Discourse context cannot make ‘or’ inclusive (only experimental task can)

Abstract

The canonical view in linguistics is that the core meaning of ‘p or q’ is inclusive disjunction, but that its most common interpretation is exclusive disjunction. Nevertheless, the idea persists that an inclusive interpretation is possible given the proper discourse context. Experimental evidence for this idea comes from studies using truth-value judgment, showing that some participants judge ‘p or q’ as true when both p and q are true. We present evidence against the idea that discourse context alone licenses an inclusive interpretation, proposing that it is only the task of truth-value judgment that ostensibly licenses this interpretation. We conducted an experiment using both truth-value judgment and a reply-anticipation task, where participants anticipated an interlocutor's answer to how many of p and q are true, based on their interpretation of ‘p or q’. Target utterances appeared in discourse contexts encouraging an inclusive interpretation by making it maximally informative and relevant. Despite this, most participants anticipated that the interlocutor would answer that exactly one of p and q were true. We therefore believe the onus is now on those claiming that discourse context can license an inclusive interpretation to explain why it is conspicuously absent in tasks other than truth-value judgment.