When I say that I am a forensic linguist, many people react in surprise. “Does that profession really exist?” The answer is yes. And not only does it exist, but in recent years it has contributed to solving crimes, clarifying high-profile cases, and opening up new lines of work in fields as diverse as education and translation.
In my book Atrapados por la lengua (Larousse, 2020) I gathered 50 cases in which words proved decisive. An anonymous threat, a contract drafted ambiguously, a disputed exam or recording… all of them show that language leaves traces as revealing as fingerprints. And that, when analysed rigorously, the traces can become crucial evidence to reveal the facts.
Forensic linguistics, however, is not limited to the the courtroom. It can also be used in the classroom. In the recent issue of INTEREOI XIX (Escuela Oficial de Idiomas) I describe how this discipline can motivate students through learning situations based on authentic cases, such as attributing authorship to an anonymous text or conducting a mock trial. It is not just a curiosity, but a teaching tool that fosters critical reflection, connectings with students’ social context, and supporting fairer assessment procedures.
Soon this content will also be included in the Master’s Degree in Specialised Translation at UVic-UCC, for subject Analysis of Specialised Discourse and Terminology, which I have redesigned to introduce the perspective of forensic linguistics. This is a further step towards integrating the discipline into the training of future translators and language specialists.
The relationship with translation and court interpreting goes further. In my expert work I also carry out evaluations of courtroom interpreting, analysing how interpreters’ decisions—their choice of register, reformulation, turn-taking and certain pragmatic markers—can affect how credible a witness statement is perceived to be. These reflections, which I recently presented at the Colegio de Traductores Públicos de la Ciudad de Buenos Aires (CTPCBA), highlight how forensic linguistics connects directly with the professional practice of translators and interpreters.
The future presents new challenges. The rise of generative artificial intelligence has given us hybrid texts: that is, texts in which one part is written by a human and another part is generated by a machine. Think of a student who writes an essay and asks a GenAI chatbot to complete a section, or a professional who uses a generator to draft the introduction of a report and then edits and adapts it.
In the CorpIdentIA project we have studied how automatic GenAI detectors work, and the results are a cause for concern: in our recent article in Inteletica, we show that they fail frequently, especially when dealing with this type of text co-authored by humans and machines.
The consequences of these errors are not abstract. For a student, it may mean an unfair accusation of plagiarism. For a teacher, the difficulty of assessing work that is no longer entirely “their own” nor entirely “external.” And for a translator, the pressure to justify whether a commission was carried out manually, with AI assistance, or through a hard-to-define combination of both. In all these cases, the boundary between the human and the artificial is unclear, and without ethical reflection we risk making unjust decisions.
This brings me to an ethical reflection: we increasingly receive requests to “certify” whether a text was produced by GenAI. But an expert report cannot be based on what an opaque detector says. Doing so would be irresponsible and could jeopardise fundamental rights. In education, it could unfairly penalise an innocent student; in research, it could question careers without foundation; and in court, it could compromise the right to a fair defence. Forensic linguistics must rely on solid, transparent methodologies, not on technological promises that offer no guarantees.
In the end, it all comes down to a simple idea: words matter. In courtrooms, in classrooms, in translation, and in the digital world. Analysing them rigorously allows us to protect rights, resolve conflicts, and better understand the society we live in.
I discuss these issues in my latest book El poder de les paraules (Ara Llibres, 2025). And also recently at La Setmana del Llibre en Català, as a guest on the Paraules habitables. Igualtat en català podcast (in collaboration with TERMCAT), where we discussed, among other things, terms such as consentiment, llum de gas and bretxa de gènere, and how their use shapes social and legal perceptions. My collaboration with TERMCAT on terminology relating to sexual consent further reinforces this idea: to name well is to make visible and to guarantee rights.
Forensic linguistics reminds us that language, far from being a mere tool, is one of the most significant traces we leave behind.
