Institut Jean Nicod

Accueil du site > Séminaires & Colloques > Séminaires > Archives > 2015-2016 > Mind & Language > Presentation


M&L Seminar

The Mind and Language Seminar


Contact : Tomoo Ueda ou Michael Murez


Past sessions

Mercredi 23 septembre de 14h à 16h - Salle Langevin
Isidora Stojanovic (IJN, CNRS),
Expressing Value-Judgements in Context

In this talk, we'll look at different semantic and pragmatic mechanisms of expressing and/or conveying judgements of value, and, in particular, of aesthetic value. While such judgements may, of course, be expressed by means of evaluative terms, such as “excellent”, “mediocre”, or “bad”, they may also be expressed using terms whose primary usage is not evaluative, such as “intense” or “somber”. What is more, in the latter case, whether a given statement conveys a positive or a negative judgement crucially depends on the context. We shall suggest that in most such cases, the application conditions of the term involve several dimensions of evaluation, and that the overall evaluation conveyed depends on which dimensions are contextually most relevant.

Mercredi 30 septembre de 14h à 16h - Salle Langevin
Geoff Nunberg (School of Information, UC Berkeley),
The social life of slurs

The words we call slurs are just plain vanilla descriptions like cowboy and coal scuttle. They don't semantically convey any disparagement of their referents, whether as content, conventional implicature, presupposition, “coloring” or mode of presentation. As plain descriptions, they have nothing in common with “thick terms” that mix categorization and attitude, nor are they semantically the direct expressions of strong emotion. What distinguishes pairs like kraut and German is a difference of metadata rather than meaning: the former is the conventional description for Germans among Germanophobes when they are speaking in that capacity, in the same way swerve! is the conventional expression that some teenagers use for “Oh, joy!” when they’re stressing that social identity. To use a slur is to exploit the Maxim of Manner (or Levinson’s M-Principle) to assert one’s affiliation with a group that has a disparaging attitude towards the slur’s referent. This account is sufficient to explain all the familiar properties of slurs, such as their speaker orientation and “nondetachability.” It also explains some of their unexplored features, such as the variation in tone and strength among the different slurs for a particular group and the role of slurs in shaping the social identity of their users, all with no need of additional linguistic mechanisms.

Mercredi 7 octobre de 14h à 16h - Salle Langevin
Enrico Terrone (Collège d'études mondiales, FMSH),
Twofileness: Fictional Characters as Generators of Mental files

The claim that fictional characters are abstract artifacts raises two issues. The first concerns function: given that artifacts normally have functions, which is the function of a fictional character? The second concerns experience: in reading a novel or watching a movie we treat fictional characters as concrete individuals; how can such a phenomenology fit with an ontology according to which fictional characters are abstract artifacts?
I will address the second issue by addressing the first one. I will claim that the function of a fictional character, as an abstract artifact, is the generation of a special pair of mental files; a ‘fiction file’ about the character as an individual in the fictional world, and a ‘source file’ about the character as an abstract artifact in the real world. I will call this hypothesis ‘twofileness’ and I will argue that it can reconcile our experience of fictional characters as concrete individuals with their ontological status, that is, abstract artifacts. 

Main References
Recanati, François, 2012, Mental Files, Oxford: Oxford University Press.
Smith, Murray, 2011, “On the Twofoldness of Character.” New Literary History 42.2: 277-294.
Thomasson, Amie, 1999, Fiction and Metaphysics, Cambridge: Cambridge University Press.

Mercredi 4 novembre de 14h à 16h - Salle Langevin
Emar Maier (University of Groningen),
Talking about Frodo: fictional anchors, descriptive imagination, or pretense?

