This paper investigates the view that digital hypercomputing is a good reason for rejection or re-interpretation of the Church-Turing thesis. After suggestion that such re-interpretation is historically problematic and often involves attack on a straw man (the ‘maximality thesis’), it discusses proposals for digital hypercomputing with Zeno-machines , i.e. computing machines that compute an infinite number of computing steps in finite time, thus performing supertasks. It argues that effective computing with Zeno-machines falls into a dilemma: either they are specified such (...) that they do not have output states, or they are specified such that they do have output states, but involve contradiction. Repairs though non-effective methods or special rules for semi-decidable problems are sought, but not found. The paper concludes that hypercomputing supertasks are impossible in the actual world and thus no reason for rejection of the Church-Turing thesis in its traditional interpretation. (shrink)

Ford’s Helen Keller Was Never in a Chinese Room claims that my argument in How Helen Keller Used Syntactic Semantics to Escape from a Chinese Room fails because Searle and I use the terms ‘syntax’ and ‘semantics’ differently, hence are at cross purposes. Ford has misunderstood me; this reply clarifies my theory.

Delusional beliefs have sometimes been considered as rational inferences from abnormal experiences. We explore this idea in more detail, making the following points. Firstly, the abnormalities of cognition which initially prompt the entertaining of a delusional belief are not always conscious and since we prefer to restrict the term “experience” to consciousness we refer to “abnormal data” rather than “abnormal experience”. Secondly, we argue that in relation to many delusions (we consider eight) one can clearly identify what the abnormal cognitive (...) data are which prompted the delusion and what the neuropsychological impairment is which is responsible for the occurrence of these data; but one can equally clearly point to cases where this impairments is present but delusion is not. So the impairment is not sufficient for delusion to occur. A second cognitive impairment, one which impairs the ability to evaluate beliefs, must also be present. Thirdly (and this is the main thrust of our chapter) we consider in detail what the nature of the inference is that leads from the abnormal data to the belief. This is not deductive inference and it is not inference by enumerative induction; it is abductive inference. We offer a Bayesian account of abductive inference and apply it to the explanation of delusional belief. (shrink)

Searle’s Chinese Room Argument (CRA) has been the object of great interest in the philosophy of mind, artificial intelligence and cognitive science since its initial presentation in ‘Minds, Brains and Programs’ in 1980. It is by no means an overstatement to assert that it has been a main focus of attention for philosophers and computer scientists of many stripes. It is then especially interesting to note that relatively little has been said about the detailed logic of the argument, whatever significance (...) Searle intended CRA to have. The problem with the CRA is that it involves a very strong modal claim, the truth of which is both unproved and highly questionable. So it will be argued here that the CRA does not prove what it was intended to prove. (shrink)

Embodied Cognition is an approach to cognition that departs from traditional cognitive science in its reluctance to conceive of cognition as computational and in its emphasis on the significance of an organism's body in how and what the organism thinks. Three lines of embodied cognition research are described and some thoughts on the future of embodied cognition offered.

Intuitively it has seemed to many that our concepts conscious state and conscious creature are sharp rather than vague, that they can have no borderline cases. On the other hand, many who take conscious states to be identical to, or realized by, complex physical states are committed to the vagueness of those concepts. In the paper I argue that conscious state and conscious creature are sharp by presenting four necessary conditions for conceiving borderline cases in general, and showing that some (...) of those conditions cannot be met with conscious state. I conclude that conscious state is sharp, and the conclusion is then extended to conscious creature. The paper ends with a brief discussion of some implications. (shrink)

The paper is concerned with John Searle’s famous Chinese room argument. Despite being objected to by some, Searle’s Chinese room argument appears very appealing. This is because Searle’s argument is based on an intuition about the mind that ‘we’ all seem to share. Ironically, however, Chinese philosophers don’t seem to share this same intuition. The paper begins by first analysing Searle’s Chinee room argument. It then introduces what can be seen as the (implicit) Chinese view of the mind. Lastly, it (...) demonstrates a conceptual difference between Chinese and Western philosophy with respect to the notion of mind. Thus, it is shown that one must carefully attend to the presuppositions underlying Chinese philosophising in interpreting Chinese philosophers. (shrink)

This paper presents the main topics, arguments, and positions in the philosophy of AI at present (excluding ethics). Apart from the basic concepts of intelligence and computation, the main topics of ar-tificial cognition are perception, action, meaning, rational choice, free will, consciousness, and normativity. Through a better understanding of these topics, the philosophy of AI contributes to our understand-ing of the nature, prospects, and value of AI. Furthermore, these topics can be understood more deeply through the discussion of AI; so (...) we suggest that “AI Philosophy” provides a new method for philosophy. (shrink)

