What is it to know more? By what metric should the quantity of one's knowledge be measured? I start by examining and arguing against a very natural approach to the measure of knowledge, one on which how much is a matter of how many. I then turn to the quasi-spatial notion of counterfactual distance and show how a model that appeals to distance avoids the problems that plague appeals to cardinality. But such a model faces fatal problems of its own. (...) Reflection on what the distance model gets right and where it goes wrong motivates a third approach, which appeals not to cardinality, nor to counterfactual distance, but to similarity. I close the paper by advocating this model and briefly discussing some of its significance for epistemic normativity. In particular, I argue that the 'trivial truths' objection to the view that truth is the goal of inquiry rests on an unstated, but false, assumption about the measure of knowledge, and suggest that a similarity model preserves truth as the aim of belief in an intuitively satisfying way. (shrink)

This paper advocates explicitness about the type of entity to be considered as content- bearing in connectionist systems; it makes a positive proposal about how vehicles of content should be individuated; and it deploys that proposal to argue in favour of representation in connectionist systems. The proposal is that the vehicles of content in some connectionist systems are clusters in the state space of a hidden layer. Attributing content to such vehicles is required to vindicate the standard explanation for some (...) classificatory networks’ ability to generalise to novel samples their correct classification of the samples on which they were trained. (shrink)

Sometimes your concept and mine have exactly the same content. When this is so, it is comparatively easy for me to understand what you say when you deploy your concept, for us to disagree, agree, and so on. But what if your concept and mine do not have exactly the same content? This question has occupied a number of philosophers, including Paul Churchland, Jerry Fodor, and Ernie Lepore. This paper develops a novel and rigorous measure of concept similarity, Proportion, such (...) that concepts with different contents but sufficiently high Proportion scores will also conduce to understanding, agreement, and disagreement. (shrink)

Suppose we have an idea of what counts as communication, more precisely as a communicative event. Then we have the further task of dividing communicative events into successful and unsuccessful. Part of this task is to find a basis for this evaluation, i.e. appropriate properties of speaker and hearer. It is argued that success should be evaluated in terms of a relation between thought contents of speaker and hearer. This view is labelled ‘classical’, since it is justifiably attributable to both (...) Locke and Frege (section 2). Most of the paper is devoted to discussing competing contemporary views, such as behaviorist/pragmatist views (section 4), requirements of knowledge (section 5), of reliability (section 6) and that success should be defined in terms of public language semantics rather than in terms of thought content (section 7). (shrink)

The predominance of machine learning based techniques in cognitive neuroscience raises a host of philosophical and methodological concerns. Given the messiness of neural activity, modellers must make choices about how to structure their raw data to make inferences about encoded representations. This leads to a set of standard methodological assumptions about when abstraction is appropriate in neuroscientific practice. Yet, when made uncritically these choices threaten to bias conclusions about phenomena drawn from data. Contact between the practices of multivariate pattern analysis (...) and philosophy of science can help to illuminate the conditions under which we can use artificial neural networks to better understand neural mechanisms. This paper considers a specific technique for MVPA called representational similarity analysis. I develop a theoretically-informed account of RSA that draws on early connectionist research and work on idealization in the philosophy of science. By bringing a philosophical account of cognitive modelling in conversation with RSA, this paper clarifies the practices of neuroscientists and provides a generalizable framework for using artificial neural networks to study neural mechanisms in the brain. (shrink)

Inferential Role Semantics is often criticized for being incompatible with the platitude that words of different speakers can mean the same thing. While many assume that this platitude can be accommodated by understanding sameness of meaning in terms of similarity of meaning, no worked out proposal has ever been produced for Inferential Role Semantics. I rectify this important omission by giving a detailed structural account of meaning similarity in terms of graph theory. I go on to argue that this account (...) has a number of attractive features, prominent among them that it makes sameness of meaning probabilistically determine co-reference. (shrink)

A similarity space is a hyperspace in which the dimensions represent various dimensions on which objects may differ. The similarity space theory of concepts is the thesis that concepts are regions of similarity spaces that are somehow realized in the brain. Proponents of such a theory of concepts include Paul Churchland and Peter Gärdenfors. This paper argues that the similarity space theory of concepts is mistaken because regions of similarity spaces cannot serve as the components of judgments. It emerges that (...) although similarity spaces cannot model concepts, they may model a kind of nonconceptual representation. (shrink)

