ABSTRACT
Any attempt to understand media and to relate them to cognition and learning through the symbol systems they employ would seem to require an explication of what symbol systems are. However, this is more easily said than done. Pictures differ qualitatively from prose, facial expressions differ from graphs, and film differs from chemical models; all this is obvious. But confusion arises when one attempts to introduce superordinate dimensions along which different symbol systems can be arrayed for the purpose of differentiating systematically among them and characterizing the media that employ them. Given the diversity of meanings ascribed to such terms as symbol, sign, icon, symbol system, realism, and likeness, we may find ourselves attempting to sort out, clarify, and define issues whose understanding is perhaps best served by intuition. To understand how symbol systems interact with cognition and learning in principle may perhaps require no more than an intuitive understanding of how symbol systems function as modes of appearance—that is, how they express and represent communicable qualities in different ways. Going beyond our intuition, we run the risk, as Berkeley put it, of first raising the dust and then claiming that we cannot see. Yet, some analysis of the general issues concerning symbol systems seems necessary. This chapter begins with a logical analysis of symbol systems, following primarily the work of Nelson Goodman (1968). The reason for choosing Goodman’s theory as the basis of my (rather sketchy) description of the nature of symbol systems is that his treatment is the logically most stringent, consistent, and clear. The logical description of symbol systems is then followed by a psychological treatment of some of Goodman’s notions. Of particular interest is the question of resemblance, similarity, or iconicity in symbolic representation, an issue that figures prominently in the treatment of media in education. There are two reasons for choosing to begin with a logical analysis and to continue with a psychological one. First, not everything that follows from a logical analysis is psychologically consistent. Second, neither of the two approaches taken alone seems to suffice as a treatment of symbol systems.