Computational complexity is a discipline of computer science and mathematics which classifies computational problems depending on their inherent difficulty, i.e. categorizes algorithms according to their performance, and relates these classes to each other. P problems are a class of computational problems that can be solved in polynomial time using a deterministic Turing machine while solutions to NP problems can be verified in polynomial time, but we still do not know whether they can be solved in polynomial time as well. A (...) solution for the so-called NP-complete problems will also be a solution for any other such problems. Its artificial-intelligence analogue is the class of AI-complete problems, for which a complete mathematical formalization still does not exist. In this chapter we will focus on analysing computational classes to better understand possible formalizations of AI-complete problems, and to see whether a universal algorithm, such as a Turing test, could exist for all AI-complete problems. In order to better observe how modern computer science tries to deal with computational complexity issues, we present several different deep-learning strategies involving optimization methods to see that the inability to exactly solve a problem from a higher order computational class does not mean there is not a satisfactory solution using state-of-the-art machine-learning techniques. Such methods are compared to philosophical issues and psychological research regarding human abilities of solving analogous NP-complete problems, to fortify the claim that we do not need to have an exact and correct way of solving AI-complete problems to nevertheless possibly achieve the notion of strong AI. (shrink)

The artificial intelligence models with machine learning that exhibit the best predictive accuracy, and therefore, the most powerful ones, are, paradoxically, those with the most opaque black-box architectures. At the same time, the unstoppable computerization of advanced industrial societies demands the use of these machines in a growing number of domains. The conjunction of both phenomena gives rise to a control problem on AI that in this paper we analyze by dividing the issue into two. First, we carry out an (...) epistemological examination of the AI’s opacity in light of the latest techniques to remedy it. And second, we evaluate the rationality of delegating tasks in opaque agents. (shrink)

This paper explores early Australasian philosophy in some detail. Two approaches have dominated Western philosophy in Australia: idealism and materialism. Idealism was prevalent between the 1880s and the 1930s, but dissipated thereafter. Idealism in Australia often reflected Kantian themes, but it also reflected the revival of interest in Hegel through the work of ‘absolute idealists’ such as T. H. Green, F. H. Bradley, and Henry Jones. A number of the early New Zealand philosophers were also educated in the idealist tradition (...) and were influential in their communities, but produced relatively little. In Australia, materialism gained prominence through the work of John Anderson, who arrived in Australia in 1927, and continues to be influential. John Anderson had been a student of Henry Jones, who might therefore be said to have influenced both main strands of Australian philosophical thought. (shrink)

This paper links the conjecture that the physical world is a virtual reality to the findings of modern physics. What is usually the subject of science fiction is here proposed as a scientific theory open to empirical evaluation. We know from physics how the world behaves, and from computing how information behaves, so whether the physical world arises from ongoing information processing is a question science can evaluate. A prima facie case for the virtual reality conjecture is presented. If a (...) photon is a pixel on a multi-dimensional grid that gives rise to space, the speed of light could reflect its refresh rate. If mass, charge and energy all arise from processing, the many conservation laws of physics could reduce to a single law of dynamic information conservation. If the universe is a virtual reality, then its big bang creation could be simply when the system was booted up. Deriving core physics from information processing could reconcile relativity and quantum theory, with the former how processing creates the space-time operating system and the latter how it creates energy and matter applications. (shrink)

We try to show that there is no difference in principle between communicating a piece of information to a human and to a machine. The argumentation depends on the following theses: Communicating is transfer of information; information has propositional form; propositional form can be modelled as categorization; categorisation can be modelled in a machine; a suitably equipped machine can grasp propositional content designed for human communication. What I suggest is that the discussion should focus on the truth and precise meaning (...) of these statements. However, in case these statements are true it follows that: For any act of communication that successfully transfers a piece of information to a human, that act could also transfer that piece of information to a machine. (shrink)

Recently, many philosophers have been inclined to ascribe mentality to animals on the main grounds that they possess certain complex computational abilities. In this paper I contend that this view is misleading, since it wrongly assumes that those computational abilities demand a psychological explanation. On the contrary, they can be just characterised from a computational level of explanation, which picks up a domain of computation and information processing that is common to many computing systems but is autonomous from the domain (...) of psychology. Thus, I propose that it is possible to conceive insects and other animals as mere computing agents, without having any commitment to ascribe mentality to them. I conclude by sketching a proposal about how to draw the line between mere computing and genuine mentality. (shrink)

