• Increase font size
  • Default font size
  • Decrease font size


E-mail Print PDF

The meetings take place on Tuesdays, from 16.30 to 18.30 in the salle de réunion, 29 rue d'Ulm, Pavillon Jardin.   



June 18, 2013: Indrek Ireland, "Rules of Use"


Throughout the 20th century it was a common idea in philosophy of language that for an expression to be meaningful is for it to be governed by a rule of use. For example, it was mentioned by Peter Strawson, David Kaplan, John Perry, and Scott Soames. However, nobody went past very general remarks in discussing it. Even worse, it came to be widely seen as inconsistent with “truth-conditional semantics” and subject to the so-called Frege-Geach problem. This led other philosophers to view the idea as vague and mystical, as too radical and obviously problematic, and think of it as ultimately not really worth our time because of there being clear and tractable formal substitutes like characters. For example, here’s Jason Stanley’s summary assessment of it in his survey article “Philosophy of Language in the Twentieth Century” (my emphasis):

Whereas the notion of a rule of use is vague and mystical, Kaplan’s notion of the character of an expression is not only clear, but set theoretically explicable in terms of fundamental semantic notions. (Stanley 2008)

 My aim in this paper is to therefore take the idea that meaningfulness has to do with rules of use and first make it precise and demystify it. I then want to also show that it’s consistent with “truth-conditional semantics” and thus not radical, and that it’s not subject to the Frege-Geach problem, and thus not obviously problematic. Finally, I will argue that it is very much worth our time because it can explain why doing descriptive semantics in terms of characters works in the first place, and because it enables us to provide a semantics for expressions which we can’t give one in terms of characters.


June 11, 2013: Dan Zeman, "Temporalism in an Extensional Framework" - CANCELLED


May 14, 2013: Ernesto Perrini, "Green Leaves Again".


Pia paints the russet leaves of her maple green, and says:

1. The leaves are green

A botanist asks if she has green leaves for a study. She answers:

2. The leaves are green,

1 seems to be true, and 2 false.

Kennedy and McNally 2010 explain the different evaluations of 1 and 2 by postulating that they have terms with different senses: Color adjectives have anongradable sense, denoting “the property of manifesting some property which is correlated with the color in question,” and a gradable one denoting that the object has the appropriate quantity or quality of the color. In 1, ‘green’ has a gradable meaning, and in 2, the nongradable one.

They have two arguments to show that color terms ambiguous: (i) the gradable sense, but not the nongradable sense, admits degree modifiers, and (ii) crossed-sense anaphora are not allowed. Both arguments are inconclusive, in each case there is an alternative explanation. I will focus on (i). We can find already in Zwicky and Sadock 1975 reasons to think that distributional arguments such as (i) are not always enough to establish the ambiguity of expressions. In this case, I propose the following reasoning, building on Zwicky and Sadock, to refuse the ambiguity of color adjectives: (a) if color adjectives were ambiguous, this ambiguity should be realizable in some language; (b) supposing that this is not the case, they are either underspecified or there is an unsystematic ambiguity; (c) the different readings of color adjectives are systematic; (d) if the supposition in (b) is right, they are underspecified.

We can also extend the argument of Kripke 1977 concerning defining definite descriptions to establish (d). Supposing a language in which color adjectives have a gradable sense, the nongradable use will be available. Unlike Kripke’s result, however, we will reach a literal use due to a conceptual extension, and not a non literal use due to our fallibility.

Kennedy and McNally propose perspicuous representations of the intended readings of 1 and 2, but they still have to show that such representations belong to the object language, and not to the metalanguage of the theorist – and a distinction in the metalanguage doesn’t show by itself an ambiguity in the object language. Their argument is inconclusive. Moreover, it seems more reasonable to explain Travis’ puzzle as a case of conceptual extension, rather than two different concepts associated with the same word.


April 30, 2013: Fiora Salis, "Thinking and talking about Marlow"


The problem of intersubjective identification arises from the difficulties of explaining how our thoughts and discourse about fictional characters can be directed towards the same (or different) characters given the assumption that there are no fictional entities. In this paper I aim to offer a solution in terms of participation in name-using practices against a recent proposal put forward by Stacie Friend in terms of notion-networks. The key idea is that if name-using practices secure reference, then participation in a practice is a sufficient condition for intersubjective identification of fictional characters.


