Knowledge representation is one way to exploit expertise in a given domain by logical means. But, what kind of knowledge does one acquire from an inference (or inference on a query result over a knowledge base)? Such a question may appear awkward since the answer seems so obvious: from an inference, one simply acquires knowledge. This is undoubtedly the case when only one type of knowledge (for instance, expert knowledge) is involved in an inference. What if several types of knowledge are involved? What type of knowledge can one deduce from a plurality of knowledge types? I claim that reasoning with different knowledge concepts requires a fine-grained representation of knowledge in which every knowledge type finds a singular expression in order to avoid some epistemic equivocity associated with a coarse-grained representation of knowledge. In the first part of the paper, I revisit the Muddy Children Puzzle, which usually serves to illustrate common knowledge in dynamic epistemic logic. I try to show that this problem also shows some sort of epistemic equivocity between concepts of knowledge and, consequently, that the problem calls for some epistemological refinements concerning the representation of the types of knowledge at play in an inference. In the second part, I address this issue from a semantic point of view, and I develop a fragment of epistemic logic capable of providing a solution to the problem of epistemic equivocity.

Christoph Jäger (Erkenntnis 61:187–201, 2004) has argued that Dretske’s (Knowledge and the flow of Information, MIT Press, Cambridge, 1981) information-based account of knowledge is committed to both knowledge and information closure under known entailment. However, in a reply to Jäger, Dretske (Erkenntnis 64:409–413, 2006) defended his view on the basis of a discrepancy between the relation of information and the relation of logical implication. This paper shares Jäger’s criticism that Dretske’s externalist notion of information implies closure, but provides an analysis based on different grounds. By means of a distinction between two perspectives, the mathematical perspective and the epistemological perspective, I present, in the former, a notion of logical implication that is compatible with the notion of information in the mathematical theory of information, and I show how, in the latter, Dretske’s logical reading of the closure principle is incompatible with his information-theoretic epistemological framework.

Ce papier présente un cadre logique capable d'exprimer un contextualisme épistémologique.Le contextualisme épistémologique repose sur la possibilité d'une interprétation indexicale des connaissances de l'opérateur, selon laquelle les conditions de vérité des attributions des connaissances manifeste une variabilité contextuelle d'une telle manière que les connaissances dans un contexte n'implique pas de connaissances dans d'autres contextes.Au moyen d'une notion de contexte épistémique définie sur la base de la notion de contexte développée par McCarthy et Buvac en Intelligence Artificielle, le papier montre comment une interprétation indexicale des connaissances de l'opérateur peut être modélisée formellement à travers un système de déduction naturellequi permet le raisonnement classique parmi des contextes gouvernés par différents concepts de connaissance.ABSTRACT.This paper aims at presenting a logical framework capable of expressing epistemological contextualism.Epistemological contextualism relies upon the possibility of an indexical interpretation of the knowledge operator, according to which the truth conditions of knowledge attributions exhibit a contextual variability in such a way that knowledge in one context does not entail knowledge in every context.By means of a notion of epistemic context defined on the basis of the notion of context developed by McCarthy and Buvač in artificial intelligence, I show how an indexical interpretation of the knowledge operator can be formally modeled through a natural deduction system that enables classical reasoning among contexts governed by different concepts of knowledge.

In this paper, I develop a framework for representing knowledge, that aims at exploiting logically epistemological objects, namely epistemic contexts. Even though this framework is mostly logical and epistemological in character, it takes advantage of many works in artificial intelligence, in particular the ones of McCarthy and Buvač (1997). In Sect. 1, I characterize the notion of epistemic contexts. In Sect. 2, I present a natural deduction system that allows for the introduction and the elimination of knowledge operators. Such a system enables classical reasoning among contexts governed by different concepts of knowledge. Finally, in Sect. 3, I discuss some corollaries of the proposed framework (knowledge transfer, introspection, and closure).

I establish and I explore a parallel between two important analogies in Wittgenstein's remarks on knowledge and certainty, and two characteristic features of Dretske's information-theoretic model of knowledge. My claim is that the analogical opposition (a static/dynamic type of opposition) behind the axis/hinge and the riverbed/flux-of-the-waters analogies finds a counterpart in Dretske's perspective with the opposition between information and conditions of the channel of communication. This comparative analysis puts into light the epistemic functions at stake in Wittgenstein's analogies and in Dretske's framework, as well as two common conclusions that both Wittgenstein and Dretske share about the status and the scope of sceptical doubts.