Many activities in Cognitive Science involve complex computer models and simulations of both theoretical and real entities. Artificial Intelligence and the study of artificial neural nets in particular, are seen as major contributors in the quest for understanding the human mind. Computational models serve as objects of experimentation, and results from these virtual experiments are tacitly included in the framework of empirical science. Simulations of cognitive functions, like learning to speak, or discovering syntactical structures in language, are the basis for (...) many claims about human capacities in language acquisition. This raises the question whether results obtained from experiments that are essentially performed on data structures are equivalent to results from "real" experiments. This paper examines some design methodologies for models of cognitive functions using artificial neural nets. The process of conducting the cognitive simulations is largely a projection of theories, or even unsubstantiated conjectures, onto simulated neural structures and an interpretation of the experimental results in terms of the human brain. The problem with this process is that results from virtual experiments are taken to refer unambiguously to the human brain; and the more the language of human cognitive function is deployed in both theory construction and (virtual) experimental interpretation, the more convincing the impression. Additionally, the complexity of the methodologies, principles, and visualization techniques, in the implementation of the computational model, masks the lack of actual similarities between model and real world phenomena. Some computational models involving artificial neural nets have had some success, even commercially, but there are indications that the results from virtual experiments have little value in explaining cognitive functions. The problem seems to be in relating computational, or mathematical, entities to real world objects, like neurons and brains. I argue that the role of Artificial Intelligence as a contributor to the knowledge base of Cognitive Science is diminished as a consequence. (shrink)

Discussions about the achievements and limitations of the various approaches to the development of intelligent systems can have an essential impact on empirically based research, and with that also on the future development of computer technologies. However, such discussions are often based on vague concepts and assumptions. In this context, we claim that the proposed `three-world ontology'' offers the most appropriate conceptual framework in which the basic problems concerned with cognition and computation can be suitably expressed and discussed, although the (...) solutions of some of these problems seem to lie beyond the horizon of our current understanding. We stress the necessity to differentiate between authentic and functional cognitive abilities; although computation is not a plausible way towards authentic intelligence, we claim that computational systems do offer virtually unlimited possibilities to replicate and surpass human cognitive abilities on the functional level. (shrink)

We describe an emerging field, that of nonclassical computability and nonclassical computing machinery. According to the nonclassicist, the set of well-defined computations is not exhausted by the computations that can be carried out by a Turing machine. We provide an overview of the field and a philosophical defence of its foundations.

According to the dominant computational approach in cognitive science, cognitive agents are digital computers; according to the alternative approach, they are dynamical systems. This target article attempts to articulate and support the dynamical hypothesis. The dynamical hypothesis has two major components: the nature hypothesis (cognitive agents are dynamical systems) and the knowledge hypothesis (cognitive agents can be understood dynamically). A wide range of objections to this hypothesis can be rebutted. The conclusion is that cognitive systems may well be dynamical systems, (...) and only sustained empirical research in cognitive science will determine the extent to which that is true. (shrink)

This paper examines an insoluble Cartesian problem for classical AI, namely, how linguistic understanding involves knowledge and awareness of u’s meaning, a cognitive process that is irreducible to algorithms. As analyzed, Descartes’ view about reason and intelligence has paradoxically encouraged certain classical AI researchers to suppose that linguistic understanding suffices for machine intelligence. Several advocates of the Turing Test, for example, assume that linguistic understanding only comprises computational processes which can be recursively decomposed into algorithmic mechanisms. Against this background, in (...) the first section, I explain Descartes’ view about language and mind. To show that Turing bites the bullet with his imitation game, in the second section I analyze this method to assess intelligence. Then, in the third section, I elaborate on Schank and Abelsons’ Script Applier Mechanism (SAM, hereby), which supposedly casts doubt on Descartes’ denial that machines can think. Finally, in the fourth section, I explore a challenge that any algorithmic decomposition of linguistic understanding faces. This challenge, I argue, is the core of the Cartesian problem: knowledge and awareness of meaning require a first-person viewpoint which is irreducible to the decomposition of algorithmic mechanisms. (shrink)