April 23, 2013: Sara Packalén, "On the Relation between Semantics and Psychology"


Do the mental states and structures of a natural language speaker constrain a semantic theory for her language? The answer seems to depend on what we want a semantic theory for a natural language to do. Minimally, we want it to make the correct predictions about the meanings of the sentences of the language and about grammaticality. However, most theorists agree on that an adequate natural language semantics should do some explanatory work over and above the purely descriptive work. The theory should fit into an explanation of certain core human linguistic capacities, such as our ability to learn a language, understand novel expressions in a language and process linguistic information with high speed and accuracy rate. Now, given that learning and understanding are psychological processes, it is natural to require of an adequate natural language semantics that it fit into a psychological explanation of the mentioned linguistic capacities. On this view, then, an adequate semantic theory for a language L describes the actual states and structures of the mind/brains of the speakers of L (at some level of abstraction). If so, a theory that makes the correct predictions about meaning and grammaticality but misrepresents the relevant mental states and structures of the speakers will be incorrect in a substantial sense. The aim of the talk is to discuss whether this final step, the 'psychological' step, is warranted.


April 16, 2013: Vasilis Tsompanidis, "Accounting for 'Now'-Thoughts"


In this talk I will argue against two traditional accounts of thoughts whose expression would involve the term ‘now’. A presentist explanation based on tensed facts or properties faces problems accounting for (i) ‘now’ thoughts that are about a temporally extended present, and (ii) the time it takes for our brain to process sensory information. Hugh Mellor's B-theoretic reply based on the tenseless properties of tensed thought tokens cannot explain the continuous updating of our ‘now’ thoughts, and why decisions based on them seem rational. I propose an alternative account based on the mental files framework, and discuss how it might be developed in order to solve these issues.


April 2, 2013: Luca Gasparri, "The Reference of Lexical Content"


Despite being a denotational approach to meaning, model-theoretic analysis suffers from severe limitations when it comes to giving a viable representation of what allows competent speakers of a language to identify the occasions of use of non-functional words. In this talk, I will present evidence to support the claim that such inability to tackle the referential dimension of lexical competence depends on some constitutive features of the analytic machinery at work in formal semantics and I will explore its bearings on the question of the inherent / inherited referentiality of word meanings.


March 19, 2013: Markus Kneer, "Relativism and Recalcitrant Objectivists"


Certain types of evaluative discourse give rise to a curious though widely discussed phenomenon called ‘faultless disagreement’. Two speakers might find themselves entangled in genuine disagreement over whether a particular painting is beautiful or not, without either being ‘at fault’ in terms of holding a mistaken belief. A popular way of making sense of the puzzling intuition consists in relativizing truth to the perspectives of the interlocutors: There is indeed a single proposition under dispute (and hence proper disagreement), but that proposition might be true with respect to the perspective of one speaker and false with respect to another.

Despite its simplicity and intuitive appeal, relativism has, as I will attempt to show, a major drawback: it cannot accommodate evaluative claims that are intended and interpreted as conveying ‘objective truths’.  Such claims, it seems, have been entirely overlooked due to the alleged unavailability of hard facts in evaluative domains.  Semantically, however, they are perfectly ordinary. Plentiful they are as well in certain relativist target domains such as aesthetics and morality. The relativist thus faces a dilemma: Either she radically revises her framework, which quite possibly involves giving up the elegant Kaplanian semantics invoked in nearly all leading relativist theories. Or else she reduces the explanatory reach of her theory to the most subjective fields of discourse, in which case we might see little reason for parting ways with a unified monadic approach to truth.


February 26, 2013: Max Seeger, "Immunity to Error through Misidentification and the Self: An Assessment of Alleged Counterexamples"


The immunity thesis regarding the first-person holds that introspection-based self-ascriptions are immune to error through misidentification. Several authors attempt to refute the thesis by appeal to psychopathologies (such as thought insertion, somatoparaphrenia, and anarchic hand syndrome) or experimentally induced body illusions (such as the rubber hand illusion or the body swapping illusion). I defend the thesis with a simple point: The immunity thesis is a thesis about self-ascriptions which are based on introspection. Although the putative counter-examples involve misattributions, they are either not self-ascriptions or they are not based on introspection. Hence, they do not refute the thesis. However, the cases do raise two issues that merit more discussion: the multimodal nature of self-representations and identification freedom as an explanation of first-person immunity.


