(From the Press's Website) -/- Winner of the 2004 Lakatos Prize, Thought in a Hostile World is an exploration of the evolution of cognition, especially human cognition, by one of today's foremost philosophers of biology and of mind. Features an exploration of the evolution of human cognition. Written by one of today’s foremost philosophers of mind and language. Presents a set of analytic tools for thinking about cognition and its evolution. Offers a critique of nativist, modular versions of evolutionary psychology, (...) rejecting the example of language as a model for thinking about human cognitive capacities. Applies to the areas of cognitive science, philosophy of mind, and evolutionary psychology. -/- . (shrink)

Contemporary realist theories of value claim to be compatible with natural science. In this paper, I call this claim into question by arguing that Darwinian considerations pose a dilemma for these theories. The main thrust of my argument is this. Evolutionary forces have played a tremendous role in shaping the content of human evaluative attitudes. The challenge for realist theories of value is to explain the relation between these evolutionary influences on our evaluative attitudes, on the one hand, and the (...) independent evaluative truths that realism posits, on the other. Realism, I argue, can give no satisfactory account of this relation. On the one hand, the realist may claim that there is no relation between evolutionary influences on our evaluative attitudes and independent evaluative truths. But this claim leads to the implausible skeptical result that most of our evaluative judgements are off track due to the distorting pressure of Darwinian forces. The realist’s other option is to claim that there is a relation between evolutionary influences and independent evaluative truths, namely that natural selection favored ancestors who were able to grasp those truths. But this account, I argue, is unacceptable on scientific grounds. Either way, then, realist theories of value prove unable to accommodate the fact that Darwinian forces have deeply influenced the content of human values. After responding to three objections, the third of which leads me to argue against a realist understanding of the disvalue of pain, I conclude by sketching how antirealism is able to sidestep the dilemma I have presented. Antirealist theories of value are able to offer an alternative account of the relation between evolutionary forces and evaluative facts —
an account that allows us to reconcile our understanding of evaluative truth with our understanding of the many non-rational causes that have played a role in shaping our evaluative judgements. (shrink)

Artificial intelligence research has foundered on the issue of representation. When intelligence is approached in an incremental manner, with strict reliance on interfacing to the real world through perception and action, reliance on representation disappears. In this paper we outline our approach to incrementally building complete intelligent Creatures. The fundamental decomposition of the intelligent system is not into independent information processing units which must interface with each other via representations. Instead, the intelligent system is decomposed into independent and parallel activity (...) producers which all interface directly to the world through perception and action, rather than interface to each other particularly much. The notions of central and peripheral systems evaporateeverything is both central and peripheral. Based on these principles we have built a very successful series of mobile robots which operate without supervision as Creatures in standard office environments. (shrink)

This article considers the issue of opacity as a problem for socially consequential mechanisms of classification and ranking, such as spam filters, credit card fraud detection, search engines, news trends, market segmentation and advertising, insurance or loan qualification, and credit scoring. These mechanisms of classification all frequently rely on computational algorithms, and in many cases on machine learning algorithms to do this work. In this article, I draw a distinction between three forms of opacity: opacity as intentional corporate or state (...) secrecy, opacity as technical illiteracy, and an opacity that arises from the characteristics of machine learning algorithms and the scale required to apply them usefully. The analysis in this article gets inside the algorithms themselves. I cite existing literatures in computer science, known industry practices, and do some testing and manipulation of code as a form of lightweight code audit. I argue that recognizing the distinct forms of opacity that may be coming into play in a given application is a key to determining which of a variety of technical and non-technical solutions could help to prevent harm. (shrink)

For millennia, philosophers have speculated about the origins of ethics. Recent research in evolutionary psychology and the neurosciences has shed light on that question. But this research also has normative significance. A standard way of arguing against a normative ethical theory is to show that in some circumstances the theory leads to judgments that are contrary to our common moral intuitions. If, however, these moral intuitions are the biological residue of our evolutionary history, it is not clear why we should (...) regard them as having any normative force. Research in the neurosciences should therefore lead us to reconsider the role of intuitions in normative ethics. (shrink)

Recent progress in artificial intelligence has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats that of humans in some respects. Despite their biological inspiration and performance achievements, these systems differ from human intelligence in crucial ways. We review progress in cognitive science suggesting that truly human-like learning and thinking (...) machines will have to reach beyond current engineering trends in both what they learn and how they learn it. Specifically, we argue that these machines should build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; ground learning in intuitive theories of physics and psychology to support and enrich the knowledge that is learned; and harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. We suggest concrete challenges and promising routes toward these goals that can combine the strengths of recent neural network advances with more structured cognitive models. (shrink)

In this essay, I draw on Haidt’s and Baron’s respective insights in the service of a bit of philosophical psychoanalysis. I will argue that deontological judgments tend to be driven by emotional responses, and that deontological philosophy, rather than being grounded in moral reasoning, is to a large extent3 an exercise in moral rationalization. This is in contrast to consequentialism, which, I will argue, arises from rather different psychological processes, ones that are more “cognitive,” and more likely to involve genuine (...) moral reasoning. These claims are strictly empirical, and I will defend them on the basis of the available evidence. Needless to say, my argument will be speculative and will not be conclusive. Beyond this, I will argue that if these empirical claims are true, they may have normative implications, casting doubt on deontology as a school of normative moral thought. (shrink)

