Language research has come to rely heavily on large‐scale, web‐based datasets. These datasets can present significant methodological challenges, requiring researchers to make a number of decisions about how they are collected, represented, and analyzed. These decisions often concern long‐standing challenges in corpus‐based language research, including determining what counts as a word, deciding which words should be analyzed, and matching sets of words across languages. We illustrate these challenges by revisiting “Word lengths are optimized for efficient communication” (Piantadosi, Tily, & Gibson, (...) 2011), which found that word lengths in 11 languages are more strongly correlated with their average predictability (or average information content) than their frequency. Using what we argue to be best practices for large‐scale corpus analyses, we find significantly attenuated support for this result and demonstrate that a stronger relationship obtains between word frequency and length for a majority of the languages in the sample. We consider the implications of the results for language research more broadly and provide several recommendations to researchers regarding best practices. (shrink)

Spoken communication occurs in a “noisy channel” characterized by high levels of environmental noise, variability within and between speakers, and lexical and syntactic ambiguity. Given these properties of the received linguistic input, robust spoken word recognition—and language processing more generally—relies heavily on listeners' prior knowledge to evaluate whether candidate interpretations of that input are more or less likely. Here we compare several broad-coverage probabilistic generative language models in their ability to capture human linguistic expectations. Serial reproduction, an experimental paradigm where (...) spoken utterances are reproduced by successive participants similar to the children's game of “Telephone,” is used to elicit a sample that reflects the linguistic expectations of English-speaking adults. When we evaluate a suite of probabilistic generative language models against the yielded chains of utterances, we find that those models that make use of abstract representations of preceding linguistic context (i.e., phrase structure) best predict the changes made by people in the course of serial reproduction. A logistic regression model predicting which words in an utterance are most likely to be lost or changed in the course of spoken transmission corroborates this result. We interpret these findings in light of research highlighting the interaction of memory-based constraints and representations in language processing. (shrink)

Dans le cadre de notre thèse, nous avons confronté les Lais de Marie de France datant du xiie siècle qui présentent la particularité alors inédite d’avoir pour auteur une femme et le Genji Monogatari de Murasaki Shikibu (une femme également), écrit au xie siècle au Japon, qui fait aussi figure d’exception dans un paysage scriptural majoritairement masculin, a fortiori pour une œuvre d’une telle ampleur. Le rapprochement entre les deux œuvres était avant tout motivé par une thématique de fond, l’imaginaire (...) féminin, déployé dans deux cultures géographiquement et sociologiquement éloignées, cependant immergées dans un milieu élitiste, celui de la Cour. Ce choix s’inscrivait surtout dans la perspective d’une approche mythocritique plutôt que strictement littéraire, dont l’intérêt se situait en deçà du cadre du texte, puisqu’il puisait dans la mythologie fondatrice des deux cultures, source de l’émergence d’un imaginaire féminin à des époques chiches de féminité. Or, cette analyse, que nous avons qualifiée de contrastive, a permis l’ouverture de la réflexion sur d’autres perspectives que seul autorisait un tel rapprochement ; elle a également cherché à distinguer, parmi les motifs communs aux deux textes, ceux qui relevaient d’une universalité (contextuelle) de ceux qui relevaient d’une commune origine eurasiatique. La réflexion était axée sur la notion d’altérité selon une acception très large mais concentrée sur la représentation par l’imaginaire féminin de l’Autre sexe et d’un Ailleurs avant tout textuel et narratif. (shrink)

À l’origine de la philosophie comme des sciences, il y a, selon Aristote, « l’étonnement de ce que les choses sont ce qu’elles sont ». Nul doute qu’Aristote aurait trouvé en Suisse maints sujets d’étonnement. Qu’est-ce qu’une vache ? Qu’est-ce qu’une montagne ? Qu’est-ce que le Röstigraben ? Qu’est-ce qu’une fondue ? Qu’est-ce qu’un trou dans l’emmental ? Qu’est-ce que l’argent ? Qu’est-ce qu’une banque ? Qu’est-ce qu’une confédération ? Qu’est-ce qu’une horloge ? Qui est Roger Federer ? Qu’est-ce qu’est (...) Anton Marty ? Qu’est-ce que le plaisir de manger du chocolat ? -/- Chaque chapitre de cet ouvrage a été écrit par un spécialiste de renommée internationale et défend une solution claire à l’une de ces questions. Le tout constitue une introduction à la fois accessible et plaisante à un domaine de recherche philosophique aujourd’hui en effervescence : la métaphysique. (shrink)