Fictional proper names pose a notoriously difficult puzzle for semantics. Since Frodo never existed, "Frodo" does not refer and, by compositionality and rigidity of proper names, (1) and (2) fail to express a proposition.
(1) Frodo is a hobbit born in the Shire.
(2) Frodo is a fictional character made up by Tolkien.
To solve this puzzle I propose a formal semantic approach based on the pretense-theoretic analysis of fiction as 'prescriptions to imagine' (Walton 1990). My account builds on the communication-theoretic semantic framework of Kamp (1990,2015), which in turn builds on the basic formalism of Discourse Representation Theory (DRT). The idea of this framework is to model interpretation dynamically in terms of an update on a formal mental state representation consisting of mental files (or anchors, in Kamp's terminology) and attitudes. Specifically, I propose to analyze the interpretation of fictional statements (like (1)) as dynamic updates on an imagination component of the interpreter’s mental state, while plain assertions (including metafictional ones, like (2)) are updates on a belief component.
Proper names are uniformly treated as presupposition triggers looking for an antecedent mental file to bind to. But how do we resolve the presuppositions of fictional names, for which we have no acquaintance-based mental files? In this talk I critically evaluate three answers to this question: (i) introduce special fictional anchors (a subspecies of what Recanati 2012 calls unloaded indexed files) (cf. Kamp 2015); (ii) interpret the name descriptively within the prescribed imagination (cf. Currie 1990); and (iii) analyze fiction in terms of a higher level mental operation, call it pretense or simulation, that transforms anchors into pretend-anchors, and beliefs into pretend-beliefs (cf. Recanati 2000).

Mercredi 18 novembre de 14h à 16h - Salle Langevin
Mind & Language : Reading Group

Mercredi 2 décembre de 14h à 16h - Salle Langevin
Robert May (UC Davis),
Pejoratives as Fiction - Christopher Hom (Texas Tech University) & Robert May (University of California, Davis)

Fictional terms are terms that have null extensions, and in this regard pejorative terms are a species of fictional terms: although there are Jews, there are no kikes. That pejoratives are fictions is the central consequence of the Moral and Semantic Innocence (MSI) view of Hom and May (2013). There it is shown that for pejoratives, null extensionality is the semantic realization of the moral fact that no one ought to be the target of negative moral evaluation solely in virtue of their group membership. In having null extensions, pejorative terms are much like mythological terms like ‘unicorn horn’ that express concepts with empty extensions, even though it was thought otherwise: people who falsely believed the mythology were mislead into thinking that ordinary objects (i.e. whale tusks) were magical objects, and pejoratives terms work likewise. For example, the term ‘kike’ is supported by the ideology of anti-Semitism, and speakers who fall prey to its influence (perniciously or not) are mislead into thinking that ordinary people (i.e. Jews) are inherently worthy of contempt. In this paper, we explore the consequences of this parallelism, with an eye to criticisms of MSI. In particular, we will re-visit identity expressivist views - those that hold that there are kikes and that they are Jews, and hence deny null extensionality - arguing that this embeds a mistake of fiction for fact. Among the issues to be discussed are the role of fictional truth in understanding pejorative sentences and the relation of the semantics of pejoratives to offensive use of language. We conclude with meta-semantic reflections on the nature of word meanings.

Mercredi 9 décembre de 14h à 16h - Salle Langevin
Gregory Bochner (Université libre de Bruxelles),
The Problem of the Essential Index.

"I try to motivate a reduction of the problem of the essential indexical, taken as a problem for what Perry calls “the doctrine of propositions,” to what would be a simple but more fundamental problem, one arguably manifested in a wide variety of puzzling phenomena. “The problem of the essential index,” as I will call it, arises for the general view that the truth-conditions of a belief are fully determined by its content. I submit that empirical beliefs falsify this view. While the contents of empirical beliefs have relative truth-conditions, the empirical beliefs themselves have absolute truth-conditions, which essentially involve some specific index of evaluation, viz. the particular situation that is both the causal source and the target of the relevant information. The problem of the essential indexical is usually taken to have highlighted a new contrast between *de se* beliefs, on the one hand, and *de dicto* and *de re* beliefs, on the other. As against this, I argue that what most generally threatens the doctrine of propositions is the old contrast between *de dicto* and *de re* beliefs (which only include the *de se* beliefs). Though the two contrasts are real, they raise distinct problems, which should not be conflated. Only the second bears interesting connections to indexicality and to the doctrine of propositions. I submit that the solution to this second problem is to recognise that *de re* beliefs are typically empirical beliefs whose index of evaluation is a world centred on an object of acquaintance. My claims will point to a more general view, according to which: (i) indexical relations are essential because they play the role of anchoring the content of an empirical belief to its index of evaluation; (ii) reference can be direct in perception, thought, and language, but if material objects enter the truth-conditions of our attitudes and statements, this is not (indirectly) by being represented by the contents of our minds, but (directly) by being present themselves in the relevant situation of evaluation of our empirical judgements. Concrete attitudes and speech acts refer to concrete situations, and it is the situations, not the contents, that involve the concrete referents."

