It is widely held in contemporary philosophy of mind that states with underived representational content are ipso facto psychological states. This view—the Content View—underlies a number of interesting philosophical projects, such as the attempt to pick out a psychological level of explanation, to demarcate genuinely psychological from non-psychological states, and to limn the class of states with phenomenal character. The most detailed and influential theories of underived representation in philosophy are the tracking theories developed by Fodor, Dretske, Millikan and others. (...) Tracking theorists initially hoped to ‘naturalize’ underived representation by showing that although it is distinctively psychological it is not irreducibly so, yet they ended up developing theories of representation that by their own lights don’t pick out a distinctively psychological phenomenon at all. Burge sets out to develop a theory of underived representation that does pick out a distinctively psychological phenomenon. His theory promises to vindicate the Content View and the various philosophical projects that depend on it. In this paper I argue that Burge’s theory dementalizes representation for the same reason tracking theories do: These theories hold that representations are states with underived accuracy conditions, yet such states are found in all sorts of mindless systems, like plants. (shrink)

There seem to be things we ought not to believe and others we are permitted to believe. Belief is treated as a normative phenomenon both in everyday and academic discourse. At the same time, normativity can be seen as a threat to a naturalistic understanding of the world. Whilst naturalistic claims are of descriptive nature, norms are prescriptive. It is usually held that they cannot be reduced to statements of fact. This problem is also pertinent to the normativity of belief. (...) How is such a phenomenon to be understood within a naturalist framework? -/- Sullivan-Bissett provides a naturalistic account of epistemic normativity in which she explains epistemic normativity in terms of biological functions of belief-producing mechanisms. Importantly, her account is error-theoretic: She argues that we mistake doxastic strategies, which are at best normative in a hypothetical sense, to be categorical epistemic norms. -/- In continuation of Sullivan-Bissett’s account, I draw attention to one of the belief-producing mechanisms which is responsible for bringing about these mistaken beliefs. I claim that normative conformity – a social-learning mechanisms – brings about our beliefs in the existence and categorical validity of epistemic norms. Keeping with the evolutionary perspective employed by Sullivan-Bissett, I answer questions about the function and phylogeny of our mistaken beliefs in epistemic normativity by means of normative conformity and in line with Sullivan-Bissett’s account. (shrink)

The central hypothesis of this book, Delusions and Beliefs: A Philosophical Inquiry (Routledge, 2019), is that delusions are malfunctional beliefs (Chapter 1); they belong to the category of belief (Chapter 2) but, unlike mundane false or irrational beliefs, they fail to perform some functions of belief (Chapter 3). More precisely, delusions directly or indirectly involve some malfunctioning cognitive mechanisms, which is empirically supported by the two-factor account of delusion formation (Chapter 4).

Delusional beliefs are typically pathological. Being pathological is clearly distinguished from being false or being irrational. Anna might falsely believe that his husband is having an affair but it might just be a simple mistake. Again, Sam might irrationally believe, without good evidence, that he is smarter than his colleagues, but it might just be a healthy self-deceptive belief. On the other hand, when a patient with brain damage caused by a car accident believes that his father was replaced by (...) an imposter or another patient with schizophrenia believes that “The Organization” painted the shops on a street in red and green to convey a message, these beliefs are not merely false or irrational. They are pathological. What makes delusions pathological? This paper explores the negative features because of which delusional beliefs are pathological. First, I critically examine the proposals according to which delusional beliefs are pathological because of (1) their strangeness, (2) their extreme irrationality, (3) their resistance to folk psychological explanations or (4) impaired responsibility-grounding capacities of people with them. I present some counterexamples as well as theoretical problems for these proposals. Then, I argue, following Wakefield’s harmful dysfunction analysis of disorder, that delusional beliefs are pathological because they involve some sorts of harmful malfunctions. In other words, they have a significant negative impact on wellbeing (=harmful) and, in addition, some psychological mechanisms, directly or indirectly related to them, fail to perform the jobs for which they were selected in the past (=malfunctioning). An objection to the proposal is that delusional beliefs might not involve any malfunctions. For example, they might be playing psychological defence functions properly. Another objection is that a harmful malfunction is not sufficient for something to be pathological. For example, false beliefs might involve some malfunctions according to teleosemantics, a popular naturalist account of mental content, but harmful false beliefs do not have to be pathological. I examine those objections in detail and show that they should be rejected after all. (shrink)

