A Critique of Artificial Reason Hubert L. Dreyfus . HUBERT L. DREYFUS What Computers Still Can't Do Thi s One XZKQ-GSY-8KDG What. WHAT COMPUTERS STILL CAN'T DO Front Cover.

Understanding Computers and Cognition presents an important and controversial new approach to understanding what computers do and how their functioning is related to human language, thought, and action. While it is a book about computers, Understanding Computers and Cognition goes beyond the specific issues of what computers can or can't do. It is a broad-ranging discussion exploring the background of understanding in which the discourse about computers and technology takes place. Understanding Computers and Cognition is written for a wide audience, (...) not just those professionals involved in computer design or artificial intelligence. It represents an important contribution to the ongoing discussion about what it means to be a machine, and what it means to be human. Book jacket. (shrink)

Can computers think? This book is intended to demonstrate that thinking, understanding, and intelligence are more than simply the execution of algorithms--that is, that machines cannot think. Written and edited by leaders in the fields of artificial intelligence and the philosophy of computing.

I offer an explication of the notion of computer, grounded in the practices of computability theorists and computer scientists. I begin by explaining what distinguishes computers from calculators. Then, I offer a systematic taxonomy of kinds of computer, including hard-wired versus programmable, general-purpose versus special-purpose, analog versus digital, and serial versus parallel, giving explicit criteria for each kind. My account is mechanistic: which class a system belongs in, and which functions are computable by which system, depends on the system's mechanistic (...) properties. Finally, I briefly illustrate how my account sheds light on some issues in the history and philosophy of computing as well as the philosophy of mind. (shrink)

This paper argues that the idea of a computer is unique. Calculators and analog computers are not different ideas about computers, and nature does not compute by itself. Computers, once clearly defined in all their terms and mechanisms, rather than enumerated by behavioral examples, can be more than instrumental tools in science, and more than source of analogies and taxonomies in philosophy. They can help us understand semantic content and its relation to form. This can be achieved because they have (...) the potential to do more than calculators, which are computers that are designed not to learn. Today’s computers are not designed to learn; rather, they are designed to support learning; therefore, any theory of content tested by computers that currently exist must be of an empirical, rather than a formal nature. If they are designed someday to learn, we will see a change in roles, requiring an empirical theory about the Turing architecture’s content, using the primitives of learning machines. This way of thinking, which I call the intensional view of computers, avoids the problems of analogies between minds and computers. It focuses on the constitutive properties of computers, such as showing clearly how they can help us avoid the infinite regress in interpretation, and how we can clarify the terms of the suggested mechanisms to facilitate a useful debate. Within the intensional view, syntax and content in the context of computers become two ends of physically realizing correspondence problems in various domains. (shrink)

Computers can mimic human intelligence, sometimes quite impressively. This has led some to claim that, a.) computers can actually acquire intelligence, and/or, b.) the human mind may be thought of as a very sophisticated computer. In this paper I argue that neither of these inferences are sound. The human mind and computers, I argue, operate on radically different principles.

In Section 2, I survey some of the ways that computers are used in mathematics. These raise questions that seem to have a generally epistemological character, although they do not fall squarely under a traditional philosophical purview. The goal of this article is to try to articulate some of these questions more clearly, and assess the philosophical methods that may be brought to bear. In Section 3, I note that most of the issues can be classified under two headings: some (...) deal with the ability of computers to deliver appropriate “evidence” for mathematical assertions, a notion that is explored in Section 4, while others deal with the ability of computers to deliver appropriate mathematical “understanding,” a notion that is considered in Section 5. Final thoughts are provided in Section 6. (shrink)

