level, with twelve segment-level tags in four groups (content operations, ideological calibration, habitus conversion and provenance erasure). Each tag carries a directional code based on van Dijk's ideological square and a confidence score. Three text-level diagnostics (PSEUDO-ORIG, HABIT-CONFORM and TROJAN) identify translations that have been converted to the discourse of the receiving field and presented as originals.The manual gives the definition, decision criteria and authentic examples for each tag, drawn from English–Greek far-right translations, together with an annotation protocol and rules for interpreting results on three criteria: tag density, directional consistency and macro diagnostics. A machine-readable specification of the scheme (JSON) is included. The scheme was developed and applied in the author's doctoral research on the translation of US far-right discourse by Golden Dawn. Version 1.0.1 extends the alignment guidance (§4, §12) and the alignment aids for EXCIS and INSERT to length-based alignment (Gale & Church, 1993), the method used by the TraMa annotation tool. The tags, definitions, directional codes and interpretation rules are unchanged.
Authors
- Valentini Kalfadopoulou (ORCID: https://orcid.org/0000-0003-2184-5435)
Institutions
- Ionian University (GR)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-18
- DOI
- https://doi.org/10.5281/zenodo.24326540
- Primary Topic
- Translation Studies and Practices
- Type
- article
- Field-Weighted Citation Impact
- 0.00