(back to list)

It is widely supposed that thought precedes and anticipates action, that conjectures precede experiments designed to test them and that science provides a rational understanding of nature, which technologies merely embody. Thus, in science as in other human activities, tools and technologies are consequences of creative, intentional, purposive thought but have no role in enabling or shaping thought. The pre-eminence of mind is also expressed by the priority assigned to representations and reasoning processes in cognitive psychology and, similarly, to ideas and arguments in the history of science, with its corollary that technologies emerge from the application of scientifically discerned principles. Until recently the design of intelligent systems also presupposed that functional competence could be achieved by providing a system with abstract representations of objects and processes in an environment and with the means of reasoning with and about these.
   This view implies that systematic, rational thought is or can be separate from the world that it seeks to understand, manipulate or control. It is not easy to show how profoundly mistaken this view is. One source of difficulty is that the priority of mental representations over physical manipulations presupposes a mind-body dualism that has an established philosophical pedigree. This dualism is associated with Descartes, whose optical ray-diagrams of eyeballs gave us some of the earliest explanations of how images of the material world are re-created in the theatre of the mind. There they can be explored and manipulated in thought, by methods perfected and popularised by Galileo, who embedded them in thought-experimental narratives. In the work of Galileo, Newton and Einstein it appeared that thought could transcend the limitations of embodied, sensory experience. Experiments and mere technological extensions to our senses could only follow the lead of ratiocination. The demonstrative force of thought experiments is taken by some to show that Plato was right: the human mind can directly intuit the nature of reality. If this view of knowledge and how we acquire it is correct then, as Dreyfus, Edelman and others have argued, the anticipatory, instructional approach to artificial intelligence should have been far more successful than it has been.
   This constraint is illustrated by an important feature of the history of science. The sciences frequently run up against the limitations of a way of representing aspects of the world – from material objects such as fundamental particles to abstract entities such as numbers or space and time. One of the most profound changes in our ability to describe aspects of experience has involved developing new conceptions of what it is possible to represent. Examples include the invention of Euclidean geometry as a way of reasoning consistently about spatially extended objects; Cartesian co-ordinates which linked geometrical and algebraic reasoning; the calculus, which provided a way of reasoning about continuously changing variables; the theory of probability as a way of calculating rather than estimating the likelihood of events, and diagrammatic techniques in quantum physics that represent properties of mathematical descriptions which themselves already transcend the phenomenological or visual approach of classical physics. Computer-based simulation methods may turn out to be a similar representational turning point for the sciences. An important point about these developments is that they are not merely ways of describing. Unlike sense-extending devices such as microscopes, telescopes or cosmic ray detectors, each enabled a new way of thinking about a particular domain.

Experiment as an Instrument of Innovation:,,Experience and Embodied Thought,,David C. Gooding, Science Studies Centre, Department of Psychology,,University of Bath, Bath BA2 7AY, UK (click for complete article)