There has been much debate whether computers can be responsible. This question is usually discussed in terms of personhood and personal characteristics, which a computer may or may not possess. If a computer fulfils the conditions required for agency or personhood, then it can be responsible; otherwise not. This paper suggests a different approach. An analysis of the concept of responsibility shows that it is a social construct of ascription which is only viable in certain social contexts and which serves (...) particular social aims. If this is the main aspect of responsibility then the question whether computers can be responsible no longer hinges on the difficult problem of agency but on the possibly simpler question whether responsibility ascriptions to computers can fulfil social goals. The suggested solution to the question whether computers can be subjects of responsibility is the introduction of a new concept, called “quasi-responsibility” which will emphasise the social aim of responsibility ascription and which can be applied to computers. (shrink)

The heyday of discussions initiated by Searle's claim that computers have syntax, but no semantics has now past, yet philosophers and scientists still tend to frame their views on artificial intelligence in terms of syntax and semantics. In this paper I do not intend to take part in these discussions; my aim is more fundamental, viz. to ask what claims about syntax and semantics in this context can mean in the first place. And I argue that their sense is so (...) unclear that that their ability to act as markers within any disputes on artificial intelligence is severely compromised; and hence that their employment brings us nothing more than an illusion of explanation. (shrink)

A technological revolution with first order implications is undeniable and underway. That is the permeation of society by computers and telecommunications technology. For western society, committed to a social, economic, and value structure premised upon an industrial society, the move to an information society is more than disruptive; it is transformational. Current changes are so rapidly paced in relation to business planning that it creates major challenges and opportunities to reach out, influence, and guide the change.The telematics revolution will affect (...) every aspect of our society since it will affect every aspect of our world which involves the generation, production, storage, or handling of information. Many ethical issues are touched upon. To sum them up, the new immorality is to choose to act in ignorance of future consequences. (shrink)

To be an agent, one must be able to formulate intentions. To be able to formulate intentions, one must have a first-person perspective. Computers lack a first-person perspective. So, computers are not agents.

The original proof of the four-color theorem by Appel and Haken sparked a controversy when Tymoczko used it to argue that the justification provided by unsurveyable proofs carried out by computers cannot be a priori. It also created a lingering impression to the effect that such proofs depend heavily for their soundness on large amounts of computation-intensive custom-built software. Contra Tymoczko, we argue that the justification provided by certain computerized mathematical proofs is not fundamentally different from that provided by surveyable (...) proofs, and can be sensibly regarded as a priori. We also show that the aforementioned impression is mistaken because it fails to distinguish between proof search (the context of discovery) and proof checking (the context of justification). By using mechanized proof assistants capable of producing certificates that can be independently checked, it is possible to carry out complex proofs without the need to trust arbitrary custom-written code. We only need to trust one fixed, small, and simple piece of software: the proof checker. This is not only possible in principle, but is in fact becoming a viable methodology for performing complicated mathematical reasoning. This is evinced by a new proof of the four-color theorem that appeared in 2005, and which was developed and checked in its entirety by a mechanical proof system. (shrink)

This book provides a sustained and penetrating critique of a wide range of views in modern cognitive science and philosophy of the mind, from Turing's famous test for intelligence in machines to recent work in computational linguistic theory. While discussing many of the key arguments and topics, the authors also develop a distinctive analytic approach. Drawing on the methods of conceptual analysis first elaborated by Wittgenstein and Ryle, the authors seek to show that these methods still have a great deal (...) to offer in the field of the cognitive theory and the philosophy of mind, providing a powerful alternative to many of the positions put forward in the contemporary literature. Amoung the many issues discussed in the book are the following: the Cartesian roots of modern conceptions of mind; Searle's 'Chinese Room' thought experiment; Fodor's 'language of thought' hypothesis; the place of 'folk psychology' in cognitivist thought; and the question of whether any machine may be said to 'think' or 'understand' in the ordinary senses of these words. Wide ranging, up-to-date and forcefully argued, this book represents a major intervention in contemporary debates about the status of cognitive science an the nature of mind. It will be of particular interest to students and scholars in philosophy, psychology, linguistics and computing sciences. (shrink)

