Mistakes were made, Torture was used: How the agent- and patient-less constructions affect hearers' perception of blame in English

The purpose of this dissertation is to study how language influences blame, examining how the inclusion/omission of predicate arguments influences the perception of the event being described. The first section examines how the inclusion/omission of agents/patients influences hearers' perceptions of the severity by asking participants to assign punishments to various crimes. As previous research by Fausey and Boroditsky (2010) has shown agentless constructions lessen the blame/punishment given to the ‘do-er’ of an action, the results should reveal a decrease in perceived severity of an action in an agentless sentence with a patient as compared to an agentive sentence. This study seeks to confirm these results. It also tests whether a further decrease occurs between the perceived severity of an action in an agentless sentence with a patient to an agentless sentence without. The second section examines hearers' awareness of the influence of agentless sentences in everyday speech. In this section, participants are given a short description of the effect of agentless sentences on perception of blame, then given the task from the first section again. Participants’ exposure to this information is expected to have no effect. This prediction is supported by previous research by Fausey and Boroditsky (2010), Fausey & Matlock (2011), and Matlock et al (2012); which has shown framing effects influence participants’ judgements even when participants were shown videos of the events.