L’objectif de cet article est de clarifier les relations qu’entretiennent trois théories de la justification des croyances, toutes considérées comme des théories internalistes 1 : la conception déontique (le déontologisme), la conception accessibiliste (l’accessibilisme) et la conception mentaliste (le mentalisme). Nous expliquerons qu’en dépit de ce que l’on pourrait penser à première vue l’adoption de l’accessibilisme n’implique pas celle du mentalisme. Dans un second temps, nous montrerons pourquoi on ne peut être un défenseur de la conception déontique de la justification (...) sans être également un partisan de l’accessibilisme.2 L’intérêt de ces démonstrations réside principalement dans le fait qu’elles nous permettent de délier —ou, au contraire d’unir— le sort de ces différentes conceptions de la justification vis- à-vis des objections qui leur sont adressées. Faire la preuve de l’autonomie du mentalisme vis- à-vis de l’accessibilisme, par exemple, c’est établir l’immunité du premier vis-à-vis des objections que l’on peut faire au second. L’examen proprement dit des relations qui rattachent ces trois conceptions internalistes de la justification occupe les sections 4 et 5. Les sections 1, 2 et 3 sont des sections introductives. Elles sont, pour leur part, consacrées à la description de ces conceptions. (shrink)

Ce chapitre discute de la justification des croyances testimoniales, c’est-à-dire de la justification des croyances que nous adoptons en nous appuyant sur le témoignage d’autrui. Plus précisément, la question à laquelle cette contribution s’intéresse est celle des conditions nécessaires et suffisantes de la justification des croyances testimoniales. Il y a deux manières classiques, et soi-disant antagonistes, d’y répondre: la réponse réductionnisme et la réponse non-réductionniste. L’objectif de ce chapitre est d’une part de présenter ces deux réponses, d’autre part, d’expliquer pourquoi (...) ce prétendu antagonisme n’en est pas un. Le plan est le suivant. La première partie de ce chapitre est consacrée à distinguer les croyances testimoniales des croyances perceptuelles (section 1. et 1.1). Dans la seconde partie (section 2 à 2.3), nous présentons la réponse réductionniste et la réponse non-réductionniste ainsi que quelques-unes des objections les plus importantes adressées à chacune d’elles. La troisième partie explique pourquoi le prétendu antagonisme entre le réductionnisme et le non-réductionnisme n’en est pas un en exploitant ce que nous appelons « la thèse du malentendu » (3 à 3.2). (shrink)

Sommes-nous, au moins occasionnellement, responsables de nos croyances ? Une chose est sûre, en pratique, nous considérons souvent que tel est le cas. Mais l’attribution d’une telle responsabilité est problématique dans la mesure où les croyances ne sont pas des états mentaux que nous contrôlons comme nous contrôlons, par exemple, nos actions. Cet article est consacré à préciser, à expliquer et à défendre l’affirmation selon laquelle les croyances ne sont pas des états mentaux que nous pouvons acquérir « à volonté (...) » . Dans un deuxième temps, nous expliquons brièvement pourquoi la vérité de cette affirmation n’empêche pas que nous soyons, au moins occasionnellement, responsables de ce que nous croyons. (shrink)

Cette contribution a deux objectifs principaux. Le premier est de montrer que les intuitions sont caractérisées par ce que j’appellerai « une capacité rationnelle », c’est-à-dire, qu’elles sont susceptibles d’être évaluées quant à leur rationalité ou leur irrationalité. Le second objectif de cet article est d’étayer l’hypothèse selon laquelle les intuitions seraient des états affectifs proches des émotions — et non pas des états doxastiques ou des expériences perceptuelles —, en montrant qu’une telle conception affective des intuitions est seule capable (...) de rendre compte de la spécificité phénoménologique, la modularité, et la capacité rationnelle des intuitions. This article has two purposes. First, it wants to show that intuitions possess, what I shall call, “a rational capacity” : they are mental states susceptible to be assessed as rational or irrational. Second, this contribution aims at providing some evidence supporting the view that intuitions are affective states similar to emotions rather than doxastic or perceptual states. Such an affective account of the intuitions is the only one able to capture the specific phenomenology, the modularity and the rational capacity of the intuitions. (shrink)

Le présent article revient sur l’expression en vieil islandais vargr í véum (« loup dans les sanctuaires »). En comparant différents récits de tels loups, notamment le mythe de l’enchaînement de Fenrir, l’article montre que plutôt qu’une vieille catégorie préchrétienne dénotant le sacrilège, l’expression doit être comprise comme un moyen discursif servant à marginaliser son objet. Elle devait ainsi s’avérer utile pour des auteurs médiévaux et donc chrétiens au moment où ceux-ci pensaient des problèmes tels que les hiérarchies sociales ou (...) encore la centralisation de l’état. (shrink)