Electronic computers form an integral part of modern mathematical practice. Several high-profile results have been proven with techniques where computer calculations form an essential part of the proof. In the traditional philosophical literature, such proofs have been taken to constitute a posteriori knowledge. However, this traditional stance has recently been challenged by Mark McEvoy, who claims that computer calculations can constitute a priori mathematical proofs, even in cases where the calculations made by the computer are too numerous to be surveyed (...) by human agents. In this article we point out the deficits of the traditional literature that has called for McEvoy’s correction. We also explain why McEvoy’s defence of mathematical apriorism fails and we discuss how the debate over the epistemological status of computer-assisted mathematics contains several unfortunate conceptual reductions. (shrink)

Digital computers play a special role in cognitive science—they may actually be instances of the phenomenon they are being used to model. This paper surveys some of the main issues involved in understanding the relationship between digital computers and cognition. It sketches the role of digital computers within orthodox computational cognitive science, in the light of a recently emerging alternative approach based around dynamical systems.

IN SEPTEMBER 1957, Herbert Simon, a pioneer in cognitive simulation, predicted that within ten years, i.e., by now, a computer would be world chess champion and would prove an important mathematical theorem. This prediction was based on Simon's early initial success in writing a program that could play legal chess and one able to prove simple theorems in logic and geometry. But the early successes turned out to be based on the solution of problems that were simple for machines, and (...) further progress turned out to be more and more difficult, until recently it has begun to be obvious to interested observers, and even to some workers in the field, that the original optimism of Simon and company was unfounded. (shrink)

Recent work on hypercomputation has raised new objections against the Church–Turing Thesis. In this paper, I focus on the challenge posed by a particular kind of hypercomputer, namely, SAD computers. I first consider deterministic and probabilistic barriers to the physical possibility of SAD computation. These suggest several ways to defend a Physical version of the Church–Turing Thesis. I then argue against Hogarth's analogy between non-Turing computability and non-Euclidean geometry, showing that it is a non-sequitur. I conclude that the Effective version (...) of the Church–Turing Thesis is unaffected by SAD computation. (shrink)

In this paper, we address the question of what current computers are from the point of view of human-computer interaction. In the early days of computing, the Turing machine (TM) has been the cornerstone of the understanding of computers. The TM defines what can be computed and how computation can be carried out. However, in the last decades, computers have evolved and increasingly become interactive systems, reacting in real-time to external events in an ongoing loop. We argue that the TM (...) does not provide a mechanistic explanation for interactive computing. The reason is that the fundamental phenomena relevant to interactive computing are out of the scope of classical computability theory. Part of the explanatory power of the TM relies on what we propose to call an execution model. An execution model belongs to a level of abstraction where it is possible to describe both the functional architecture and the execution in mechanistic terms. An updated execution model is warranted to provide the minimal mechanistic description for interactive computation as a counterpart of what the TM could explain regarding Church-Turing computation. It would support an explanation of the ubiquitous computing devices we know - those interacting with humans, e.g., through digital interfaces. We show that such a model is not available within interactive models of computation and that relevant abstractions and concerns are available in computer engineering but need to be identified and gathered. To fill this void, we propose to reflect on the level of abstraction required to support the mechanistic description of an interactive execution and propose some preliminary requirements. (shrink)

Could it be that Deleuze's most lasting legacy will lie in his ‘Postscript on Control Societies’, a mere 2,300-word essay from 1990? While he discussed computers and new media infrequently, Deleuze admittedly made contributions to the contemporary discourse on computing, cybernetics and networks, particularly in his late work. From the concepts of the rhizome and the virtual, to his occasional interjections on the digital versus the analogue, there is a case to be made that the late Deleuze has not only (...) influenced today's discourse on new media but also proposes an original set of arguments about society and politics at the turn of the new millennium. Focusing on the ‘Postscript on Control Societies’ and a handful of texts that surround it, we will reconstruct an image of what it means to live in the information age. This will have consequences for how we define the digital and the analogue, what the computer means, and ultimately provide some insight into one of the more elusive terms in all of Deleuze, the superfold. (shrink)

Discusses the problems that surround the developing science of Artificial Intelligence (AI). This title introduces and clarifies the basic concepts for understanding these problems and also discusses opposing views and possible solutions. It describes the kinds of research that seem to improve our understanding of the mechanisms of intelligence.