Mercredi 6 janvier de 14h à 16h - Salle Langevin
Emile Thalabard (Université Paris IV Sorbonne),
Accessibility, attention and qualia.

Abstract: In the current debates, it is usually held that the alledged richness of conscious sensory episodes is hard to handle for Global Workspace accounts of consciousness. Ned Block's overflow argument, which relies on the high capacity of short term visual memory stores, constitutes a challenge given the limited capacity of working memory.
In this talk, I will argue that, while retaining the core ideas of the GW model (ie: consciousness is an all or nothing phenomenon, which consists in the global availability of sensory contents for further cognitive processing), it is possible to account for phenomenal overflow. This account relies on a distinction between weak and strong access, and accordingly, on a distinction between entering working memory, and being maintained in working memory. Contrary to the usual anti-Blockean strategy, I will accept phenomenal overflow,while arguing that a suitably qualified necessity claim regarding the relationship between attention and consciousness may be maintained. I will contrast this account with two neighbouring propositions: Prinz's AIR Theory, and Carruthers' account of phenomenal consciousness.

Mercredi  13 janvier de 14h à 16h - Salle Langevin
Max Kölbel (ICREA, Barcelona),
Pants on Fire and Knickers in a Twist.

Abstract : Suppose (i) we want to say that what explains the difference between “My pants are on fire.” and “Max’s pants are on fire.” is that the former expresses a de se content” or “centered content”, while the latter does not. Suppose further, (ii) that we want to say that “It might be green.” expresses a de se or centered content about epistemic possibility, and that this is what differentiates it from, for example, “It might be green given my information.”. Then we have a problem in specifying the contents of the following three sentences:
1a It might be green.
1b It might be green given my information.
1c It might be green given Max’s information.
The aim of this paper is to explore a solution which consists in claiming that 1a and 1b both express the same centered content, but that they differ in the conversational contribution they make.

Mercredi  20 janvier de 14h à 16h - Salle Langevin
Olivier Massin (Université de Genève),
Chromatic Mixtures: Purple, Orange… Green(?)

Mercredi 3 février de 14h à 16h - Salle Langevin
Olivier Massin (Université de Genève),
Affective Mixtures: On Mixed Feelings

Mercredi 10 février de 14h à 16h - Salle Langevin
Michele Palmira (Barcelona, LOGOS),
Propositions and Arbitrary Reference

In his 1965 seminal article “What Numbers Could Not Be”, Paul Benacerraf draws on the existence of multiple and equally adequate reductions of natural numbers to sets to conclude that arithmetical realism, viz. the view that natural numbers exist, is false. Several authors (e.g. Bealer 1998, Crane 1992, Jubien 2001, King 2007, Melia 1992) raise a Benacerraf-style problem for the view that propositions are sets of possible worlds or functions from worlds to truth-values. The aim of the talk is twofold: to clarify what it takes to raise a Benacerraf-style problem for propositions, and to offer an as yet unexplored solution to this problem. The solution hinges on the idea that propositions-naming expressions arbitrarily refer to world-theoretic entities. In order to argue for this solution, a specific semantics for arbitrarily referring expressions is offered, and some objections against the very idea of arbitrary reference are addressed.

Mercredi 17 février de 14h à 16h - Salle Langevin
Tom Avery (IJN),
"Characterising Consciousness"

Abstract: In this talk I consider a peculiar situation that obtains in the philosophy of mind: There is widespread agreement to the effect that there’s a generic feature which distinguishes sensory experiences as such; this is standardly characterised in terms of their possessing a ‘phenomenal character’, or their being such that there’s ‘something  it’s like’ to undergo them. There is then a collection of disputes over what it takes for our sensory experiences to be specific ways that they are in respect of this generic feature; these include the Dispute between qualia theory and representationalism, and the dispute between intentionalism and naïve realism/disjunctivism, as well as a wide range of disputes concerning the kinds of contents that our sensory experiences possess. At a glance, then, there appears to be a deep overlap in concern across a variety of distinct disputes, in which the participants are all interested in accounting for what it takes for our sensory experiences to be the specific ways that they are in respect of the generic feature which distinguishes sensory experiences as such. However, the appearance of overlap here is undermined by the fact that, though theorists often agree in their labels for the relevant feature, there’s really no widespread agreement as to what the generic feature is, that distinguishes sensory experiences as such. In this talk, I consider whether there’s any more illuminating way of characterising the relevant feature, so that the positions advanced across these disputes can all be accurately construed as positions on what it takes for our sensory experiences to be the ways that they are in respect of this feature, so characterised. I also consider the dialectical benefits that such a characterisation would bring.

