| Annotation in original language: |
This study examines grammatical differences between news articles published by Czech online mainstream media and disinformation (or antisystem) outlets, focusing specifically on morphological features. The analysis compares the relative frequencies of selected morphological categories in texts from both media types. The material for the study is drawn from the CLIC24 corpus, which consists of nearly 700 Czech online news articles representing both mainstream and disinformation sources. All texts address the topic of climate change, ensuring thematic consistency and minimizing differences arising from topic variation.
Methodologically, we calculate the relative frequencies of selected morphological features, including parts of speech, grammatical cases, types of pronouns, deverbative adjectives, past participles, and perfect participles. MorphoDiTa tool was used for morphological annotation (Straková et al. 2014).
The results reveal systematic differences between the two media types. Mainstream media texts show a higher frequency of nouns, numerals, and prepositions, as well as greater use of genitive and locative cases. They also contain more past participles and singular nouns. In contrast, disinformation texts exhibit higher frequencies of adjectives, adverbs, interjections, conjunctions, pronouns, and particles. They are characterized by greater use of dative, vocative, and instrumental cases, as well as possessive, indefinite, negative, and demonstrative pronouns.
Overall, the findings suggest that mainstream and disinformation media differ not only in content and framing, but also in systematic grammatical patterns.
|