Michael Israel

Minimizers, Maximizers and the Rhetoric of Scalar Reasoning

This paper examines the lexicalization patterns of polarity items with a view to understanding the range of possible polarity items and the reasons why such forms should exist in the first place. My starting point is the Scalar Model of Polarity (Israel 1996, 1998), which predicts a reliable correlation between a polarity items sensitivity and its scalar semantic properties: specifically, it predicts that forms denoting a minimal scalar degree may be emphatic negative polarity items (NPIs), while forms denoting maximal degrees can be emphatic positive polarity items. A variety of anomalous polarity items are discussed which flout this prediction, including both emphatic NPIs denoting maximal degrees (e.g. for all the tea in China, wild horses) and emphatic PPIs denoting minimal degrees (e.g. for a pittance, in a jiffy). The exceptional behaviour of these forms is shown to be a direct function of the participant roles they denote, reflecting the fact that different roles license different kinds of scalar inferences depending on how they contribute to the likelihood of an expressed proposition. In addition to establishing a link between thematic structure and the lexical semantics of polarity sensitivity, this result is shown to have important implications for the nature of scalar reasoning generally, and for the role it plays in structuring rhetorical discourse.