Brainchildren: Essays on Designing Minds

摘要

Reviewed by: Brainchildren: Essays on Designing Minds* Stuart Bennett (bio) Brainchildren: Essays on Designing Minds. By Daniel C. Dennett. Cambridge, Mass.: MIT Press, 1998. Pp. ix+417; illustrations, notes/references, bibliography, index. $20. Daniel C. Dennett’s philosophical enquiries into belief, consciousness, and the mind are important and influential. This book, a selection of his essays published during the past twelve years, demonstrates the breadth of his thinking—the essays cover fundamental issues relating to consciousness and belief as well as artificial intelligence, artificial life, ethology, and the animal mind—and the extent of his knowledge of the relevant science and technology. Two essays, “Real Patterns” (1991) and “Real Consciousness” (1994), set out the core of Dennett’s thinking. In the former he considers the [End Page 720] question “are there really beliefs?” and argues for a position lying between a materialism that requires, for beliefs to be real, physical existence in some sense, and a materialism that denies the existence of beliefs. This is an uncomfortable position to occupy if you also reject dualism: if beliefs are not real in a physical sense but exist then how do they exist? He answers this question in the second essay, which deals with his so called Multiple Drafts Model, in which he argues that conscious experience is “a succession of states constituted by various processes occurring in the brain and not something over and above these processes which is caused by them” (p. 136). Is there a resonance here with Lyotard’s idea of “little stories” and the absence of a master narrative? There is no master judge in the mind who oversees the construction of conscious experience (our narrative) from each tiny experience but a continuous process of interaction that in itself constitutes consciousness. And contained within our narrative are our beliefs. Dennett makes the effort to write for both the professional philosopher and the “General Reader.” Not surprisingly, this general reader, a professional engineer, found the papers concerned with issues such as “can machines think?” and with the practical requirements for creating a conscious robot more accessible than papers concerned with theories of consciousness. His essays on machines (except for the one on the requirements for making a conscious robot) are dated in their conclusions; for example, the concerns about the subtle dangers of expert systems are now well known and heeded (or are they?). Nevertheless, the arguments presented are interesting and worth reading. Dennett takes the view that we will reach a time when we will be able to construct a machine that can pass the strong Turing test, that is, give answers to a human’s questioning that are indistinguishable from a human’s answering. Moreover, he asserts, any computer (robot, or other form of artificial intelligent entity) that passes the test must be accepted as a self-conscious thinking entity. But he doubts that we will build such a machine, simply because it will be pointless to do so: machines that meet restricted, special-purpose needs will suffice. The essays in the book are themselves interesting, but for whom is book intended? There is no introductory essay to provide context or Dennett’s own reflections on his earlier work. The short notes at the beginning of each essay are too brief and fragmented to serve this purpose. Some are also tantalizing: Dennett tells us, in the opening remarks for the final essay in the book, which dates from 1986, that he has returned to the issue discussed and it is more complicated than he thought. The issue in question is the proposition that we have reached a point at which advances in technology are such that, rather than its supporting us in achieving interesting and morally good lives, we may be forced to choose between lives that are interesting and lives that are morally good. The book is offered as a student text, and it will serve thi