Salvador Mascarenhas
Assistant Professor

Ecole Normale Supérieure
Department of Cognitive Studies
29 rue d'Ulm, room PJ005
75005 Paris, France


I am an Assistant Professor at Ecole Normale Supérieure’s Department of Cognitive Studies (DEC), a member of Institut Jean-Nicod, and the principal investigator in the Agence Nationale de Recherche grant Between Language and Reasoning. I got my PhD at New York University’s Department of Linguistics in 2014, where I wrote a dissertation on the relationship between human reasoning, natural language semantics, and the philosophy of language. Between 2014 and 2016, I was a Junior Research Fellow at St Catherine’s College, Oxford, affiliated with the Faculty of Philosophy.


Research in linguistic semantics and the philosophy of language in the past forty years has produced sophisticated mathematical models that represent the meanings of natural language utterances and explain how meanings relate to one another to form entailment patterns. At the same time, research on human reasoning within psychology has discovered a wealth of fallacious inference patterns, establishing that human reasoning is fallible in highly predictable ways. The psychological study of reasoning has been characterized by extensive experimental work analyzed in terms of theories focusing on the processes of reasoning. On the other hand, linguistic semantics has a longstanding tradition of formal rigor and a focus on logical thinking, but it has so far largely ignored fallacies. The two fields overlap significantly, but they have progressed almost completely in parallel, with little interaction. The main aim of my research is to fill that gap by extending linguistic semantics to the study of human reasoning.

My approach has two main components. The first involves recasting the account of human reasoning known as mental model theory in a formally explicit system, incorporating relevant insights from my work on inquisitive semantics. Together with Philipp Koralus I have developed a version of mental-models theory that locates the origin of reasoning failures in a question-asking and answering process. The second component is an interpretation-based theory of (some) reasoning failures that explains these failures in terms of modern theories of implicature. This second theory contrasts with its reasoning-based competitors in that it assumes as a working hypothesis a logically sound reasoning module that operates on interpretations more complex than meets the eye.

From these two rigorously defined theories, different predictions emerge. These predictions allow for a comparison between the two theories, providing a principled way to tease apart the contributions of general-purpose reasoning mechanisms and of interpretive procedures in our failures of reasoning.

I also have an active interest in, and have worked on, the following topics in semantics and philosophy of language

  • the semantics of natural language questions and erotetic logics;
  • dialog coherence in the pragmatics of questions and answers;
  • dynamic semantics for anaphora and presupposition;
  • alternative semantics for indefinites and disjunction;
  • positive polarity and free choice.

Articles and manuscripts


Fall 2020

LING302 Semantics II

Thursdays 2pm–4pm, provisionally salle séminaire du DEC on 29 Ulm.

Course description

This is the second module on natural language semantics offered at the Department of Cognitive Studies. We will cover advanced topics in semantics that weren’t covered in Semantics I, and we will revisit topics from Semantics I under a new light. We will cover the following broad topics.

  • Quantification
  • Dynamic and update semantics
  • Intentional semantics: propositional attitudes, modals, conditionals
  • Meaning beyond declaratives: questions and imperatives


Semantics I, or permission of the instructor


  1. a small number of homework assignments (4 or 5)
  2. discussion of primary readings in class
  3. term paper (8–16 pages)

COGSCI302 Super Semantics

with Philippe Schlenker and Emmanuel Chemla

Mondays 2pm–4.30pm, provisionally salle Ribot on 29 Ulm.

Course description

While formal semantics has been a success story of contemporary linguistics, it has been narrowly focused on spoken language. Systematic extensions of its research program have recently been explored: beyond spoken language, beyond human language, beyond language proper, and even beyond systems with an overt syntax. First, the development of sign language semantics calls for systems that integrate logical semantics with a rich iconic component. This semantics-with-iconicity is also crucial to understand the interaction between co-speech gestures and logical operators, an important point of comparison for sign languages. Second, several recent articles have proposed analyses of the semantics/pragmatics of primate alarm calls, an important topical extension of semantics. Third, recent research has developed a semantics/pragmatics for music, based in part on insights from iconic semantics. Finally, the methods of formal semantics have newly been applied to reasoning and to concepts, which do not have a syntax that can be directly observed. The overall result is a far broader typology of meaning operations in nature than was available a few years ago. The course will offer a survey of some of these results.


Students should have an ability to follow formal analyses, and they should thus have taken a serious introduction to formal logic or to formal semantics, or have significant experience with mathematical theories. If in doubt, please check with the instructors.


  • two two-page squibs (short discussions) due early in the semester; squib #2 will ideally be a term-paper outline
  • Term paper