“Teleosemantics and Pushmi-Pullyu Representations” (call it “TP-PR,” this journal 2014 79.3, 545–566) argues that core teleosemantics, particularly as defined in Millikan (Language, thought and other biological categories, MIT Press, Cambridge, 1984, J Philos 86(6):281–297, 1989, White queen psychology and other essays for Alice, MIT Press, Cambridge, 1993, Philosophical perspectives, Ridgeview Publishing, Alascadero, 1996, Varieties of meaning, MIT Press, Cambridge, 2004–2008), seems to imply that all descriptive representations are at the same time directive and that directives are at the same time (...) descriptive, hence that all representations are pushmi-pullyu representations. A pushmi-pullyu representation is at once indicative and imperative, telling both what the case is and what to do about it. The usefulness at all of the notion of a pushmi-pullyu representation is then questioned. I have recently found several new papers, dissertations, discussions, in which TP-PR is treated as definitive. However, the conclusions of TP-PR rest on misunderstandings of my position, so although the date is late, clarification seems still to be in order. I will indicate a few significant corrections that need to be made to TP-PR’s description of my position (originally called “biosemantics”) and explain central features of biosemantics that are omitted in TP-PR’s description of basic teleosemantics and which are incompatible with the main argument and conclusions of TP-PR. I will try to clarify and explain the value of the notion of a pushmi-pullyu representation. (shrink)

Significant variations in the way objects appear across different viewing conditions pose a challenge to the view that they have some true, determinate color. This view would seem to require that we break the symmetry between multiple appearances in favor of a single variant. A wide range of philosophical and non-philosophical writers have held that the symmetry can be broken by appealing to daylight viewing conditions—that the appearances of objects in daylight have a stronger, and perhaps unique, claim to reveal (...) their true colors. In this note we argue that, whatever else its merits, this appeal to daylight is not a satisfactory answer to the problem posed by perceptual variation. (shrink)

Reliable misrepresentation is getting things wrong in the same way all the time. In Mendelovici 2013, I argue that tracking theories of mental representation cannot allow for certain kinds of reliable misrepresentation, and that this is a problem for those views. Artiga 2013 defends teleosemantics from this argument. He agrees with Mendelovici 2013 that teleosemantics cannot account for clean cases of reliable misrepresentation, but argues that this is not a problem for the views. This paper clarifies and improves the argument (...) in Mendelovici 2013 and response to Artiga's arguments. Tracking theories, teleosemantics included, really do need to allow for clean cases of reliable misrepresentation. (shrink)

It is a live possibility that certain of our experiences reliably misrepresent the world around us. I argue that tracking theories of mental representation have difficulty allowing for this possibility, and that this is a major consideration against them.

I construct a tempting anti-physicalist argument, which sharpens an explanatory gap argument suggested by David Chalmers and Frank Jackson. The argument relies crucially on the premise that there is a deep epistemic asymmetry (which may be identified with the explanatory gap) between phenomenal truths and ordinary macroscopic truths. Many physicalists reject the argument by rejecting this premise. I argue that even if this premise is true, the anti-physicalist conclusion should be rejected, and I provide a detailed, physicalist-friendly explanation of the (...) relevant premise. Along the way, I sketch an account of a priori conceptual knowledge that is compatible with naturalistic accounts of intentionality. I conclude by noting that the resulting view is a version of the popular phenomenal concept strategy that avoids a potentially worrying dilemma facing earlier incarnations of this strategy. (shrink)

According to Schellenberg’s capacitism, perception is constituted by employing perceptual capacities to discriminate and single out particulars, including objects, events and property instances. To say that perception is so constituted, for her, is to say that perceptual states have the content, phenomenal character, and evidential force they do in virtue of the fact that they are yielded by employing perceptual capacities.1 1.

