John Symons

Computational modeling plays an increasingly important explanatory role in cases where we investigate systems or problems that exceed our native epistemic capacities. One clear case where technological enhancement is indispensable involves the study of complex systems.1 However, even in contexts where the number of parameters and interactions that define a problem is small, simple systems sometimes exhibit non-linear features which computational models can illustrate and track. In recent decades, computational models have been proposed as a way to assist us in understanding emergent phenomena.

An asymmetry between the demands at the computational and algorithmic levels of description furnishes the illusion that the abstract profile at the computational level can be multiply realized, and that something is actually being shared at the algorithmic one. A disembodied rendering of the situation lays the stress upon the different ways in which an algorithm can be implemented. However, from an embodied approach, things look rather different. The relevant pairing, I shall argue, is not between implementation and algorithm, but rather between algorithm and computation. The autonomy of psychology is a result of the failure to appreciate this point.

This paper argues that the difference between contemporary software intensive scientific practice and more traditional non-software intensive varieties results from the characteristically high conditionality of software. We explain why the path complexity of programs with high conditionality imposes limits on standard error correction techniques and why this matters. While it is possible, in general, to characterize the error distribution in inquiry that does not involve high conditionality, we cannot characterize the error distribution in inquiry that depends on software. Software intensive science presents distinctive error and uncertainty modalities that pose new challenges for the epistemology of science

Skeptics argue that the acquisition of knowledge is impossible given the standing possibility of error. We present the limiting convergence strategy forresponding to skepticism and discuss the relationship between conceivable error and an agent’s knowledge in the limit. We argue that the skeptic must demonstrate that agents are operating with a bad method or are in an epistemically cursed world. Such demonstration involves a significant step beyond conceivability and commits the skeptic to potentially convergent inquiry.

Horgan’s perceptive discussion of Freudian psychology, Prozac and evolutionary biology cannot mitigate the problems that seriously weaken his book (Horgan, 1999). While he certainly manages to deflate some of the more outrageous hype surrounding the scientific and often not-so-scientific study of the mind, his criticism of the brain and behavioral sciences contains a number of flaws, some of which I will address below. My response focuses on his discussion of neuroscience. As we shall see, the three mysteries that Horgan believes cripple neuroscience are certainly not as serious as he insists.

In this paper, we argue for the centrality of prediction in the use of computational models in science. We focus on the consequences of the irreversibility of computational models and on the conditional or ceteris paribus, nature of the kinds of their predictions. By irreversibility, we mean the fact that computational models can generally arrive at the same state via many possible sequences of previous states. Thus, while in the natural world, it is generally assumed that physical states have a unique history, representations of those states in a computational model will usually be compatible with more than one possible history in the model. We describe some of the challenges involved in prediction and retrodiction in computational models while arguing that prediction is an essential feature of non-arbitrary decision making. Furthermore, we contend that the non-predictive virtues of computational models are dependent to a significant degree on the predictive success of the models in question.

SPECIAL INTRODUCTORY PRICE! Daniel Dennett has been one of the central voices in the philosophy of mind for at least the past forty years. Unlike most philosophers of his generation, Dennett’s work has resonated far and wide. It has powerfully influenced the development of cognitive science, robotics, developmental psychology, and artificial intelligence. Indeed, his work has led to many new lines of inquiry. For example, he has developed a theory of consciousness which provides an approach to naturalizing mind which circumvents many of the most significant philosophical arguments against the possibility of a scientific explanation of consciousness. The daunting quantity of literature available on Dennett makes it difficult to discriminate the useful from the tendentious, superficial, and otiose. Moreover, because no comparable philosopher has had a profound impact across such a wide range of disciplines and on intellectual culture in general, responses to Dennett’s philosophy are dispersed across a broad range of scientific, philosophical, and cultural domains. That is why this new title in the highly regarded Routledge series, Critical Assessments of Leading Philosophers, is so urgently needed. Edited by John Symons, this new Routledge Major Work is a four-volume collection of the best scholarship on Dennett; the collected materials have been carefully selected from a wide range of academic journals, edited collections, research monographs, and other sources. The tightly focused organization of this collection allows users quickly and easily to access both established and cutting-edge assessments of Dennett’s work. The set is also made for irresistible browsing. With comprehensive introductions to each volume, providing essential background information and relating the various works to each other, Daniel Dennett is destined to be an indispensable resource for research and study