A Hypothesis on the Origin of the Sign Types

Abstract

The functioning of natural language communication depends crucially on the different kinds of signs, i.e., symbol, indexical, and name, and their characteristic mechanisms of reference. In this paper, the meanings of these sign types are traced to cognitive agents with memory but without language.

The argument is based on the task of relating the view-dependent format of a content stemming from an agent’s current situation to a view-independent format for storage in memory. The view-dependent analysis of a content, called a task analysis, is built from cognitive structures suitable to serve as the literal meanings of the different sign types. Furthermore, the transfer between the view-dependent format relating to the current situation and the view-independent format of content stored in memory in [-language] agents is shown to be closely related to the alternation between the hearer and the speaker mode in [+language] agents.

While perception is a necessary condition for life, it is not sufficient. For example, we may grant that robots have perception, yet deny that they are alive.