Natural selection comes in degrees. Some biological traits are subjected to stronger selective force than others, selection on particular traits waxes and wanes over time, and some groups can only undergo an attenuated kind of selective process. This has downstream consequences for any notions that are standardly treated as binary but depend on natural selection. For instance, the proper function of a biological structure can be defined as what caused that structure to be retained by natural selection in the past. (...) We usually think of proper functions in binary terms: storing bile is a function of the gall bladder, but making stones is not. However, if functions arise through natural selection, and natural selection comes in degrees, then a binary approach to proper functions is in tension with the biological facts. In order to resolve this tension, we need to revise our standard accounts of proper function. In particular, we may have to seriously consider the possibility that functions themselves come in degrees, in spite of the ramifications this will have for the way we speak about functions and related concepts such as dysfunction, disease, and teleosemantic content. (shrink)

In the first part of the paper, I present a framework for the description and evaluation of teleosemantic theories of intentionality, and use it to argue that several different objections to these theories (the various indeterminacy and adequacy problems) are, in a certain precise sense, manifestations of the same underlying issue. I then use the framework to show that Millikan's biosemantics, her own recent declarations to the contrary notwithtanding, presents indeterminacy. In the second part, I develop a novel teleosemantic proposal (...) which makes progress in the treatment of this family of problems. I describe a procedure to derive a (unique) homeostatic property cluster [HPC] from facts having to do with the properties that a certain indicator relied on, in the events leading to its fixation in a certain population. This HPC is the one that should figure in the content attribution to the indicator in question. (shrink)

Hutto and Myin claim that teleosemantics cannot account for mental content. In their view, teleosemantics accounts for a poorer kind of relation between cognitive states and the world but lacks the theoretical tools to account for a richer kind. We show that their objection imposes two criteria on theories of content: a truth-evaluable criterion and an intensionality criterion. For the objection to go through, teleosemantics must be subject to both these criteria and must fail to satisfy them. We argue that (...) teleosemantics meets the truth-evaluable criterion and is not required to meet the intensionality criterion. We conclude that Hutto and Myin’s objection fails. (shrink)

The free energy principle is notoriously difficult to understand. In this paper, we relate the principle to a framework that philosophers of biology are familiar with: Ruth Millikan’s teleosemantics. We argue that: systems that minimise free energy are systems with a proper function; and Karl Friston’s notion of implicit modelling can be understood in terms of Millikan’s notion of mapping relations. Our analysis reveals some surprising formal similarities between the two frameworks, and suggests interesting lines of future research. We hope (...) this will aid further philosophical evaluation of the free energy principle. (shrink)

One holistic thesis about symbols is that a symbol cannot exist singly, but only as apart of a symbol system. There is also the plausible view that symbol systems emerge gradually in an individual, in a group, and in a species. The problem is that symbol holism makes it hard to see how a symbol system can emerge gradually, at least if we are considering the emergence of a first symbol system. The only way it seems possible is if being (...) a symbol can be a matter of degree, which is initially problematic. This article explains how being a cognitive symbol can be a matter of degree after all. The contrary intuition arises from the way a process of interpretation forces an all-or-nothing character on symbols, leaving room for underlying material to realize symbols to different degrees in a way that Daniel Dennett’s work can help illuminate. Holism applies to symbols as interpreted, while gradualism applies to how the underlying material realizes symbols.Selon une thèse holistique sur les symboles, un symbole ne peut exister isolément mais doit faire partie d’un systéme symbolique. Une opinion, elle aussi plausible, veut que les systèmes symboliques émergent graduellement chez un individu, un groupe ou une espèce. Le problème c’est qu’on voit mal, si le holisme des systèmes symboliques tient, comment un système symbolique peut émerger graduellement, du moins pour la première fois. Ce n’est possible, semble-t-il, que si être un symbole est affaire de degré, thèse au départ problématique. Cet article explique comment être un symbole cognitif peut après tout être affaire de degré. L’intuition contraire vient de ce que le processus d’interprétation nous force au tout ou rien, ce qui laisse un jeu dans la façon dont le matériel sous-jacent réalise les symboles à des degrés divers. Les travaux de Daniel Dennett sont à cet égard éclairants. Le holisme vaut pour les symboles tels qu’ils sont interprétés, tandis que le gradualisme vaut pour la façon dont le matériel sous-jacent réalise les symboles. (shrink)

In recent years, the ‘mechanistic view’ has developed as a popular alternative to the ‘semantic view’ concerning the identity of physical computation. However, semanticists have provided powerful arguments that suggest the mechanistic view fails to deliver essential distinctions between paradigmatic computational operations. This article reviews responses on behalf of the mechanist and uses this opportunity to propose a type of pluralism about computational identity. This pluralism contends that there are multiple ‘levels’ of properties and relations pertaining to computation that can (...) inform different kinds of individuation. As such, for the pluralist, there are multiple legitimate ways of classifying computation, depending on the nature of the system in question, and one’s own epistemic priorities. (shrink)