Jerry Fodor and Ernest Lepore have launched a powerful attack against Paul Churchland's connectionist theory of semantics—aka State Space Semantics. In one part of their overall attack, they exploit the potentially orthogonal histories of different individuals to introduce what they labeled ‘the collateral information problem’. Aarre Laakso and Gary Cottrell have recently put forward a mathematical technique for measuring conceptual similarity across neural networks. Churchland uses Laakso and Cottrell's tecnique to defend State Space Semantics. In this paper I shall highlight (...) a potential problem for Laakso and Cottrell's technique, and for Churchland's subsequent defence of connectionist semantics that has been ignored in the connectionist literature. I shall argue that a connectionist sympathiser of Churchland could not make use of Laakso and Cottrell's neurosimulations to address Fodor and Lepore's collateral information challenge. (shrink)

The article investigates characteristics of legal concepts as found in academic articles, focusing upon the knowledge base of legal experts. It is a cognitively oriented study of one of the semiotic basics of communication for academic legal purposes. The purpose is to study the structure of knowledge elements connected to the concept of “Criminal liability of corporations” from US law in and across individual experts in order to look for individual differences and similarities. The central concern is to investigate the (...) conditions for the observable efficiency of semiosis in academic discourse. In a first basic section I discuss aspects relevant for a cognitively oriented study of academic discourse. The empirical part of the article consists of an analysis of text passages from two articles in American law journals. The results of the study support the assumption that high efficiency and precision of semiosis is due rather to the use of specific cognitive processing skills than to total identity of cognitive structures across individual experts. (shrink)

In the first section of this paper I define meaning holism (MH) and compare it to related theses. In the second section I review several theories of meaning that incorporate MH as a feature, and in the third section I discuss the question whether and how MH is consistent with the assignment of semantic values to linguistic expressions. Finally, in the fourth section I present the main objections raised against MH in the literature and the answers given to them.

Proponents of meaning holism often invoke notions of meaning similarity and semantic spatiality in order to counter accusations that holism renders language unstable and chaotic. However, talk of such notions often falls short of being explicit and formal. In this paper I present an algebraically couched theory of inferential similarity and spatiality, motivated by measurement theory, and I apply it to the discussion of meaning holism. I argue that the proposed theory offers new and improved conceptual resources for facing the (...) challenges raised against the thesis. (shrink)

John is currently thinking that the sun is bright. Consider his occurrent belief or judgement that the sun is bright. Its content is that the sun is bright. This is a truth- evaluable content (which shall be our main concern) because it is capable of being true or false. In virtue of what natural, scientifically accessible facts does John’s judgement have this content? To give the correct answer to that question, and to explain why John’s judgement and other contentful mental (...) states have the contents they do in virtue of such facts, would be to naturalize mental content. (shrink)

ABSTRACT: A traditional objection to inferentialism states that not all inferences can be meaning-constitutive and therefore inferentialism has to comprise an analytic-synthetic distinction. As a response, Peregrin argues that meaning is a matter of inferential rules and only the subset of all the valid inferences for which there is a widely shared corrective behaviour corresponds to rules and so determines meaning. Unfortunately, Peregrin does not discuss what counts as “widely shared”. In the paper, I argue for an empirical plausibility of (...) Peregrin’s proposal. The aim of the paper is to show that we can find examples of meaning-constitutive linguistic action, which sustain Peregrin’s response. The idea is supported by examples of meaning modulation. If Peregrin is right, then we should be able to find specific meaning modulations in which a new meaning is publicly available and modulated in such a way that it has a potential to be widely shared. I believe that binding modulations – a specific type of meaning modulations – satisfy this condition. (shrink)

Most people find it obvious that concepts like APPLE, DOG, WATER, CACTUS, SWIM, CHIRP, FURRY, and SMOOTH are acquired from perceptual experiences along with some kind of inferential procedure. Models of how these concepts are inferentially acquired, however, force the acquired concepts to be representationally complex, built from, and composed by, the more primitive representations. Since at least the time of Plato, philosophers and psychologists have struggled to find complex sets of representations that have the same meanings, definitionally or probabilistically, (...) as these concepts. For example, to think about the property-kind being gold is not the same as to think about the complex property-kind being yellowish & shiny & malleable.... I call this Fodor's Challenge: Find an acquisition process that is genuinely inferential and yields a concept that genuinely is one of these lexical concepts. Rather than continue the pursuit of a complex representation that has the same meaning as our concept GOLD. I offer a model on which many lexical concepts are acquired from perception and inference, without being built up from, and composed by, the representations involved. The model, Baptizing Meanings for Concepts, is inspired in part by Saul Kripke's baptism process for assigning meanings to linguistic terms. Many lexical concepts, according to the BMC, are acquired by inferring the presence a new property-kind, picking out the property-kind in terms of those perceptible features, and then assigning a simple mental term, a concept, for that property-kind. (shrink)