February 5, 2013: Alexandre Billon, "Perspectives and Possibility"


From Descartes onwards philosophers have taken certainty, understood as unconceivable falsity, as a guide to necessity. Kant then introduced the distinction between a priori and a posteriori claims, which he took to mirror that between certain and uncertain claims. Without questioning this last thesis, Kripke notoriously put forward cases (hereafter Kripkean cases) which, he argued, revealed that apriority and necessity do not always coincide. I argue that we should interpret the Kripkean cases as showing that a given utterance expresses different propositions from different contexts of interpretation. Once this semantic perspectivism is accepted, however, it becomes clear that (i) Kripke’s arguments cannot establish that the a priority of propositions and their necessity come apart (ii) and that they cannot either establish that the a priority of utterances and their necessity come apart unless a contentious form of “objectivism” is assumed. I further argue that semantic perspectivism allows us to save the link between a notion of utterance certainty —certainty*— and necessity and that it prompts a reconception of the role of conceivability methods and conceptual analysis within philosophy.


January 22, 2013: Santiago Echeverri, "Cognitive Value without Modes of Presentation"


According to an orthodox view, co-referential expressions like “Hesperus” and “Phosphorus” are associated with different modes of presentation. This explains why a competent and rational speaker who finds the sentence “Hesperus is Hesperus” trivial can find the sentence “Hesperus is Phosphorus” informative. This paper introduces a novel account based on the notion of identification abilities. Instead of explaining differences in cognitive value in semantic terms, I propose to account for them by means of psychological principles that govern identification routines. I call these principles “constraints,” and show that they differ from traditional modes of presentation in various ways. To this end, I show how perceptual tracking is governed by constraints, and how it is enhanced by the use of the shape of words as a ‘proxy’ of sameness and difference in reference.


January 15, 2013: reading and discussion of Brendan Gillon, "Implicit Complements: A Dilemma for Model-Theoretic Semantics".


December 11, 2012: Emanuel Viebahn, "Speaker Intentions and Semantic Content"


This talk will deal with the role of speaker intentions in semantic theories. It is widely held that the semantic content in context of many indexicals (e.g. of demonstratives and gradable adjectives) is determined by the speaker’s intentions. I will argue that even the best-developed version of this view, due to King (forthcoming), is unsatisfactory.  I will discuss a number of cases in which it either predicts unintuitive semantic content or demands too much of speakers and their intentions. This will lead me to argue that speaker intentions should generally be kept out of semantics. I will then consider alternatives to intention-sensitive semantics and will end by proposing a new version of non-propositional semantic minimalism. On this view, the semantic content of many sentences is a propositional fragment that systematically constrains, but does not determine, the proposition a literal speaker can communicate.


December 4, 2012: Alexandre Billon, "The Present Problem Problem"


We know that our time is present. We are actually certain of that. What might be less obvious is how we can have such a knowledge. Some philosophers have indeed argued that this question is problematic and that it could not be answered unless we embrace presentism or the Block view. I argue that presentism and the Block view are neither necessary nor sufficient to explain our knowledge that we are present and to solve the "Present Problem". I argue more broadly that no transcendental argument deriving a non trivial ontological claim from our epistemological situation toward the present can be cogent and that there is no Present Problem. The only genuine question in the vicinity is why there might have seemed to be one in the first place, something we might call "the Present Problem Problem".


November 13, 2012: Felipe Carvalho, "Conscious Attention, Demonstrative reference and Immunity to Error through Misidentification"


Philosophers have long noticed that there is a tight connection between conscious attention and the way we understand, verify, and come to know propositions containing demonstrative expressions. If I point to an apple on the ground and say ‘this is red and round’, in order to understand my utterance you need to know which object I am referring to, something you can do by allocating your attention in the relevant direction and visually selecting the apple among other concurrently presented objects in the scene. And if you want to verify if other propositions are true of the apple, i.e., whether the thing which is red and round is also shiny, you redeploy the same procedure you used in order to grasp which object I was talking about –you consciously attend to it. In doing so, you come to have knowledge of a singular proposition: that this appleis red, round and shiny.

As such, it seems plausible to suppose that we can use this general point about the epistemic role of attention in order to account for an important epistemological property of perceptual demonstrative thoughts and judgments: that they are, in Sydney Shoemaker’s terms, immune to error through misidentification (henceforth IEM). After all, IEM has been traditionally explained in terms of one’s grounds for making a certain singular judgment, and if conscious attention provides us with the most important way in which we come to know about the properties of objects in our external environment, it is plausible to suppose it might also shed some light on the phenomenon of immunity to error through misidentification.