Shannon information is commonly assumed to be the wrong way in which to conceive of information in most biological contexts. Since the theory deals only in correlations between systems, the argument goes, it can apply to any and all causal interactions that affect a biological outcome. Since informational language is generally confined to only certain kinds of biological process, such as gene expression and hormone signalling, Shannon information is thought to be unable to account for this restriction. It is often (...) concluded that a richer, teleosemantic sense of information is needed. I argue against this view, and show that a coherent and sufficiently restrictive theory of biological information can be constructed with Shannon information at its core. This can be done by paying due attention some crucial distinctions: between information quantity and its fitness value, and between carrying information and having the function of doing so. From this I construct an account of how informational functions arise, and show that the “subject matter” of these functions can easily be seen as the natural information dealt with by Shannon’s theory. (shrink)

Whether “information” exists in biology, and in what sense, has been a topic of much recent discussion. I explore Shannon, Dretskean, and teleosemantic theories, and analyze whether or not they are able to give a successful naturalistic account of information—specifically accounts of meaning and error—in biological systems. I argue that the Shannon and Dretskean theories are unable to account for either, but that the teleosemantic theory is able to account for meaning. However, I argue that it is unable to account (...) for error. Thus I conclude that while talk of informational meaning is justifiable within a naturalistic framework, talk of informational error is not, and must be used in a metaphorical sense only. (shrink)

Reductive representationalist theories of consciousness are yet to produce a satisfying account of pain’s affective component, the part that makes it painful. The paramount problem here is that that there seems to be no suitable candidate for what affective experience represents. This article suggests that affective experience represents the Darwinian fitness effects of events. I argue that, because of affective experience’s close association with motivation, natural selection will work to bring affect into covariance with the average fitness effects of types (...) of event, and that this covariance makes fitness effects a promising candidate for what affect represents. I also argue that this account is to be preferred to Cutter and Tye’s recent proposal that affect represents harmfulness, and answer an objection that Aydede and Fulkerson recently offered against representational accounts of affect. (shrink)

Hutto and Myin famously argue that basic minds are not contentful and content exists only as far as it is scaffolded with social and linguistic practices. This view, however, rests on a troublesome distinction between basic and scaffolded minds. Since Hutto and Myin have to account for language purely in terms of joint action guidance, there is no reason why simpler communication systems, such as cellular signaling pathways, should not give rise to scaffolded content as well. This conclusion remains valid (...) even if one rejects the view of language as mediated through public symbols and embraces global antirepresentationalism. Content evolves spontaneously in complex regulatory systems, such as human, animal, and cellular communication. (shrink)

Ruth Millikan’s teleological theory of mental content is complex and often misunderstood. This paper motivates and clarifies some of the complexities of the theory, and shows that paying careful attention to its details yields answers to a number of common objections to teleological theories, in particular, the problem of novel mental states, the problem of functionally false beliefs, and problems about indeterminacy or multiplicity of function.

In this paper, we consider how certain longstanding philosophical questions about mental representation may be answered on the assumption that cognitive and perceptual systems implement hierarchical generative models, such as those discussed within the prediction error minimization framework. We build on existing treatments of representation via structural resemblance, such as those in Gładziejewski :559–582, 2016) and Gładziejewski and Miłkowski, to argue for a representationalist interpretation of the PEM framework. We further motivate the proposed approach to content by arguing that it (...) is consistent with approaches implicit in theories of unsupervised learning in neural networks. In the course of this discussion, we argue that the structural representation proposal, properly understood, has more in common with functional-role than with causal/informational or teleosemantic theories. In the remainder of the paper, we describe the PEM framework for approximate Bayesian inference in some detail, and discuss how structural representations might arise within the proposed Bayesian hierarchies. After explicating the notion of variational inference, we define a subjectively accessible measure of misrepresentation for hierarchical Bayesian networks by appeal to the Kullbach–Leibler divergence between posterior generative and approximate recognition densities, and discuss a related measure of objective misrepresentation in terms of correspondence with the facts. (shrink)

