What this project is about
Anne believes that Bob assumes that Anne believes that Bob’s assumption is false.
Does Anne believe that Bob’s assumption is false?
Don’t try too hard answering the question - any straightforward attempt will lead to paradox.
But what are we to make of sentences such as “Anne believes that Bob’s assumption is false.”
Is the sentence true or false?
On the face of it, it would seem that answering this question is a pressing problem for natural language semantics that assigns truth conditions to sentences of natural language.
However, semanticists have largely ignored problems of this kind, leaving the field to philosophical logicians working on paradoxes, in particular, the paradoxes of truth such as the Liar paradox.
But research on the paradoxes of truth has often focused on exploring the space of possible coherent “solutions” to the paradoxes thereby ignoring desiderata of natural language semantics.
The project provides a unified perspective on natural language semantics, conceived of as truth-conditional semantics, and the research on the so-called semantic paradoxes in form of theories of self-applicable truth.
A unified approach to truth and semantics will need to answer two principal challenges, which divides the research project into two interrelated parts:
The project constitutes the first systematic study of truth and natural language semantics from such a combined perspective.
Based on work in natural language, I argue that ‘true’ is not a genuinely gradable expression. I also provide an explanation of the apparent evidence for gradability. Hence there is no reason to think that there is a truth property that comes in degrees.
There has been increasing recognition of the appeal of contextualist solutions to the Liar paradox. The article investigates two prominent contextualist accounts as to their potential to adequately motivate contextualism.
Is it possible to provide a semantic theory for first-order languages in which the quantifiers are absolutely unrestricted? It has been argued that such a semantics can and indeed must be given in a plural or higher-order metalanguage. I argue that it is possible to provide such a semantic theory in a first-order metalanguage as well.
Minimalism about truth is one of the main contenders for our best theory of truth, but minimalists face the charge of being unable to properly state their theory. This paper shows how to properly state the theory by appealing to propositional functions that are given by definite descriptions
I argue that a full understanding of how relativism is conceived within theories of natural language shows that neither of the purported connections can be maintained. There is no reason why a solution to the Liar paradox needs to accept relativism.
The aim of this volume is to open up new perspectives and to raise new research questions about a unified approach to truth, modalities, and propositional attitudes.
We develop a semantics for truth and belief that combines ideas from contextualist theories of attitude reports and Awareness semantics for non-idealized belief.
Visitors to the project are very welcome.
A number of European countries offer Visiting Research Fellowships for visiting established ERC projects.
Several non-European countries offer International Arrangement Funding for researchers to temporarily join ERC teams in Europe.
Contact Johannes Stern to discuss these or further opportunities.