This paper examines the ethical pitfalls and challenges that non-ethicists, such as researchers and programmers in the fields of computer science, artificial intelligence and robotics, face when building moral machines. Whether ethics is “computable” depends on how programmers understand ethics in the first place and on the adequacy of their understanding of the ethical problems and methodological challenges in these fields. Researchers and programmers face at least two types of problems due to their general lack of ethical knowledge or expertise. (...) The first type is so-called rookie mistakes, which could be addressed by providing these people with the necessary ethical knowledge. The second, more difficult methodological issue concerns areas of peer disagreement in ethics, where no easy solutions are currently available. This paper examines several existing approaches to highlight the ethical pitfalls and challenges involved. Familiarity with these and similar problems can help programmers to avoid pitfalls and build better moral machines. The paper concludes that ethical decisions regarding moral robots should be based on avoiding what is immoral in combination with a pluralistic ethical method of solving moral problems, rather than relying on a particular ethical approach, so as to avoid a normative bias. (shrink)

I survey recent work on mental causation. The discussion is conducted under the twin presumptions that mental states, including especially what subjects believe and desire, causally explain what subjects do, and that the physical sciences can in principle give a complete explanation for each and every bodily movement. I start with sceptical discussions of various views that hold that, in some strong sense, the causal explanations offered by psychology are autonomous with respect to those offered by the physical sciences. I (...) then proceed to views that see the problem of mental causation as that of identifying where in the physical story about us and our world lie the parts that in effect tell us abut mental causation - the kind of position that is pretty much standard in the cognitive science community - and consider issues raised by various forms of functionalism and externalism. The general thrust of my discussion is sympathetic to the story about mental causation suggested by those type-type versions of the mind-brain identity theory that allow for the possiblity of multiple realisability. I include a brief discussion of how a map-system account of belief, by contrast with a language of thought one, should understand explanations of behaviour in terms of what a subject believes. (shrink)

The defensibility of the extended mind thesis (EMT) is often thought to hinge on the possibility of extended selves. I argue that the self cannot extend and consider the ramifications of this finding, especially for EMT. After an overview of EMT and the supposed cruciality of the extended self to the defensibility of the former thesis, I outline several lines of argument in support of the possibility of extended selves. Each line of argument appeals to a different account of diachronic (...) personal identity. I argue that no such argument for extended selves succeeds, as no account of diachronic personal identity is both plausible and supports the view that the self can extend. Next, I consider three objections that, if successful, would undercut the preceding argument that the self cannot extend. I conclude by reflecting on the implications of the conclusion that the self cannot extend, including the prospects for EMT. (shrink)

Mind attribution may be divided into the subcategories of attribution of agency, associated with moral agency, and attribution of experience and emotion, associated with moral concern and moral patiency (Gray et al. Science 315(5812):619, 2007; Gray et al. Proceedings of the National Academy of Sciences of the United States of America 108(2):477–479, 2011b; Robbins and Jack Philosophical Studies 127(1):59–85, 2006). In this paper I attend to social context and the different psychological needs influencing the different types of mind attribution. A (...) need for social connection drives the attribution of experiences and emotions. The individual’s capacity to regulate emotional reactions is crucial for empathic concern. I further relate differences of mind attribution, attention, and emotional processing, to two different modes of functioning: In an interactive state of mind (comparable to the “second-person perspective” described by Schilbach et al. Behavioral and Brain Sciences, in press), social and emotional cues are attended to; In a detached task-oriented state of mind it may be beneficial to predict the behaviors of others but emotional information may not be attended to. The complexity and plasticity of mind attribution, including the possibility to train attentional mechanisms and regulation of emotions, is promising for the field of moral enhancement. (shrink)

In this paper, I explore some of the ways in which Artificial Intelligence (AI) is mediated discursively. I assume that AI is informed by an “ancestral dream” to reproduce nature by artificial means. This dream drives the production of “cyborg discourse”, which hinges on the belief that human nature (especially intelligence) can be reduced to symbol manipulation and hence replicated in a machine. Cyborg discourse, I suggest, produces AI systems by rhetorical means; it does not merely describe AI systems or (...) reflect a set of prevailing attitudes about technology. To support this argument, I analyse a set of research articles about an “embodied conversational agent” called the Real Estate Agent (REA). The articles about REA mobilise a set of rhetorical strategies that systematically downplay the system’s artificiality and bolster its humanlike qualities. Within the context of the dream of AI to produce humanlike machines, and given our strong bias for human-human interaction, the designers’ claim to REA’s humanness in their research articles, as I argue in the final section of this paper, needs little justification. (shrink)

Turingin testi, interrogatiivimalli ja tekoäly.