In the paper I defend the idea of metametasemantics against the arguments recently presented by Ori Simchen. Simchen attacks the view, according to which metametasemantics incorporating all possible metasemantic accounts is necessary to protect the metasemantic theories from the notorious problem of inscrutability of reference. Simchen claims that if metametasemantics is allowed it ‘absorbs’ metasemantic theories to the extent that it diminishes their explanatory value. Furthermore, in this way Simchen sets up two main metasemantic paradigms i.e. productivism and interpretationism as (...) the rival theories inevitably excluding each other. I endeavour to undermine Simchen’s point by demonstrating that his argumentation mixes up deflationary reading of the predicate ‘is true’ with its substantial reading. Consequently, I demonstrate that accepting metametasemantics does not diminish explanatory value of various metasemantic theories and thus that there is no good reason to forbid metametasemantics. I also argue that even if we ignore the above-mentioned confusion in Simchen’s reasoning, his arguments still fail when considering various problems with the notion of diminishment of explanatory value and because the analogy that his arguments are based on is fairly weak. Eventually, I conclude that metametasemantics does not pose any danger to metasemantics and that it provides a solid ground for developing a theory that benefits from both productivism and interpretationism. (shrink)

Sceptics of informational terminology argue that by attributing content to signals, we fail to address nonhuman animal communication on its own terms. Primarily, we ignore that communication is sender driven: i.e. driven by the intrinsic physical properties of signals, themselves the result of selection pressures acting on signals to influence receivers in ways beneficial for senders. In contrast, information proponents argue that this ignores the degree to which communication is, in fact, receiver driven. The latter argue that an exclusive focus (...) on the intrinsic mechanical properties of signals cannot explain why receivers respond as they do. This is because receivers are not prisoners of sender influence. They possess response flexibility, and so we can only explain why receivers respond to signals as they do by positing that receivers ‘derive information’ from signals. I argue that, while basically true, this response flexibility can take one of two forms depending on the causal-explanatory role of information in understanding the response of the receiver: diachronic, on the one hand; and synchronic, on the other. In species with diachronic response flexibility only, information is derived by receivers from signals in a minimal sense. In such cases, information is an ultimate explanatory construct: one underpinned by historical facts at the population level. Alternatively, in species with synchronic response flexibility, information is derived by receivers from signals in a richer sense. Here, information is a proximate explanatory construct: one underpinned by cognitive-mechanistic facts at the level of the individual organism. Without recognising the different ways information can be derived from signals, and the different causal-explanatory roles information can play in understanding alternate kinds of receiver flexibility, proponents of information leave themselves open to the charge of anthropomorphising some signalling systems. (shrink)

Within theoretical and empirical enquiries, many different meanings associated with consciousness have appeared, leaving the term itself quite vague. This makes formulating an abstract and unifying version of the concept of consciousness – the main aim of this article –into an urgent theoretical imperative. It is argued that consciousness, characterized as dually accessible (cognized from the inside and the outside), hierarchically referential (semantically ordered), bodily determined (embedded in the working structures of an organism or conscious system), and useful in action (...) (pragmatically functional), is a graded rather than an all-or-none phenomenon. A gradational approach, however, despite its explanatory advantages, can lead to some counterintuitive consequences and theoretical problems. In most such conceptions consciousness is extended globally (attached to primitive organisms or artificial systems), but also locally (connected to certain lower-level neuronal and bodily processes). For example, according to information integration theory (as introduced recently by Tononi and Koch, 2014), even such simple artificial systems as photodiodes possess miniscule amounts of consciousness. The major challenge for this article, then, is to establish reasonable, empirically justified constraints on how extended the range of a graded consciousness could be. It is argued that conscious systems are limited globally by the ability to individuate information (where individuated information is understood as evolutionarily embedded, socially altered, and private), whereas local limitations should be determined on the basis of a hypothesis about the action-oriented nature of the processes that select states of consciousness. Using these constraints, an abstract concept of consciousness is arrived at, hopefully contributing to a more unified state of play within consciousness studies itself. (shrink)

