In the previous chapter, the characteristics of the three approaches to cognitive science were reviewed, highlighting important distinctions between the classical, connectionist, and embodied approaches. This was done by exploring the analogy between cognitive science and classical music. It was argued that each of the three approaches within cognitive science was analogous to one of three quite different traditions within classical music, and that these differences were apparent in how each approach studied music cognition. However, at the end of the chapter the possibility of hybrid theories of music cognition was raised.
The possibility of hybrid theories of music cognition raises the further possibility that the differences between the three approaches within cognitive science might not be as dramatic as could be imagined. The purpose of the current chapter is to explore this further possibility. It asks the question: are there marks of the classical? That is, is there a set of necessary and sufficient properties that distinguish classical theories from connectionist and embodied theories?
The literature suggests that there should be a large number of marks of the classical. It would be expected that classical theories appeal to centralized control, serial processing, local and internal representations, explicit rules, and a cognitive vocabulary that appeals to the contents of mental representations. It would also be expected that both connectionist and embodied theories reject many, if not all, of these properties.
In the current chapter we examine each of these properties in turn and make the argument that they do not serve as marks of the classical. First, an examination of the properties of classical theories, as well as a reflection on the properties of the computing devices that inspired them, suggests that none of these properties are necessary classical components. Second, it would also appear that many of these properties are shared by other kinds of theories, and therefore do not serve to distinguish classical cognitive science from either the connectionist or the embodied approaches.
The chapter ends by considering the implications of this conclusion. I argue that the differences between the approaches within cognitive science reflect variances in emphasis, and not qualitative differences in kind, amongst the three kinds of theory. This sets the stage for the possibility of hybrid theories of the type examined in Chapter 8.