The evolution of human symbolic capacity must have been very rapid even in some intermediate stage (e.g. the proto-symbolic behavior of Homo erectus). Such a rapid process requires a runaway model. The type of very selective and hyperbolically growing self-organization called “hypercyle” by Eigen and Schuster could explain the rapidity and depth of the evolutionary process, whereas traditional runaway models of sexual selection seem to be rather implausible in the case of symbolic evolution. We assume two levels: at the first (...) level the species is adapted to ecological demands and accumulates the effects of this process in the genome. At the second level a kind of social/cultural knowledge is accumulated via a set of symbolic forms, one of which is language. Bühler’s model of three basic functions of signs can also be elaborated so that its cyclic structure becomes apparent. We assume that the hypercyclic process of semiosis and functional differentiation was triggered in 2 my BP (with the Homo erectus) and got more and more speed with the species Homo sapiens and later. The consequences for the evolutionary stratification of human languages will be drawn in the last section of the paper. The basic aim of the paper is to provide a semiotic (and not just a linguistic) explanation of the origin of language which can be linked to relevant models in evolutionary biology and which exploits the possibilities contained in self-organizing systems. (shrink)

In this paper I assess the explanatory role of internal representations in connectionist models of cognition. Focusing on both the internal‘hidden’units and the connection weights between units, I argue that the standard reasons for viewing these components as representations are inadequate to bestow an explanatorily useful notion of representation. Hence, nothing would be lost from connectionist accounts of cognitive processes if we were to stop viewing the weights and hidden units as internal representations.