COVID-19 vaccine refusal seems like a paradigm case of irrationality. Vaccines are supposed to be the best way to get us out of the COVID-19 pandemic. And yet many people believe that they should not be vaccinated even though they are dissatisfied with the current situation. In this paper, we analyze COVID-19 vaccine refusal with the tools of contemporary philosophical theories of responsibility and rationality. The main outcome of this analysis is that many vaccine-refusers are responsible for the belief that (...) they should not be vaccinated and epistemically rational in holding it. This is an important result because it provides insights into the legitimacy of certain public health policies. In particular, this result shows that a public health policy that would abandon the project of convincing certain vaccine-refusers with reasons – e.g., by simply making vaccination compulsory – is prima facie illegitimate. (shrink)

In Working Virtue: Virtue Ethics and Contemporary Moral Problems, leading figures in the fields of virtue ethics and ethics come together to present the first ...

Mimesis as Make-Believe is important reading for everyone interested in the workings of representational art.

Our thesis is that there is a notion of justification, corresponding to the active exercise of a competence in order to attain truth, whose value is explained neither by reliabilism, nor by the usual versions of credit theory.

The standard view of ignorance is that it consists in the mere lack of knowledge or true belief. Duncan Pritchard has recently argued, against the standard view, that ignorance is the lack of knowledge/true belief that is due to an improper inquiry. I shall call, Pritchard’s alternative account the Normative Account. The purpose of this article is to strengthen the Normative Account by providing an independent vargument supporting it.

It is commonly accepted – not only in the philosophical literature but also in daily life – that ignorance is a failure of some sort. As a result, a desideratum of any ontological account of ignorance is that it must be able to explain why there is something wrong with being ignorant of a true proposition. This article shows two things. First, two influential accounts of ignorance – the Knowledge Account and the True Belief Account – do not satisfy this (...) requirement. They fail to provide a satisfying normative account of the badness of ignorance. Second, this article suggests an alternative explanation of what makes ignorance a bad cognitive state. In a nutshell, ignorance is bad because it is the manifestation of a vice, namely, of what Cassam calls “epistemic insouciance”. (shrink)

We are occasionally responsible for our beliefs. But is this doxastic responsibility analogous to any non-attitudinal form of responsibility? What I shall call the consequential conception of doxastic responsibility holds that the kind of responsibility that we have for our beliefs is indeed analogous to the kind of responsibility that we have for the consequences of our actions. This article does two things, both with the aim of defending this somewhat unsophisticated but intuitive view of doxastic responsibility. First, it emphasizes (...) the advantage of preserving, as the consequential conception does, the analogy with the non-attitudinal realm, i.e., with the realm of actions and their consequences. Second, this article regiments the most important objections to the consequential conception and answers them. The upshot is that there are no serious drawbacks to the consequential conception. There is, therefore, no reason not to favour it over accounts of doxastic responsibility that do not preserve the analogy with the non-attitudinal realm. (shrink)

Situations where it is not obvious which of two incompatible actions we ought to perform are commonplace. As has frequently been noted in the contemporary literature, a similar issue seems to arise in the field of beliefs. Cases of doxastic divergence are cases in which the subject seems subject to two divergent oughts to believe: an epistemic and a practical ought to believe. This article supports the moderate pragmatist view according to which subjects ought, all things considered, to hold the (...) practically right belief in, at least, some cases of doxastic divergence. Unlike many defences of pragmatism, this paper does not aim to overcome exclusivism (briefly, the view that only epistemic, but not practical, considerations have an influence on what a subject ought to believe). Another major challenge that pragmatism faces is to show that the epistemic and the practical ought to believe are comparable. This article makes a case for their comparability. (shrink)

We frequently praise or blame people for what they believe or fail to believe as a result of their having investigated some matter thoroughly, or, in the case of blame, for having failed to investigate it, or for carelessly or insufficiently investigating. for instance, physicists who, after years of toil, uncover some unknown fact about our universe are praised for what they come to know. sometimes, in contrast, we blame and may even despise our friends for being ignorant of certain (...) ecological facts as a result of their refusing to countenance the evidence. the purpose of this paper is to explore what underlies the legitimacy of this practice—the praise or blame of people for what they know or fail to know as a result of investigation or otherwise—namely, the ability to exercise control over one’s doxastic states, and, in particular, as i will argue, one’s ability to exercise indirect generic control over one’s doxastic states. (shrink)