Este trabajo presenta un marco epistemológico capaz de hacer frente a la oposición entre conocimiento indígena y conocimiento científico, una oposición generalizada en la antropología, especialmente en relación con el problema del desarrollo sostenible. En la primera parte del artículo, proporcionamos un marco contextualista que satisfaga dos restricciones: la neutralidad a priori respecto a formas, o tipos, de conocimiento, y (2) la explicitud de las condiciones con respecto a la posibilidad de transferencia de conocimiento. En la segunda parte, aplicamos el marco para el caso específico del conocimiento algonquino del norte. A través de tal marco, se investiga y hace explícito el estándar epistémico subyacente puesto en juego, mediante la identificación de las condiciones en que creencias particulares son calificadas como conocimiento en el contexto algonquino del norte.

This paper aims at providing additional support to the defense of epistemological contextualism. Epistemological contextualism relies on the possibility of an indexical interpretation of the knowledge operator, according to which the truth of knowledge attributions exhibit a contextual variability in such a way that knowledge in one context does not entail knowledge in every context. By means of a notion of epistemic context defined on the basis of the notion of context developed by McCarthy and Buvač in artificial intelligence, I show how an indexical interpretation of the knowledge operator can be formally modelled and how the formal resources required to disambiguate the knowledge operator can shed new light on epistemological contextualism.

I argue in favor of the principle of deductive closure. I provide a characterization of the notion of epistemic context in terms of presuppositions in order to show how one can preserve deductive closure by limiting its applicability to a given epistemic context. This analysis provides also a redefinition of relevant alternatives from an epistemological point of view, that avoids any appeal to a counterfactual interpretation of relevance. In that perspective, relevance is conceived as a presuppositional quality. In regard to the debate with the skeptic, it is shown that her argument relies upon a context shift that compromises deductive closure because it alters the presuppositions at play. Finally, deductive closure is not only vindicated but is acknowledged as a contextual marker in the sense that a failure of deductive closure indicates a context shift.

In this paper, I show the complementarity of foundationalism and coherentism with respect to any efficient system of beliefs by means of a distinction between two types of proposition drawn from an analogy with an axiomatic system. This distinction is based on the way a given proposition is acknowledged as true, either by declaration (F-proposition) or by preservation (C-proposition). Within such a perspective, i.e., epistemological complementarism, not only can one see how the usual opposition between foundationalism and coherentism is irrelevant, but furthermore one can appreciate the reciprocal relation between these two theories as they refer to two separate epistemological functions involved in the dynamics of constituting and expanding an epistemic system.

The general principle of epistemic closure stipulates that epistemic properties are transmissible through logical means. According to this principle, an epistemic operator, say e, should satisfy any valid scheme of inference, such as: if e(p entails q), then e(p) entails e(q). The principle of epistemic closure under known entailment (ECKE), a particular instance of epistemic closure, has received a good deal of attention since the last thirty years or so. ECKE states that: if one knows that p entails q, and she knows that p, then she knows that q. It is widely accepted that ECKE constitutes an important piece of the skeptical argument, but the acceptance of an unrestricted version of ECKE is still a matter of debate. On the side of the defenders of ECKE, one finds Stine (1976), Brueckner (1985), Vogel (1990), and Feldman (1995). Others proposed a refutation or a limitation of the principle, like Dretske (1970), Nozick (1981), Hales (1995), Williams (1996), and Sosa (1999). As it turns out, the relevant alternatives view (RAV) elaborated by Dretske, which restricts the scope of ECKE, has been discussed extensively and acknowledged as one of the most important contributions. There is nonetheless a major unsolved difficulty pertaining to Dretske-RAV: the notion of relevant alternatives is defined in such a way that it is bounded by counterfactual possibilities. This ontological import leaves open the door to the skeptic. Some have tried to give more precision to this notion, like Stine (1976), who appealed to a Gricean approach to define relevant alternatives in conversational contexts. My proposal is in accordance with the gist of Dretske’s strategy, i.e. to restrict the validity of ECKE, and I claim that in order to escape the difficulties inherent to RAV one has to introduce a more robust notion, the notion of epistemic context. Epistemic contexts are a subclass of propositional contexts. In that perspective, the closure property is expressed in terms of a property of a relation between epistemic contexts. ECKE holds when and only when either the epistemic context of the premisses is the same as the epistemic context of the conclusion, or the epistemic context change between the premisses and the conclusion is permissible. Permissibility of epistemic context change is a function of consistency. By means of this epistemic context approach, I will show that: (1) epistemic contexts are defined by basic propositions (unchallenged justified beliefs), (2) ECKE holds only under very specific constraints, and (3) the skeptical argument involves a non-permissible change of epistemic context and, by the same token, cannot rely upon ECKE.

