Institut Jean Nicod

Accueil > Séminaires/Colloques > Archives > Cycle de Conférences > 2012-2013 > Paul Pietroski > Paul Pietroski

Paul Pietroski

Paul Pietroski (University of Maryland
Professeur invité à l’ENS, juin 2013.
Site web

Les 6, 13, 20 et 27 juin 2013 de 14h30 à 16h30

Institut Jean Nicod, Pavillon Jardin,
Ecole Normale Supérieure - 29, rue d’Ulm 75005 Paris. Salle de réunion au RDC.

Meanings First

Some papers will be available at pietro


Lecture 1 : Framing Event Variables
Lecture 2 : I-Languages, T-Sentences, and Liars
Lecture 3 : Word, Concepts, and Conjoinability
Lecture 4 : Meanings as Concept Assembly Instructions

Main Questions : what are linguistic meanings, and how are they related to concepts ?

Background Assumptions : Let’s say that a Human Language is a spoken or signed language that any biologically normal child can acquire given an ordinary course of linguistic experience. Each such language connects boundlessly many articulations—roughly, sounds or gestures—with meanings of some kind. Human Languages permit ambiguity, but only in constrained ways. For example, the sound of ‘The duck is ready to eat’ has more than one meaning. However, ‘The duck is eager to eat’ is less ambiguous, as is ‘The duck is easy to eat’. So each Human Language connects articulations with meanings in an unbounded but constrained way ; and each human child can connect articulations with meanings in this way. So we can ask : what are these meanings that children can connect with articulations in these unbounded but constrained ways ?

Two Caveats (which turn out to be important)

First, humans naturally enjoy one or more mental languages, whose atomic expressions are symbols that have “contents” of some kind. We may share these mental languages with other animals. But these languages do not connect articulations with meanings in the relevant sense.
Second, humans can invent languages whose well-formed formulae have semantic properties by stipulation. But ordinary children do not ordinarily acquire such languages.

Three Theses (at least one of which is wrong)

1. Human Languages are I-languages in Chomsky’s sense :

 each Human Language is a biologically implemented generative procedure 

 that connects articulations with meanings in a certain unbounded but constrained way

2. Human Languages are T-languages in Davidson’s sense :

 for each Human Language, there a Tarski-style theory of truth that

can serve as the core of a theory of meaning for that Language

3. If Human Languages are I-Languages, then Human Languages are not T-Languages.

Plan for the Lectures (after initially explaining the assumptions, caveats, and theses)

A. Briefly argue for thesis 1, contrasting I-languages with E-languages

B. Argue for thesis 3, focusing on three objections concerning truth, denotation, and predication
—an objection concerning predicates that are allegedly true of “events”
—a “Liar Paradox” objection, concerning the relation of truth/satisfaction to meaning
—an objection concerning expressions that allegedly denote “entities”

C. Sketch an alternative to thesis 2, and develop it in as much detail as time permits
—Human Language meanings are instructions for how to build concepts of a special sort
—more specifically, meanings are mechanically executable instructions (or recipes) for 
 how to build systematically composable concepts that are monadic and conjunctive



Sponsored by the ERC Advanced Grant ‘Context, Content and Compositionality’ (F. Recanati) and by the Labex IEC (program ‘New Ideas in the Philosophy of Mind and Language’)