I will begin by introducing two models that seek to account for IEM phenomena in this manner: the traditional 'way of knowing' (WOK) model, originally developed by Evans (1982) and Peacocke (1983) and the 'reference-fixing property' (RFP) model, introduced by Campbell (1997/1999/2002) as a theoretical alternative to WOK. According to the WOK model, IEM can be traced to the way in which you come to know that a certain visual feature F is instantiated in the scene. If you come to know that F-ness is instantiated by consciously attending to a location L where F-ness is perceived, if you then attribute F-ness to the object at L your judgment ('this is F') will be immune to error through misidentification.

John Campbell, however, has argued that the WOK model cannot be right. In certain circumstances, even if the conditions specified in WOK are met, the corresponding judgment ‘this is F’ may still turn out to be in error through misidentification. In its place, he proposes the RFP model, where IEM is traced to the role of the property F itself in visually selecting the object‘this’ refers to. If F is the property we exploit in order to visually select the object ‘this’ refers to, judgments of the form‘this is F’ will be IEM. And spatial location is, quite plausibly, the property we exploit in order to single out objects in our external environment. This explains why judgments of the form ‘this is there (at the attended location)’ are IEM.

I think this is a mistake, and my goal in this talk will be to defend the orthodox WOK model against Campbell's objections. Once we have a correct understanding of IEM, it should be clear that the counter-examples Campbell deploys against the WOK model are spurious, and do not reveal errors through misidentification. But if we now grant that a wider range of properties, besides spatial location, can figure in perceptual demonstrative judgments that are IEM, the RFP model loses much of its appeal, and ends up yielding some very counter-intuitive results.

This gives us enough reasons to keep the WOK model. This model allows us to maintain Campbell’s correct insights on the role of conscious spatial attention in visual selection and demonstrative reference, without leading to the problems faced by the RFP model. In addition, the WOK model has already been successfully applied in explaining IEM in judgments made with indexicals like ‘I’ and ‘here’,so adopting it in the perceptual demonstrative case would allow us to explain the phenomenon in a uniform way.


November 6, 2012: reading and discussion of Gareth Evans, "Does Tense Logic Rest upon a Mistake?"


October 23, 2012: Mark Richard, "Reference to rabbits"


There seems to be a quick argument for the conclusion that what we refer to with practically all of our names and nouns is indeterminate.  The major premise is that names like ‘Peter Rabbit’ and predicates like ‘rabbit’ are vague since  there are a host of different collections of molecules such that it is utterly indeterminate which of those collections collects all and only the molecules that are part of Peter Rabbit.  Thus, there are a host of different objects such that it is indeterminate which of them the name ‘Peter Rabbit’ refers to.  Likewise for the noun ‘rabbit’ –there are countless disjoint candidates for its extension.  Likewise for pretty much every name and noun.

The issues here are partially semantic, partially metaphysical.  One way to respond to the argument is concede the major premise but insist that, since vagueness is epistemic, no interesting conclusions about reference follow.  A second way to respond takes the vagueness of 'rabbit' to be substantive –there is no fact of the matter as to what 'rabbit' is true of, and thus the term does not determinately refer to anything.  This doesn't mean that it doesn't refer –'rabbit' refers to rabbits, what else?  Rather, it means that the only serviceable notion of reference is a lightweight, "disquotational" one.  A third way to respond to the argument is to question whether 'rabbit' is vague in a way that supports the argument's conclusion.  The argument, after all, moves pretty quickly from the premise, Peter's physical constitution is a vague matter, to the conclusion that Peter is a vague matter.  Perhaps here it moves too quickly.

So far as I can see, these are the only responses to the argument worth taking seriously. In this talk, I propose to sort through them. I'll conclude that a bunny's being a determinate object doesn't require that it have a perfectly determinate collection of parts, or a perfectly precise location, so reference to rabbits is not threatened by the problem of the many.  On the way there I will have a good deal to say about and against various views of vagueness.