Reflections on Meaning refines Paul Horwich’s use theory of meaning. Horwich holds that the meaning of a word is constituted by the nonsemantic property that best explains a certain law. For a given word, the law to be explained governs that word’s use by specifying the “acceptance conditions” of a privileged class of sentences containing the word (26). Horwich devotes considerable energy to details in Reflections on Meaning and focuses on especially pressing problems for his use theory of meaning. As (...) a result the book’s topics run the gamut, and the connections between its chapters are not always strong. Rather than try to provide a synoptic overview, I’ll discuss three areas where it seems further clarification and detail could be fruitful: the distinction between semantic and nonsemantic properties, context sensitivity, and compositionality. Horwich thinks ours is a “fundamentally non-semantic world” (27), making it crucial that meaning be explained in nonsemantic terms. In particular, he insists that we “exclude from the analyzing-properties [of word meaning] anything that would itself require analysis in terms of meaning”: we can’t appeal to reference, belief, or intention, for example (37). But Horwich does not object to “accounts of meaning in psychological terms,” and his own theory relies heavily on a psychological, nonsemantic relation that Horwich calls “acceptance” (37). It’s difficult to see a substantive difference between this technical notion and belief. Acceptance, for Horwich, is “the psychological (but nonsemantic) relation to a sentence that is manifested in our relying on it as a premise in theoretical and practical inference” (40–41). Belief, on the other hand, is a relation with these properties except that it is semantic. Horwich’s other characterizations of acceptance don’t sharpen the distinction very much: “S accepts a sentence just in case that sentence, or its mental correlate, is in S’s belief box” (41); “believing a given proposition is nothing more than accepting some sentence that expresses it” (61).. (shrink)

There are two main accounts of what it is for external objects to be presented in visual experience. According to particularism, particular objects are built into the representational contents of experiences. Existentialism is a quite different view. According to existentialism, the representational contents of perceptual experiences are general rather than particular, in the sense that the contents can be fully captured by existentially quantified statements. The present paper is a defense of existentialism. It argues that existentialism is much better equipped (...) than particularism to explain the contents of hallucinations, and it defends existentialism against three objections. (shrink)

There are two main accounts of what it is for external objects to be presented in visual experience. According to particularism, particular objects are built into the representational contents of experiences. Existentialism is a quite different view. According to existentialism, the representational contents of perceptual experiences are general rather than particular, in the sense that the contents can be fully captured by existentially quantified statements. The present paper is a defense of existentialism. It argues that existentialism is much better equipped (...) than particularism to explain the contents of hallucinations, and it defends existentialism against three objections. (shrink)

Mirror neurons and systems are often appealed to as mechanisms enabling mindreading, i.e., understanding other people’s mental states. Such neural mirroring processes are often treated as instances of mental simulation rather than folk psychological theorizing. I will call into question this assumed connection between mirroring and simulation, arguing that mirroring does not necessarily constitute mental simulation as specified by the simulation theory of mindreading. I begin by more precisely characterizing “mirroring” (Sect. 2) and “simulation” (Sect. 3). Mirroring results in a (...) neural process in an observer that resembles a neural process of the same type in the observed agent. Although simulation is often characterized in terms of resemblance (Goldman, Simulating minds: The philosophy, psychology, and neuroscience of mindreading, 2006), I argue that simulation requires more than mere interpersonal mental resemblance: A simulation must have the purpose or function of resembling its target (Sect. 3.1). Given that mirroring processes are generated automatically, I focus on what is required for a simulation to possess the function of resembling its target. In Sect. 3.2 I argue that this resemblance function, at least in the case of simulation-based mindreading, requires that a simulation serve as a representation or stand-in of what it resembles. With this revised account of simulation in hand, in Sect. 4 I show that the mirroring processes do not necessarily possess the representational function required of simulation. To do so I describe an account of goal attribution involving a motor mirroring process that should not be characterized as interpersonal mental simulation. I end in Sect. 5 by defending the conceptual distinction between mirroring and simulation, and discussing the implications of this argument for the kind of neuroscientific evidence required by simulation theory. (shrink)

Dennett's intended rapprochement between physical realism and intentional relativism fails because it is premised upon conflicting arguments governing the status of design. Indeed, Dennett's remarks on design serve to highlight tensions buried deep within his theory. For inasmuch as Dennett succeeds in objectifying attributions of design, attributions of intentionality readily follow suit, leading to a form of intentional realism. But inasmuch as Dennett is successful in relativizing attributions of design, scientific realism at large is subject to renewed anti-realistic criticism. Dennettian-inspired (...) considerations of adaptationism substantiate the former move towards intentional realism, while considerations of the relativity of artifactual design encourage the latter move towards physical relativism. The ambivalence intrinsic to Dennett's ``mild realism'' can be viewed as a function of these two conflicting positions on design, for Dennett can no more avoid objectifying intentionality when he is realistic about design than he can avoid relativizing physical causality when relativistic about design. (shrink)