Four experiments examined people’s folk-psychological concept of intentional action. The chief question was whether or not _evaluative _considerations — considerations of good and bad, right and wrong, praise and blame — played any role in that concept. The results indicated that the moral qualities of a behavior strongly influence people’s judgements as to whether or not that behavior should be considered ‘intentional.’ After eliminating a number of alternative explanations, the author concludes that this effect is best explained by the hypothesis (...) that evaluative considerations do play some role in people’s concept of intentional action.<b> </b>. (shrink)

When a chimpanzee stockpiles rocks as weapons or when a frog sends out mating calls, we might easily assume these animals know their own motivations--that they use the same psychological mechanisms that we do. But as Beyond the Brain indicates, this is a dangerous assumption because animals have different evolutionary trajectories, ecological niches, and physical attributes. How do these differences influence animal thinking and behavior? Removing our human-centered spectacles, Louise Barrett investigates the mind and brain and offers an alternative approach (...) for understanding animal and human cognition. Drawing on examples from animal behavior, comparative psychology, robotics, artificial life, developmental psychology, and cognitive science, Barrett provides remarkable new insights into how animals and humans depend on their bodies and environment--not just their brains--to behave intelligently. Barrett begins with an overview of human cognitive adaptations and how these color our views of other species, brains, and minds. Considering when it is worth having a big brain--or indeed having a brain at all--she investigates exactly what brains are good at. Showing that the brain's evolutionary function guides action in the world, she looks at how physical structure contributes to cognitive processes, and she demonstrates how these processes employ materials and resources in specific environments. Arguing that thinking and behavior constitute a property of the whole organism, not just the brain, Beyond the Brain illustrates how the body, brain, and cognition are tied to the wider world. (shrink)

Some philosophers have recently argued that decision-makers ought to take normative uncertainty into account in their decisionmaking. These philosophers argue that, just as it is plausible that we should maximize expected value under empirical uncertainty, it is plausible that we should maximize expected choice-worthiness under normative uncertainty. However, such an approach faces two serious problems: how to deal with merely ordinal theories, which do not give sense to the idea of magnitudes of choice-worthiness; and how, even when theories do give (...) sense to magnitudes of choice-worthiness, to compare magnitudes of choice-worthiness across different theories. Some critics have suggested that these problems are fatal to the project of developing a normative account of decision-making under normative uncertainty. The primary purpose of this article is to show that this is not the case. To this end, I develop an analogy between decision-making under normative uncertainty and the problem of social choice, and then argue that the Borda Rule provides the best way of making decisions in the face of merely ordinal theories and intertheoretic incomparability. (shrink)

The computational argument for individualism, which moves from computationalism to individualism about the mind, is problematic, not because computationalism is false, but because computational psychology is, at least sometimes, wide. The paper provides an early, or perhaps predecessor, version of the thesis of extended cognition.

Moral judgments about an agent's behavior are enmeshed with inferences about the agent's mind. Folk psychology—the system that enables such inferences—therefore lies at the heart of moral judgment. We examine three related folk-psychological concepts that together shape people's judgments of blame: intentionality, choice, and free will. We discuss people's understanding and use of these concepts, address recent findings that challenge the autonomous role of these concepts in moral judgment, and conclude that choice is the fundamental concept of the three, defining (...) the core of folk psychology in moral judgment. (shrink)

Computationalism has been the mainstream view of cognition for decades. There are periodic reports of its demise, but they are greatly exaggerated. This essay surveys some recent literature on computationalism. It concludes that computationalism is a family of theories about the mechanisms of cognition. The main relevant evidence for testing it comes from neuroscience, though psychology and AI are relevant too. Computationalism comes in many versions, which continue to guide competing research programs in philosophy of mind as well as psychology (...) and neuroscience. Although our understanding of computationalism has deepened in recent years, much work in this area remains to be done. (shrink)

This collection of eleven papers by Elijah Millgram (nine of which have been previously published) is ostensibly united by the thesis that the best way to go about assessing moral theories is to identify the view of practical reasoning that each such theory rests upon, and evaluate the adequacy of these respective theories of practical reasoning. The correct moral theory, Millgram assures us, will be the one that is paired with the best theory of practical reasoning. He outlines this methodology (...) in a substantial (32 pp.) introduction. Why should we adopt Millgram’s method? A host of concerns immediately leap to mind. If two or more different moral theories rest on the same theory of practical reasoning, then how would discovering the latter to be the correct theory of practical reasoning help us decide among the moral theories? What if a given moral theory is consistent with two or more different theories of practical reasoning? What if we cannot evaluate theories of practical reasoning independently of having adopted a moral perspective? Millgram doesn’t address these natural questions head on, but rather proposes that the essays of the volume collectively constitute a “feasibility demonstration” (p. 3) of the method. In other words, the only way that we will be persuaded that the pairings between moral theories and practical reasoning theories are tight enough to support this grand project is to get our hands dirty in detailed discussion of particular moral theories, particular theories of practical reasoning, and the relations between them. It is through seeing the method at work that we will become convinced, Millgram hopes, of several interlocking theses: (1) that each of the major moral theories of the past has had a distinctive take on practical reasoning; (2) that pivotal structural elements of these theories are due to the underlying theory of practical reasoning; (3) that problems in a moral theory can often be traced to problems in the underlying theory of practical reasoning; (4) that theories of practical reasoning are “engines” (p.. (shrink)