October 16, 2012: Matias Gariazzo"Truth relativism about epistemic modals and the argument from ignorant assessors"


The data on rejection and retraction that has been appealed to for supporting truth relativism about epistemic modals involve (i) a better-informed subject that assesses a claim made by a less informed subject and (ii) a retraction made by the latter based on the assessment made by the former. Richard Dietz (2008) argues that we do not obtain data on rejection and retraction supporting truth relativism once we consider the opposite type of situation where someone assesses a claim made by a better-informed subject. This criticism has been taken to imply that at most epistemic modal claims are asymmetrically assessment sensitive: they are assessment sensitive insofar as the assessor knows more than the speaker. Based on such a construal of the objection MacFarlane (2011, 2012) and Egan & Weatherson (2011) have replied to it. I argue that the Dietz’s objection can be generalized in order to show that the accuracy of epistemic modal claims cannot be taken to be assessment sensitive in any case.


October 2, 2012: discussion with Georges Rey. Readings: Peter Carruthers, The Opacity of Mind; Georges Rey, "We're Not All 'Self-Blind': A Defense of a Modest Introspectionism" and "Postscript".






PAST SEMINARS (2011-2012):


July 3, 2012: Dan Zeman"On How the Operator Argument Could Support Relativism"


One traditional argument for introducing parameters in the circumstances of evaluation has been "the Operator Argument" (Kaplan (1989)). Recently, this type of argument has been taken to support relativism about several types of expressions. In the talk I consider a type of objection to the claim that the Operator Argument could support relativism that comes in two versions, one belonging to Lopez de Sa (2011), the other to Ninan (2010). I explore several routes that the relativist could take in answering the two versions of the objection. The particular answer the relativist will give depends on certain assumptions about semantic content (namely, what I call "semantic value monism" and "semantic value pluralism"). I make it clear what is the form of the counterargument that the relativist needs to provide in each case and then attempt to provide it.


June 26, 2012: Isidora Stojanovic"Metasemantics, Postsemantics, Prepragmatics: Widening the Semantics/Pragmatics Boundary"


The debate over the semantics-pragmatics distinction has been vibrant for a decade or two, but seems to be reaching an impasse. This paper argues that a possible reason for this may be the failure to recognize a distinct realm, which falls neither under semantics nor pragmatics, but may be labeled "postsemantics" – or "prepragmatics". The ultimate goal of the paper is to put forward and defend a new picture of our language architecture, according to which: semantic content is strictly poorer than the lexically encoded content (hence doesn't involve or depend on any contextually determined elements, not even the reference of indexicals and demonstratives); pragmatics, as it is commonly assumed, does not reach into truth-conditions and does not affect truth-value, and its mechanisms require the capacity of representing, and reasoning about, one's beliefs and intentions; and, finally, there is a distinctive postsemantic/prepragmatic level, at which sentences and/or utterances get evaluted for their truth value (but also for other properties, such as their modal status or their assertoric 'content'), and which takes into account various contextual information.


The paper is structured as follows. In section 1, I turn to a topic that has long been of interest to philosophers of language, and has been one of the "stumbling stones" in the discussions of the semantics/pragmatics distinction: the topic of demonstratives. In the mainstream, direct-referentialist view, the semantic contribution of a pronoun (demonstrative or indexical) is some contextually determined object or individual. It is also held that the issue of how the semantic value of a demonstrative is determined (e.g. whether speaker's intentions, pointing gestures, contextual salience etc. play a role in determining it, and to which extent) is an issue that does not pertain to semantics, but to meta-semantics. My aim in sect. 1 is twofold. First, I show that the mainstream view has difficulties in maintaining such an approach to the semantics of demonstratives and, at the same time, drawing a principled line of demarcation between semantics and pragmatics. Secondly, the semantics vs. metasemantics dichotomy, while legitimate from a theoretical point of view, does not square well with any distinctions that psycholinguistics would want to draw among the various processes involved in language processing. In section 2, I turn to yet another topic much discussed in the semantics-pragmatics literature: that of quantifier domain restriction (i.e. the question of which class of individuals is relevant to interpreting a given use of e.g. 'most students'). Drawing on some ideas from my paper "Domain Sensitivity" (Synthese 184: 137-155), I argue that even though domain-restriction affects truth value (and is therefore not merely a matter of pragmatic implicature), it does not reach into any properly semantic level. The upshot of the first two sections is to show that, if there is indeed room for a middle ground between semantic and pragmatic levels or interpretation mechanisms, then the resolution of demonstrative reference and of the domain over which a quantifier ranges will be candidates par excellence to belong there. In section 3, I put forward my own proposal, which, in a nutshell, views semantic mechanisms as intimately linked with stable lexical meaning, but takes pragmatic processes to require a full-fledged capacity to reason about the speaker's mental states, and to deploy general principles akin to Gricean maxims. While the resolution of reference may, in certain cases, require the latter, it typically doesn't; the same holds for domain-resolution. I end the paper by discussing how this "middle ground" level between semantics and pragmatics relates to the idea of "postsemantics" that has seen light in the literature on the relativism/contextualism debate (in particular regarding predicates of personal taste and epistemic modals).