Information about the environment is captured in human biological systems on a variety of interacting levels – in distributions of genes, linguistic particulars, concepts, methods, theories, preferences, and overt behaviors. I investigate some of the basic principles which govern such a hierarchy by constructing a comparatively simple three-level selection model of bee foraging preferences and behaviors. The information-theoretic notion of ''''mutual information'''' is employed as a measure of efficiency in tracking a changing environment, and its appropriateness in epistemological applications is (...) discussed at some length. In particular, information accumulated in mid-level preference distributions exhibits suggestive properties for the purposes of naturalistic epistemology. It also appears that the novelty of scientific objects and representations and the rapidity of scientific change relative to genetic change need present no obstacle to the use of such models in explanations of scientific progress and the reliability of scientific judgement. (shrink)

Representations seem to play a major role in many neuroscientific explanations. Philosophers have long attempted to properly define what it means for a neural state to be a representation of a specific content. Teleosemantic theories of content which characterize representations, in part, by appealing to a historical notion of function, are often regarded as our best path towards an account of neural representations. This paper points to the anti-representationalist consequences of these accounts. I argue that assuming such teleosemantic views will (...) deprive representations of their explanatory role in computational explanations. My argument rests on the claim that many explanations in cognitive neuroscience are entirely independent of any historical considerations. In making this claim, I will also offer an adapted version of the famous Swampperson thought experiment, which is better suited to discussions of subpersonal neural representations. (shrink)

In this article, I develop a loosely Wittgensteinian conception of what it takes for a being, including an AI system, to understand language, and I suggest that current state of the art systems are closer to fulfilling these requirements than one might think. Developing and defending this claim has both empirical and conceptual aspects. The conceptual aspects concern the criteria that are reasonably applied when judging whether some being understands language; the empirical aspects concern the question whether a given being (...) fulfills these criteria. On the conceptual side, the article builds on Glock’s concept of intelligence, Taylor’s conception of intrinsic rightness as well as Wittgenstein’s rule-following considerations. On the empirical side, it is argued that current transformer-based NNLP models, such as BERT and GPT-3 come close to fulfilling these criteria. (shrink)

Is temporal representation constitutively necessary for perception? Tyler Burge (2010) argues that it is, in part because perception requires a form of memory sufficiently sophisticated as to require temporal representation. I critically discuss Burge’s argument, maintaining that it does not succeed. I conclude by reflecting on the consequences for the origins of temporal representation.

The etiological approach to ‘proper functions’ in biology can be strengthened by relating it to Robert Cummins' general treatment of function ascription. The proper functions of a biological trait are the functions it is assigned in a Cummins-style functional explanation of the fitness of ancestors. These functions figure in selective explanations of the trait. It is also argued that some recent etiological theories include inaccurate accounts of selective explanation in biology. Finally, a generalization of the notion of selective explanation allows (...) an analysis of the proper functions of human artifacts. (shrink)

Epistemologists and philosophers of mind both ask questions about belief. Epistemologists ask normative questions about belief—which beliefs ought we to have? Philosophers of mind ask metaphysical questions about belief—what are beliefs, and what does it take to have them? While these issues might seem independent of one another, there is potential for an interesting sort of conflict: the epistemologist might think we ought to have beliefs that, according to the philosopher of mind, it is impossible to have. This essay argues (...) that this conflict does arise and that it creates problems for traditional skeptical views in epistemology. In particular, it argues that on certain popular views about the nature of belief, it is impossible to adopt the near-global agnosticism recommended by the skeptical epistemologist. On other plausible views, it is possible only in special circumstances, and this limitation undermines skeptical epistemological claims. The only views about the nature of belief on which there are no metaphysical hurdles to adopting the agnosticism recommended by the skeptic are views that face powerful objections—objections that are completely independent of antiskeptical epistemological considerations. (shrink)