Mercredi 16 mars de 14h à 16h - Salle Celan
Filipe Drapeau Vieira Contim (EA 1270, U. Rennes 1),
Descriptive Names Without Paradox

Résumé. Les noms descriptifs tels que « Jack l'Éventreur » ou « Neptune » (utilisé au temps de Le Verrier) posent un défi au théoricien de la référence directe dans la mesure où ils brouillent la distinction entre référence et description : d'un côté, leur référence semble déterminée par la satisfaction d'une condition descriptive, au même titre que les descriptions définies ; de l'autre, ce sont des désignateurs rigides de jure dont on peut présumer qu'ils contribuent à exprimer des propositions singulières à propos de leur référent, à l'instar des noms propres ordinaires. Ce caractère hybride induit une conséquence paradoxale : l'introduction d'un nom descriptif permettrait de connaître a priori et sans effort des vérités contingentes à propos du référent.

Je me propose de dissiper le paradoxe en m'appuyant sur une taxonomie des noms descriptifs. Je distinguerai trois catégories de noms : les noms descriptifs proxys (« Julius », « Newman 1 »,...) dont la référence est à la fois fixée par description et connue par description ; les noms descriptifs singuliers (« Jack l'Éventreur », « Gorge Profonde »,?) dont la référence est fixée par description mais connue par accointance indirecte, via les traces du référent ; et enfin les noms pseudo-descriptifs (« Neptune ») auprès desquels les descriptions ne jouent, contrairement aux apparences, aucun rôle référentiel (fixation de la référence) ou cognitif (contenu des pensées). Je tâcherai de montrer que seuls les noms descriptifs singuliers combinent les traits sémantique et cognitif requis pour générer de l'a priori contingent. Je soutiens néanmoins que cette forme d'a priori contingent est inoffensive dans la mesure où l'on se contente de connaître a priori des contenus qu'il faut déjà connaître empiriquement pour comprendre un nom descriptif singulier. L'introduction d'un tel nom n'augmente donc pas notre connaissance du monde.

Abstract. Descriptive names such as ‘Jack the Ripper’ or ‘Neptune’ (as used by Le Verrier) raise a challenge for the direct reference theorist as they blur the distinction between reference and description. On the one hand, they designate their object via the satisfaction of a descriptive condition, on a par with definite descriptions. On the other hand, descriptive names are de jure rigid designators, and as such they presumably contribute to express singular propositions about their referent, just like ordinary names. This hybrid nature induces a well-known paradox: just by coining a descriptive name, one could know a priori and without effort some contingent truths about the world. 

I will try to dissipate the paradox of a priori contingent by offering a refined taxonomy of descriptive names. I distinguish between three kinds of such names: (i) proxy descriptive names (‘Julius’, ‘Newman1’…) whose referent is both determined and known by description; (ii) singular descriptive names (‘Jack the Ripper’, ‘Deep Throat’…) whose reference is fixed by description yet known by acquaintance; (iii) dummy descriptive names (‘Neptune’) with respect to which definite descriptions play no role, neither referential (fixation of reference) nor cognitive (content of thoughts), appearances notwithstanding. If I am right, only singular descriptive names combine the semantic and cognitive features required to generate the paradox of a priori contingent.  I claim, however, that this form of a priori contingent is harmless. I will show indeed that introducing a descriptive name does not bring new knowledge about the world, as it merely puts one in a position to know a priori what was already known empirically in order to understand the name.   

Mercredi 23 mars de 14h à 16h - Salle Celan
Elmar Geir Unnsteinsson (University College Dublin),
The Edenic theory of reference

Abstract: In this talk I argue for a theory of the optimal or proper function of referring to an object with a linguistic expression, called the edenic theory of reference. First, I define linguistic reference in terms of Gricean communicative intentions. Roughly, to refer with an expression is to speaker mean a singular proposition and intend the hearer to use the occurrence of the expression as evidence for which object is intended. Secondly, I propose a doxastic optimality constraint on such acts of referring, stating that the speaker must not have any false beliefs about the identity or distinctness of the object in question. Lastly, I develop two independent arguments for the constraint. One, that such false beliefs - called separatory and combinatory confusion - constitutively corrupt the evidence provided by the utterance. Two, that the constraint is part of  pragmatic competence and shows up in criticisms we make of each other as speakers and hearers. In uttering a singular term one normally represents oneself to the audience as not having any confused beliefs about the object.