Dans le cadre de la representation des connaissances au sein des environnements virtuels d'apprentissage (EVA) et du fait de la nature relationnelle et structuree de ces connaissances, leur exploitation necessite de faire appel a des processus d'analyse capables de traiter et de manipuler un grand volume de donnees structurees. Dans ce sens, un modele de representation trouve sa pleine utilite pour la recherche et le traitement automatique d'information guidant le raisonnement, le diagnostic et la prediction d'un systeme tuteur dont la tâche est de fournir des suggestions, des exemples et des explications bien adaptes a chaque apprenant. Nous presentons dans cet article un modele pour representer la connaissance du domaine et celle des usagers utilisant des EVA. Le modele s'inspire des recherches en intelligence artificielle sur la modelisation computationnelle et des theories de la psychologie cognitive qui expliquent l'activite cognitive humaine en termes de sous-systemes de la memoire et de leurs processus. Ces sous-systemes (implementes dans un EVA), parce qu'ils sont similaires a ceux utilises par l'apprenant, facilitent, d'une part, l'identification de la connaissance acquise correcte ou erronee, et suggerent, d'autre part, la planification d'une sequence d'activites pedagogiques pour ameliorer de maniere significative le niveau cognitif de cet apprenant.

In recent years, encouraging results in the field of knowledge representation within virtual learning environments (VLEs) confirm findings by artificial intelligence research in this area: namely, the benefits of integrating knowledge accumulated by research into psychology (i.e. the cognitive mechanisms of human learning) and the positive results yielded by computational modelling theories. This paper introduces a new cognitive and computational approach to knowledge representation inspired by cognitive theories that explain human cognitive activity in terms of memory subsystems and their processes, and whose aim is to suggest formal computational models of knowledge that offer efficient and expressive representation of virtual learning structures. Practical studies serve both to validate the new approach and allow general conclusions to be drawn.

In this paper, I want to characterize Wittgensteins epistemology, namely his contextualism, presented in his notes published under the title On Certainty. This characterization will be made in terms of a comparison with four control points in order to put into light the peculiarities of Wittgensteins epistemological position. The markers I will use for that purpose will give an indication of the level of conceptual commitment toward four important epistemological theses: holism, internalism, fallibilism, and egalitarianism. These markers have been chosen in function of the power of discrimination they provide regarding the two main trends in the past 40 years (or so) of debate in contemporary epistemology, i.e., foundationalism and coherentism. Using these conceptual markers, one can easily distinguish between strict foundationalism, which shows a low level of each markers, and strict coherentism, which shows a high level of each markers.

This paper aims at defending the idea that the possibility of infinite regress in a chain of reasons is not incompatible with the possibility for an epistemic agent to be justified at some given point. Consequently, the infinite regress argument usually developed against the coherence theory of knowledge does not undermine coherentism as an adequate theory of epistemic justification. Furthermore, I put into light the two necessary conditions that any belief system must satisfy in order to remain epistemically operative: a coherentist con- dition, i.e., the logical possibility of infinite regress in a chain of reasons, and a foundationalist condition, i.e., the pragmatic necessity of regress- stoppers.

Résumé: Cet article a pour but de mettre en lumière une tension conceptuelle relative à la théorie du jugement de Kant par le biais d'un examen du rapport entre analyticité et temporalité. Kant affirme que le jugement analytique ne saurait être marqué par le temps, puisque cela impliquerait une forme de synthèse. Or, il est possible d'établir dans une perspective strictement kantienne que la temporalité est impliquée dans le principe d'analyticité, en vertu des notions d'objectivité et de nécessité. Ce lien entre temps et analyticité permet également de montrer une affinité importante entre Kant et J. Hintikka Abstract: This paper aims at putting into light a conceptual tension pertaining to Kant's theory of judgment by means of an examination of the relation between analyticity and temporality. Kant claims that the analytic judgment cannot be affected by time, since that would imply a kind of synthesis. But it is possible to show, within a strict Kantian perspective, that temporality is involved in the principle of analyticity, in virtue of the notions of objectivity and necessity. This connection between time and analyticity enables one also to make explicit an important affinity between Kant and J. Hintikka.