June 12, 2012: Sam Wilkinson, "Recognition and Identification in Capgras Delusion"


I want to highlight the difference between two cognitive tasks or judgments: what I call recognition and identification. These often get conflated due to their unregimented use in ordinary language, but are clearly very different. Once I have characterized these tasks, I apply them to the Capgras delusion, and to delusions of misidentification generally, and draw some important lessons.


April 17, 2012: Felipe Carvalho, "Varieties of Sortal-Dependence"


The sortal-dependence of demonstrative thought is the thesis that our capacity to have demonstrative thoughts about perceived objects depend on our capacity to classify these objects as objects of specific kinds. But from the way it is formulated, it is not yet clear what this thesis amounts to; when we look at the different ways in which it has been discussed in the relevant literature, for example, we can distinguish between at least two different theses: (1) the thesis that our capacity to perceive particular objects as such cannot be explained outside the apparatus of sortal concepts, and (2) the thesis that our capacity to think about particular objects as such cannot be explained outside the apparatus of sortal concepts.

In light of this distinction, one could argue for the sortal dependence of demonstrative thought in two different ways. The first would be to hold that our capacities for demonstrative thought are explainable in terms of our capacities for object perception, but that the latter capacity needs the apparatus of sortal concepts. Another way would be to argue that, no matter what we end up saying about our capacities for object perception, the norms that govern singular thoughts are essentially different from those that govern object perception, which require us to bring, over and above perceptual kinds, the sortal concept a subject thinks of the object as.

In this talk I will examine different arguments that have been proposed in the literature in favor of either (1) or (2) above, and see whether we have good reasons to suppose that some form of sortal-dependence is true of demonstrative thought. First, I will argue that sortalism about object-perception is false, and hence that one could only argue in favor of the sortal-dependence of demonstrative thought by appealing to the distinction between conditions for object perception vs. conditions for demonstrative thought. Then, I will argue that even granting this distinction, as it has been elaborated in the literature, it does not follow that sortal concepts are required to capture it, which leads us to conclude that we have no good reasons to accept the sortal-dependence of demonstrative thought.


February 21, 2012: Joulia Smortchkova, "Arguments for the Rich Content View of Perceptual Experience"


What kinds of properties enter into the experiential contents of perception? Do only low-level properties (such as being blue, square, etc.) enter into perceptual experiences (poor content view) or do high level properties (such as being an agent, a banana, sad, etc.) enter into experiential content as well (rich content view)? I focus on the arguments for the rich content view. I first critically examine the division between low-level properties and high-level properties which it presupposes. I then underline the limitations of a method recently proposed to argue for the rich content view: the phenomenology first method. I suggest one way to improve the method is by checking its predictions against the experimental data provided by psychological research on visual agnosia and perceptual adaptation. My aim in so doing is not simply to imply that the methods used by psychology are the right ones: taken on their own they do not cut finely enough between competing possible interpretations of certain data. Instead, I outline how a combination of approaches might positively impact future research on such issues.


February 7, 2012: Alexandra Arapinis, "Whole-for-Part Metonymy as Classification Exploiting Functional Integrity"


Since the early 80s, metonymy has progressively gained central stage in linguistic investigations. The advent of cognitive linguistics marked a new turn in the study of this trope conceived, not as a deviation from semantic conventions (contra classical rhetorical theories), but as a phenomenon rooted in non-language-specific mechanisms of conceptualization and structuring of the world. Focusing on the particular case of whole-for-part (WP) metonymy, the general stand of this presentation will be to argue for the need to re-inject properly semantic (descriptive and formal) considerations in the study of metonymy. It will be shown that semantic typologies of meronymic relations, as well as recent results in the semantics of partial predicates and colour predicates, can be jointly exploited to attain a better understanding of the mechanism of WP-metonymy. The resulting view will in particular connect such metonymies to mechanisms of reference and property attribution to functional parts for classificatory purposes.


