“The facts speak for themselves”: dismantling conspiracy theories as disinformation

Abstract Disinformation has been commonly approached as fake news, i.e., news that does not comply with the principles of factuality, objectivity, and neutrality. However, not all pieces of disinformation are damaging (e.g., satire) or rely on lack of factuality. Rather, it is the combination of lack of factuality and intention to deceive that embodies the most serious form of disinformation. By taking advantage of echo chambers and filter bubbles, disinformers use distorted facts to disseminate alternative forms of disinformation and manipulate readers’ views. This article discusses the relevance of establishing a sociolect of conspiracy theories (CTs) as alternative sources of disinformation. It builds on a small corpus of CTs published in Portuguese to explore the use of forensic linguistics methods to assist the detection of disinformation. A holistic linguistics approach is adopted, which operates by scrutinizing metadata, structure, and discourse to identify which linguistic features depart from mainstream sources and which ones overlap, and thus understanding the linguistic materializations of CTs. The findings reveal promising results in detecting CTs as alternative sources of disinformation, with the additional advantage of substantiating judgements of disinformation with linguistic evidence. This article concludes with a discussion of the limitations of this exploratory research.