As important as it is to realize the potential of computer technology to improve education, it is just as important to understand how the social organization of schools and classrooms influences the use of computers, and in turn is effected by that technology in unanticipated ways. In Computers and Classroom Culture, first published in 1996, Janet Schofield observes the fascinating dynamics of the computer-age classroom. Among her many discoveries, Schofield describes how the use of an artificially-intelligent tutor in a geometry (...) class unexpectedly changes aspects like the level of peer competition and the teachers' grading practices. She also discusses why many teachers fail to make significant instructional use of computers and how gender appears to have a crucial impact on students' reactions to computer use. All educators, sociologists, and psychologists concerned with educational computing and the changing shape of the classroom will find themselves compellingly engaged. (shrink)

Within psychiatric research, the field of 'technotherapy' has been centred primarily on attempts to assess the computer as a treatment tool. The situation of daily clinical usage is, however, often ignored within such research, as for instance in controlled clinical trials. Our empirical study illustrates how health professionals and clients use different concepts of science and health in the attempts of formulating standards for using computers in psychiatric practice. The psychiatrists at a major psychiatric hospital decided and justified clients' use (...) of computers on the basis of a 'techno-medical' quality assurance. At the same hospital the occupational therapists stressed the improvement of social relations as a treatment goal. And, at a psychiatric outside clinic the clients used concepts of 'normality' for articulating quality in computer use. Our study exemplifies how the use of computers is a multifaceted 'performance'. What is called for is a kind of research not limited by artificial borders of 'the context' and the 'user-perspective'. In much humanistic research as well as in action research concepts of 'context' and 'user-perspective' imply a somehow romantic view on practice as pure and uncontaminated by the outside world contrasted to a 'general' or an 'objective' way of knowing the world. These sharp distinctions were however difficult to maintain in our study, where health professionals and clients took local contingencies into account when they interpreted computer use, while they simultaneously drew on a socio-historical reservoir of resources. (shrink)

In Computers Ltd, David Harel, best-selling author of Algorithmics, explains and illustrates one of the most fundamental, yet under-exposed facets of computers - their inherent limitations. Looking at the bad news that is proven, lasting, and robust, discussing limitations that no amounts of hardware, software, talents, or resources can overcome, the book presents a disturbing and provocative view of computing at the start of the 21st century.

The distinction at the heart of van Gelder’s target article is one between digital computers and dynamical systems. But this distinction conflates two more fundamental distinctions in cognitive science that should be keep apart. When this conflation is undone, it becomes apparent that the “computational hypothesis” (CH) is not as dominant in contemporary cognitive science as van Gelder contends; nor has the “dynamical hypothesis” (DH) been neglected.

This work addresses a broad range of questions which belong to four fields: computation theory, general philosophy of science, philosophy of cognitive science, and philosophy of mind. Dynamical system theory provides the framework for a unified treatment of these questions. ;The main goal of this dissertation is to propose a new view of the aims and methods of cognitive science--the dynamical approach . According to this view, the object of cognitive science is a particular set of dynamical systems, which I (...) call "cognitive systems". The goal of a cognitive study is to specify a dynamical model of a cognitive system, and then use this model to produce a detailed account of the specific cognitive abilities of that system. The dynamical approach does not limit a-priori the form of the dynamical models which cognitive science may consider. In particular, this approach is compatible with both computational and connectionist modeling, for both computational systems and connectionist networks are special types of dynamical systems. ;To substantiate these methodological claims about cognitive science, I deal first with two questions in two different fields: What is a computational system? What is a dynamical explanation of a deterministic process? ;Intuitively, a computational system is a deterministic system which evolves in discrete time steps, and which can be described in an effective way. In chapter 1, I give a formal definition of this concept which employs the notions of isomorphism between dynamical systems, and of Turing computable function. In chapter 2, I propose a more comprehensive analysis which is based on a natural generalization of the concept of Turing machine. ;The goal of chapter 3 is to develop a theory of the dynamical explanation of a deterministic process. By a "dynamical explanation" I mean the specification of a dynamical model of the system or process which we want to explain. I start from the analysis of a specific type of explanandum--dynamical phenomena--and I then use this analysis to shed light on the general form of a dynamical explanation. Finally, I analyze the structure of those theories which generate explanations of this form, namely dynamical theories. (shrink)

