Orthogonality and Presuppositions: A Bayesian Perspective

Abstract

Abstract. This work focuses on the reasons why the projection properties of presuppositional elements are not uniform. Presupposition projection designates the fact that operators that suspend the truth of a proposition have no effect on the presupposition it is associated with. For instance, with Paul did not stop smoking, the main content (that Paul does not smoke) is canceled by the negation whereas the presupposition that Paul has been smoking is left untouched. It has been observed that projection is not uniform across the different expressions that convey a presupposition (its triggers), and a distinction between weak and strong triggers has emerged. It seems that this distinction correlates with an intuition of (non-)orthogonality, that is, of (non-)independence between the main content and the presupposition. I argue that this intuition is correct to some extent but has to be clarified. To this aim, I propose a Bayesian approach, which provides a more precise rendering of orthogonality.

Experimental Perspectives on Presuppositions

“…The present results support a more precise claim in line with (Jayez 2014). There are in fact three categories of triggers.…”

“…Jayez (2014) argues that what we observe with such triggers is a side-effect of a very general discourse constraint, studied in . Non-main content information, including presupposition and implicatures is in general not involved in the normal, that is, non-metalinguistic, flow of discourse.…”

“…(Jayez 2014) discusses at some length and the remarks made there extend to (Amaral and Cummins 2014) to a large extent. The main goal of these two papers is to investigate experimentally the difference posited by between resolution and lexical presupposition triggers.…”