Rhetorical strategies and their influence on audiences are often studied through social media posts and comments. However, this focus overlooks the "universal audience", which is the majority of readers who remain silent and do not explicitly express how a message affects them. This study investigates how two classical modes of persuasion, ethos and pathos, resonate in the silent audience’s interpretations of meaning. Using a dataset of social media sentences paired with human-written interpretations, we label both sources for ethos and pathos and assess whether these rhetorical appeals are preserved. Our analyses show that interpretations diverge from the original sentences in 30% of cases, with rhetorically charged content eliciting greater variability than neutral content. We further find that ethos and pathos in original sentences can predict audience attitudes toward the author, underscoring the subtle ways rhetoric shapes perception beyond visible engagement.

A standard majority-based approach to text classification is challenged with an individualised approach in the Semeval-2023 Task 11. Here, disagreements are treated as a useful source of information that could be utilised in the training pipeline. The team proposal makes use of partially disaggregated data and additional information about annotators provided by the organisers to train a BERT-based model for offensive text classification. The approach extends previous studies examining the impact of using raters’ demographic features on classification performance (Hovy, 2015) or training machine learning models on disaggregated data (Davani et al., 2022). The proposed approach was ranked 11 across all 4 datasets, scoring best for cases with a large pool of annotators (6th place in the MD-Agreement dataset) utilising features based on raters’ annotation behaviour.