Skip to main content

Lexical Meaning and Logical Inference

Final Report Summary - LEXMEAN (Lexical Meaning and Logical Inference)

1. General overview

“Lexical Meaning and Logical Inference” (LexMean, PIEF-GA-2011-302596) is a project at the interface between philosophy of language, linguistics and logic. Its aim has been to advance our understanding of the mechanisms through which language and grammar represent logical properties and relations. The researcher, Isidora Stojanovic (Jean Nicod Institute/CNRS) comes from an interdisciplinary background in philosophy, logic, and cognitive science. The scientist in charge, Louise McNally (Universitat Pompeu Fabra) specializes in semantic theory and the semantics/pragmatics/morpho-syntax interfaces. The project, hosted by the Formal Linguistics Group (GLiF), has given rise to an extremely fruitful collaboration. The researcher has benefitted by acquiring advanced training, competences and skills in a new discipline (linguistics) while the host group has benefitted through interaction with a researcher with background in philosophy and cognitive science, the result being new lines of research and a broadened scope for the group's research interests. The research conducted in this projects has initiated a new dynamics in the interdisciplinary area that unites linguistics, philosophy and logic.

2. Overview of objectives, hypotheses and conclusions

The main objective of the project "Lexical Meaning and Logical Inference" has been to gain a better understanding of the ways in which language and grammar encode logical relations. We examined the connections between three debates in three different disciplines: the problem of logical constants in philosophy of logic, the debate on the functional vs. lexical distinction in linguistics, and the debate on the semantics vs. pragmatics distinction in philosophy of language, under the conviction that a cross-fertilization between disciplines will promote qualitatively new advances in the respective fields. Our hypothesis was that these seldom-connected debates actually addressed the same set of underlying issues. In particular, we have aimed to show that in order to give an account of semantic content, and to provide a principled answer to the question of where to draw the line between semantics and pragmatics, one needed to inquire into the criteria that distinguish the logically valid inferences from other inference patterns, as well as understand the distinction and the relationship between the lexical and the functional properties of the expressions of a given language.
By establishing a tight link between those aspects of language that are reponsible for logical inferences and those that play an essential role for grammar (such as connectives, determiners, or tense morphemes), the project has provided new incentives to the interdisciplinary research that brings together logic, linguistics and philosophy of language.

While the core objectives of our project belong to a general level of abstraction across languages and were largely approached using conceptual analysis (a standard methodology in analytic philosophy), it has been crucial to our project to support these general claims with specific case studies. In those case studies, we have looked at several classes of expressions that have typically posed problems for the lexical vs. functional dychotomy. The range of categories investigated in the course of the project include: (i) adjectival structure, evaluative markers and aesthetic predicates; (ii) indexical and personal pronouns; (iii) doxastic expressions (such as 'believes that'); (iv) verbs of communication (such as 'says that'); (v) the auxiliary 'will' and future tense morphemes; (vi) quantifiers and, more specifically, implicit quantifier domain restriction.

In addition to its scientific objectives, importance has been given to two other sets of goals: on the one hand, training and career-development goals for the researcher, and on the other, impact and outreach objectives for the project as a whole. With respect to the former, the researcher acquired the desired competences and skills in linguistics and defended her “Habilitation à diriger les recherches” (March 2014, at EHESS, Paris), which, in France, is seen as the highest academic distinction. With respect to the latter, the project brought a new dynamics into the host group, initiated unexpected collaborations, and gave rise to novel interdisciplinary approaches to certain issues. As evidence of the general impact on interdisciplinarity and outreach to other disciplines, the researcher has been invited to be the Program Co-Chair of CONTEXT 2015, a major interdisciplinary conference that brings together computer science, artifical intelligence, linguistics, philosophy, psychology, and neuroscience, with proceedings published in Springer LNAI series (
Among the many collaborations that have emerged from the project, worthy of mention is the symposium “Lexical semantics, grammatical structure, and the conceptualization of events” to be held at the Center for Language, Logic and Cognition at the Hebrew University of Jerusalem (HUJI). The idea for this symposium, for which the scientist in charge is one of the convenors and the researcher, one of the keynote speakers, was inspired by the themes of the LexMean project.

3. Overview of dissemination activities

In terms of publications, the project has resulted in two research articles in journals, two research articles in peer-reviewed collective volumes, two books reviews, and a conference article, all published within the project duration, as well as a number of forthcoming publications, including two other research articles in peer-reviewed collective volumes (in press, available online as prepublications), two survey articles, and two more articles in preparation. In the 24 months that the project lasted, the research gave a total of 22 presentations. For half of those, she was an invited speaker at international conferences or workshops (including the keynote speaker at the congress Philosophers' Rally, Warsaw 2-4 July 2014), five were submissions to peer-reviewed calls for papers/abstracts, one was an outreach presentation to broader audience, and the rest were given in department colloquia, seminars, and project or network-related workshops.

4. Examples of concrete results

To illustrate the kind of work that was conducted and the results to which it has led, we may mention two articles, one more conceptual and the other, more focused on a specific class of expressions: (a) Stojanovic, I. “Prepragmatics: Widening the Semantics-Pragmatics Boundary”, in Burgess, A. and Sherman, B. (Eds.), Metasemantics: New Essays on the Foundations of Meaning. Oxford: Oxford University Press (2014): pp. 311-326. (b) McNally, L. and Stojanovic, I. “Aesthetic Adjectives”, forthcoming in Young, J. (Ed.) The Semantics of Aesthetic Judgements, Oxford: OUP.

In (a), with the general aim of understanding what demarcates genuinely semantic properties or phenomena from the rest, it is argued that the dyadic conception of the semantics-pragmatics interface leads to an impasse. It is proposed that the best way out of the impasse is to recognize a third, separate level, which we call 'prepragmatics'. Prepragmatic phenomena, in contrast to full-fledged pragmatic reasoning, do not involve any higher-level metarepresentational capacities and are truth-conditionally relevant. The article thus offers a novel, albeit controversial picture of our language architecture.

In (b), we studied adjectival structure in relation to aesthetic adjectives, of which 'beautiful' is the paradigm example. We investigated various linguistic criteria, some known from the literature and others that we put forward for the first time, which we use to classify adjectives. Some of these criteria (notably, gradability and the presence or not of an experiencer argument) are related to logical aspects of meaning and have been proposed to manifest themselves in the functional part of morpho-syntax. This work relates to the general aim of the project in that it shows how the categorization of a class of linguistic expressions (in this case, of adjectives) based on various logical and non-logical aspects of meaning leads to new predictions about language use (in this case, aesthetic discourse). This article also nicely illustrates the fruitful collaboration between the researcher and the scientist in charge that the project made possible. Although still forthcoming, a prepublication draft of the article has been made available online and the expected impact of the article is evidenced by the fact that within a week of posting it figured in the top 3% by 30-day views on website.

5. Project webpage and contact information

Isidora Stojanovic (researcher), Institut Jean-Nicod, ENS – Pavillon Jardin, 29 rue d'Ulm, 75005
Paris, France; email: isidora.stojanovic [that character]
Louise McNally Seifert (scientist in charge), Universitat Pompeu Fabra, c. Roc Boronat 138, 08018
Barcelona, Spain; email: louise.mcnally [that character]