January 24, 2012: Michael Murez, "Mental Files and the Dynamics of Identity"


Proponents of mental files claim that when we think about identities (for example, when we judge that Samuel Clemens is Mark Twain) we "merge" or "link" mental files. I'll present a few arguments against such a view.


January 17, 2012: François Recanati & Anouch Bourmayan, "Transitive Meanings for Intransitive Verbs"


In context, ‘John is eating’ can be used to communicate that John is eating some contextually salient piece of food (e.g. a dangerous mushroom that has been the focus of attention for some time) ; in this case, Recanati (2002) argued, a contextually understood object is supplied for intransitive ‘eat’. In Deirdre Wilson’s example ‘Socrates took the hemlock and drank’, intransitive ‘drink’ takes a definite object through some kind of anaphoric process : what Socrates is said to have drunk is the hemlock mentioned in the first conjunct. The contextually understood object can also be bound, as in ‘John is anorexic, but whenever his father cooks mushrooms, he eats’, which means that John eats the mushrooms cooked by his father, or in ‘when someone like Socrates is offered hemlock, he drinks’, where ‘drinks’ means drinks the hemlock he is offered.

That an object can be contextually supplied for an intransitive verb like ‘eat’ is somewhat paradoxical ; for to say that ‘eat’ is intransitive in those examples is to say that it does not take any object (over and beyond the ‘internal accusative’ of the verb). To eat (intransitive) is to eat something, or to eat some food. How, then, can this literal meaning of intransitive ‘eat’ be pragmatically enriched so as to yield what we may call the transitive meaning eat the dangerous mushroom or eat the mushrooms cooked by his father ?

In this paper, we will provide a situation-theoretic answer. We will show that the tacit reference to (or quantification over) contextually provided objects for intransitive verbs is a byproduct of the tacit reference to (or quantification over) situations that takes place whenever we say something. If the situation at stake is a minimal situation involving only one type of food (e.g. the dangerous mushroom, or the mushrooms cooked by John’s father), then to say that John is eating (i.e. eating something) is equivalent to saying that he is eating the mushroom(s).


December 14, 2011: Katrina Przyjemski, "Anaphoric Theories of Proper Names"


This will be a relatively informal discussion of the prospects for an anaphoric theory of proper names. I discuss two recent accounts that fall into this category, Sam Cumming's Variabilism, and Kit Fine's Semantic Relationism. Both authors aim to account for differences in the contributions of co-referent proper names by appealing to anaphoric relations between name tokens.

I argue that differences in the contributions of co-referent proper name tokens cannot be reduced to differences in anaphoric relations of the kind that Fine and Cumming propose. Moreover, neither author successfully demonstrates the need for a semantic account of proper names as anaphors. I sketch an alternative, presemantic account of anaphoric relations between name tokens that draws on elements of Francois Recanati's mental files model of de jure co-reference. I argue that appealing to pre-semantic relations between proper name tokens provides the referentialist with an attractively conservative way to explain data involving substitution failure.


November 23, 2011: Gregory Bochner, "Towards an Internalist Theory of Direct Reference"


I sketch a causal picture of proper names combining Linguistic Referentialism and Mental Descriptivism. This combination affords a purely Fregean solution to epistemic problems arising from Kripke's widely accepted views. The picture, which is a version of Two-Dimensionalism, requires to regard names as non-mental entities individuated by their meaning/content/reference, and explains cognitive significance problems (the co-reference, no-reference, Twin-Earth, empty, contingent a priori, and Kripke's puzzle cases) by invoking an imperfect identification of external names/words inside the mind. The resulting account, which emphasises the importance of external-objective factors in the (automatic) determination of linguistic content, defends the plausibility of a dramatic gap between language (expression) and thought (apprehension).


October 26, 2011: Dan Zeman, "Temporal Binding in the Event Analysis"


In this paper I investigate one answer to the so-called "argument from binding" for locations consisting in quantification over events instead of quantification over locations. The particular view I will focus on is Cappelen and Hawthorne's "event analysis". After a brief presentation of the view and of how it answers the argument from binding, I provide some examples of temporal binding that show the need to modify the account. I then envisage some ways the account could be modified in order to deal with the examples given, and conclude with a more general discussion of what kind of views could benefit from the event analysis.