This chapter will deal with an issue that is as much a problem for moral philosophy as it is for the computer world. My basic theme is that high technology, and computer technology in particular, raises ethical problems of a new sort that require considerable restructuring of our traditional ethical categories. It follows that our job as philosophers is not, as it is often thought to be, simply to apply ready-made categories to new situations; rather, it is to find new (...) categories, or new ways of interpreting old categories, in order to accommodate the challenges presented by new technologies. (shrink)

Open peer commentary on the article “Constructivism and Computation: Can Computer-Based Modeling Add to the Case for Constructivism?” by Manfred Füllsack. Upshot: While I do agree with Füllsack’s positive assessment of the use of computer simulations in advancing constructivism’s program, I am less convinced by the alleged opposition between computers and constructivism he builds up. In my opinion, his depiction of computers and computation is inaccurate in several respects. As a result, the alleged incompatibility with constructivism Füllsack objects to disappears, (...) and his whole essay could then be construed as a classic straw man fallacy. This is unfortunate, because it could lead to an unwarranted dismissal of his positive contribution. (shrink)

The production of mathematical formalism on state of the art computers is quite different than by pen and paper. In this paper, I examine the question of how different media influence mathematical writing. The examination is based on an investigation of professional mathematicians' use of various media for their writing. A model for describing mathematical writing through turn-takings is proposed. The model is applied to the ways mathematicians use computers for writing, and especially it is used to understand how interaction (...) with the computer system LaTeX is different in the case of formulas than in the case of diagrams. (shrink)

The most cursory examination of the history of artificial intelligence highlights numerous egregious claims of its researchers, especially in relation to a populist form of ‘strong’ computationalism which holds that any suitably programmed computer instantiates genuine conscious mental states purely in virtue of carrying out a specific series of computations. The argument presented herein is a simple development of that originally presented in Putnam’s (Representation & Reality, Bradford Books, Cambridge in 1988 ) monograph, “Representation & Reality”, which if correct, has (...) important implications for turing machine functionalism and the prospect of ‘conscious’ machines. In the paper, instead of seeking to develop Putnam’s claim that, “everything implements every finite state automata”, I will try to establish the weaker result that, “everything implements the specific machine Q on a particular input set ( x )”. Then, equating Q ( x ) to any putative AI program, I will show that conceding the ‘strong AI’ thesis for Q (crediting it with mental states and consciousness) opens the door to a vicious form of panpsychism whereby all open systems, (e.g. grass, rocks etc.), must instantiate conscious experience and hence that disembodied minds lurk everywhere. (shrink)

We comment on the collection of papers inspired by our book Why Machines Will Never Rule the World published in volume 12 (5+6) of the journal Cosmos+Taxis. We summarize the arguments made by the contributors about what we say in the book, and then show where we disagree.

A response to a recent critique by Cem Bozşahin of the theory of syntactic semantics as it applies to Helen Keller, and some applications of the theory to the philosophy of computer science.

I’m one of those who is awed and impressed by the potential of this field and have devoted some part of my energy to persuading people that it is a positive force. I have done so largely on the grounds of its economic benefits and it potential for making the fruits of computer technology more generally available to the public — for example, to help the overworked physician; to search for oil and minerals and help manage our valuable resources; to (...) explore, mine, and experimentindangerousenvironments;toallowthenon-computingpublicaccesstovast libraries of important information and even advice, and in the process give real meaning to the.. (shrink)