The general aim of this article is to consider whether various affective phenomena – feelings like the feeling of knowing, of familiarity, of certainty, etc., but also phenomena like curiosity, interest, surprise and trust – which have been labelled “epistemic emotions” in fact constitute a unified kind, i.e., the kind of the so-called “epistemic emotions”. Obviously, for an affective phenomenon to belong to the kind of the epistemic emotions, it has to meet two conditions: it has to qualify, first, as (...) an emotion, and, second, as an epistemic one. The paper is structured accordingly. The first part is devoted to the question whether the aforementioned affective phenomena really are emotions, while the second part bears on their hypothetical common epistemicity. (shrink)

In the course of our daily lives we make lots of evaluations of actions. We think that driving above the speed limit is dangerous, that giving up one’s bus seat to the elderly is polite, that stirring eggs with a plastic spoon is neither good nor bad. We understand too that we may be praised or blamed for actions performed on the basis of these evaluations. The same is true in the case of certain beliefs. Sometimes we blame people for (...) what they believe falsely or irrationally. On occasion, we praise them for their intellectual discoveries. The goal of the present study is to describe the foundations which allow for these kinds of judgments. If you think that people are not always forgivable for what they believe, you will find in this book the resources to defend this thought. (shrink)

Many philosophers have defended the view that we are subject to the following evidential ought: “One ought to believe in accordance with one's evidence.” Although they agree on this, a more fundamental question keeps dividing them: from where does the evidential ought derive its normative force? The instrinsicalist answer to this question is sometimes described as the claim that "there is a brute epistemic value in believing in accordance with one's evidence" (Cowie, 2014, 4005). But what does this really mean? (...) Intrinsicalists' negative claim is that the normative source of the evidential ought has nothing to do with the fulfilment of our epistemic purposes. But what is their positive claim exactly? This paper aims at improving our understanding of intrinsicalism. Its conclusion is that there is only one possible way of being an intrinsicalist. Intrinsicalists need to be constitutivists. (shrink)

We are occasionally responsible for our beliefs. But is this doxastic responsibility analogous to any non‐attitudinal form of responsibility? What I shall call the consequential conception of doxastic responsibility holds that the kind of responsibility that we have for our beliefs is indeed analogous to the kind of responsibility that we have for the consequences of our actions. This article does two things, both with the aim of defending this somewhat unsophisticated but intuitive view of doxastic responsibility. First, it emphasizes (...) the advantage of preserving, as the consequential conception does, the analogy with the non‐attitudinal realm, i.e., with the realm of actions and their consequences. Second, this article regiments the most important objections to the consequential conception and answers them. The upshot is that there are no serious drawbacks to the consequential conception. There is, therefore, no reason not to favour it over accounts of doxastic responsibility that do not preserve the analogy with the non‐attitudinal realm. (shrink)

In several papers (2013, 2014, 2015) Conor McHugh defends the influential view that doxastic responsibility, viz. our responsibility for our beliefs, is grounded in a specific form of reasons-responsiveness. The main purpose of this paper is to show that a subject’s belief can be responsive to reasons in this specific way without the subject being responsible for her belief. While this specific form of reasons-responsiveness might be necessary, it is not sufficient for doxastic responsibility.

It was proposed that the data from Monte Carlo statistical investigations be subjected to analysis of variance methods rather than the conventional techniques of tabling, graphing, and inspecting the data. Two examples in which analysis of variance methods were applied to published Monte Carlo studies were presented. It was suggested that balanced factorial designs should be used whenever possible in Monte Carlo studies so that analysis of variance methods would be directly applicable. Finally, three advantages of analysis of variance methods (...) over conventional techniques were described and analysis of variance methods were contrasted with a possible alternative analytical method, multiple regression analysis. (shrink)

A satisfying theory of knowledge has to explain why knowledge seems to be better than mere true belief. In this paper, I try to show that the best reliabilist explanation (ERA+) is still not able to solve this problem. According to an already elaborated answer (ERA), it is better to possess knowledge that p because this makes likely that one’s future belief of a similar kind will also be true. I begin with a metaphysical comment which gives birth to ERA (...) +, a better formulation of ERA. Then, I raise two objections against ERA+. The first objection shows that the truth of the reliabilist answer requires the conception of a specific theory of instrumental value. In the second objection, I present an example in order to show that ERA+ actually fails to explain why it is better to possess knowledge than a mere true belief. (shrink)