Objects are central in visual, auditory, and tactual perception. But what counts as a perceptual object? I address this question via a structural unity schema, which specifies how a collection of parts must be arranged to compose an object for perception. On the theory I propose, perceptual objects are composed of parts that participate in causally sustained regularities. I argue that this theory falls out of a compelling account of the function of object perception, and illustrate its applications to multisensory (...) perception. I also argue that the account avoids problems faced by standard views of visual and auditory objects. (shrink)

Realism about cognitive or semantic phenomenology, the view that certain conscious states are intrinsically such as to ground thought or understanding, is increasingly being taken seriously in analytic philosophy. The principle aim of this paper is to argue that it is extremely difficult to be a physicalist about cognitive phenomenology. The general trend in later 20th century/early 21st century philosophy of mind has been to account for the content of thought in terms of facts outside the head of the thinker (...) at the time of thought, e.g. in terms of causal relations between thinker and world, or in terms of the natural purposes for which mental representations have developed. However, on the assumption that consciousness is constitutively realised by what is going on inside the head of a thinker at the time of experience, the content of cognitive phenomenology cannot be accounted for in this way. Furthermore, any internalist account of content is particularly susceptible to Kripkensteinian rule following worries. It seems that if someone knew all the physical facts about what is going on in my head at the time I was having a given experience with cognitive phenomenology, they would not thereby know whether that state had ‘straight’ rather than ‘quus-like’ content, e.g. whether the experience was intrinsically such as the ground the thought that two plus two equals four or intrinsically such as to ground the thought that two quus two equals four. The project of naturalising consciousness is much harder for realists about cognitive phenomenology. (shrink)

Some emotional representations seem to be unreliable. For instance, we are often afraid when there is no danger present. If emotions such as fear are so unreliable, what function do they have in our representational system? This is a problem for representationalist theories of emotion. I will argue that seemingly unreliable emotional representations are reliable after all. While many mental states strike an optimal balance between minimizing inaccurate representations and maximizing accurate representations, some emotional representations only aim at maximizing accuracy. (...) They detect important phenomena such as danger based on little evidence, which will lead to a lot of false alarms. When it matters, however, these emotional representations will detect danger and other important phenomena. Often, one is better scared (and wrong) than sorry. (shrink)

It is well known that informational theories of representation have trouble accounting for error. Informational semantics is a family of theories attempting a naturalistic, unashamedly reductive explanation of the semantic and intentional properties of thought and language. Most simply, the informational approach explains truth-conditional content in terms of causal, nomic, or simply regular correlation between a representation and a state of affairs. The central work is Dretske (1981), and the theory was largely developed at the University of Wisconsin by Fred (...) Dretske, Dennis Stampe, and Berent Enc. Recently, informational semantics has roamed far beyond its Wisconsin home, and built a sizeable collection of followers. Converts include Jerry Fodor (1987), Robert Stalnaker (1984) and, less faithfully, Paul and Patricia Churchland (1983) and Hartry Field (1986). But for some years informational semantics has been hounded by a problem with error – the classic presentation is Fodor (1984) – and no other problem has hounded the theory so persistently. (shrink)

This paper examines the relationship between a family of concepts involving reliable correlation, and a family of concepts involving adaptation and biological function, as these concepts are used in the naturalistic semantic theory of Dretske's "Explaining Behavior." I argue that Dretske's attempt to marry correlation and function to produce representation fails, though aspects of his failure point the way forward to a better theory.

The original publication can be found at www.springerlink.com.

Perceptual Dogmatism holds that if it perceptually seems to S that p, then S has immediate prima facie justification for the belief that p. Various philosophers have made the notion of a perceptual seeming more precise by distinguishing perceptual seemings from both sensations and beliefs to accommodate a) the epistemic difference between perceptual judgments of novices and experts, and, b) the problem of the speckled hen. Using somewhat different terminology, perceptual seemings are supposed to be high-level percepts instead of low-level (...) sensations. I argue that although it is right that perceptual seemings should not be identified with sensations, they should also not be identified with percepts. There is no strong reason to assume that sensations and percepts exist as separate conscious states, and it appears therefore best to identify perceptual seemings simply with perceptual experiences interpreted as entities incorporating aspects from both sensations and percepts. However, even with this plausible identification in hand, the speckled hen will remain problematic for PD. (shrink)