Mercredi 30 mars de 14h à 16h - Salle Celan
Sajed Tayebi (IPM, Teheran)
In Defence of Predicativism against Jeshion's Challenge

Abstract: According to predicativism, a proper name is always used as a predicate, and where it might seem not to be, we found on closer scrutiny that the name is the predicate part of a complex non-predicate phrase. The name N, then, is regarded as a multiply applicable predicate which is true of an object o if and only if o is called N. Following the Burge’s inspirational defense of predicativism, most predicativists regard the power of predicativism to provide a unified semantic theory that assimilates apparently referential uses of names with apparently predicative uses of them as the main argument to the effect that predicativism is superior to the orthodox view about names (i.e. referentialism). This argument, which is usually called the Unification of Argument, is recently challenged by Robin Jeshion. The unification argument is rested on the assumption that predicative uses of a proper name to which the argument appeals, are normal, literal uses of the name. Jeshion argues that the predicativist cannot simply assume this, and she needs to somehow justify it, because there are a series of examples in which, though a name is used predicatively, it does not seem to be true of the objects called N. Predicativist seems to have no choice but to regard Jeshion’s examples as involving non-literal predicative uses of names. But then, and this is the challenge, the predicativist owes us an explanation of why the referentialist cannot  regard the original predicative uses, which inspired predicativism in the first place, as non-literal uses as well.

In this talk, I am going to provide an answer to Jeshion’s challenge. For this purpose, after arguing for the inadequacy of the Fara’s answer to the challenge, I will provide a series of examples exactly parallel to Jeshion’s examples in which a name N is used in them for an object that is not called N. My examples, however, differ from Jeshion’s in the position where names are apparently used. While in Jeshion’s examples names are used in predicate position, in my examples names are used in subject position. I will argue that both the predicativist and the referentialist have no choice but to regard uses of the names in my examples as not-literal uses, while they regard the standard uses of names in subject position as literal ones. I will argue that, due to the parallelism of my examples and Jeshion’s, the justification for this approach regarding names in subject position provides a firm ground to take a similar approach with respect to names in predicate position. At the end, I will suggest that my examples serve even more than as an answer to Jeshion’s challenge. They provide a new basis for superiority of predicativism over referentialism.

Mercredi 6 avril  de 14h à 16h - Salle Celan
Ethan Jerzak (University of California, Berkeley)
Two Ways to Want

Abstract: I present hitherto unexplored and unaccounted-for uses of ‘wants’. I call them advisory uses, on which information inaccessible to the desirer herself helps determine what it’s true to say she wants. I show that extant theories by Stalnaker, Heim, and Levinson do not predict this use. I consider a few ways to fix this deficiency, and conclude by plumping for a relativist semantics, according to which desire attributions express information-neutral propositions. The truth of a desire-attribution, on the view I arrive at, depends on the state of information at the context of assessment. I conclude with a pragmatic account of the purpose of desire attributions that explains why it made sense for them to develop in this way.

Mercredi 13 avril de 14h à 16h - Salle Celan
José Luis Bermudez (Texas A&M University)
The Rationality of Self-Control
See : Lecture series

May 25th, 2-4 pm - Institut Nicod, Pavillon Jardin, ENS, 29 rue d’Ulm Paris, Salle du RDC.
Eliot Michaelson (King's College London)
Towards a Grand Unified Theory of Reference 

Sometime in the 20th century, philosophers largely abandoned the project of offering a unified theory of reference. This, I suggest, was partly the result of the temptation to handle a certain range of hard cases—involving speakers who are confused about some relevant aspect of the world—in different ways when different expressions were involved. The result has been an explosion of theories of reference: one for names, another for demonstratives, yet another for pronouns, and so on. I shall argue that the resulting theories are highly unsatisfactory, and that we can in fact do better by returning to the project of offering a unified theory of reference. I then proceed to sketch just such a theory, inspired by some rather cryptic remarks of Strawson’s. On reflection, these remarks shall turn out not to be so cryptic after all, and the theory they suggest will prove capable of explaining the full spectrum of hard cases in a unified, but not too unified, manner.