Ways of Understanding: Semiotics and Hermeneutics
When asked what it is they study, semioticians of all cultures tend to respond by formulating dichotomies such as “signals and symbols” or “signification and communication”. Their critics tend to reply by claiming that semiotics evidently has no unified subject matter and they draw the conclusion that semiotics may certainly be an interesting hermeneutic practice, but is not entitled to conceive of itself as a scientific discipline.
In this controversy the present lecture takes the position that semiotics does have a unified subject matter as well as the status of a scientific discipline: It studies sign processes, i.e., the processes that make individual organisms and groups of organisms as well as machines and systems of Artificial Intelligence reach an understanding of their environment. Semiotics investigates everything for which it makes sense to ask whether it can understand or be understood, calling them “interpreter” and “sign”, respectively. The processes which an interpreter must perform in understanding the signs given are a suitable basis for establishing a typology of signs which clarifies the relationship between the various types of sign processes including signification and communication.
In constructing such a sign typology, four increasingly specialized basic sign types are introduced: If something causes the interpreter to respond with an overt or covert change of state, it is a SIGNAL, if something causes the interpreter to make a specific assumption, it is an INDICATOR. If something causes the interpreter to assume that its producer is in a specific state, it is an EXPRESSION (of this state). If something causes the interpreter to assume that its producer has the intention to perform a specific further action, it is an INTENTION EXPRESSION (also called “GESTURE”) announcing that action.
The basic sign types can occur on increasingly complex levels of sign use (levels of reflection): Responding to an inarticulate cry by assuming that its producer is in pain amounts to taking it as an expression and involves what is called “SIGN RECEPTION” (level 1a). Uttering such a cry with the intention to bring about such a reaction amounts to expressing and involves what is called “MANIPULATION” (SIGN PRODUCTION; level 1b). When the cry is stylized (e.g., manifesting some language sign such as German “aua”, English “oh” or French “ai”), it is not only received as an expression but also interpreted as the product of an expressing attempt (SIMULATION, RECEPTION OF SIGN PRODUCTION; level 2a). When such words are pronounced without crying they are no longer taken as an expressing attempt but only as indicating an expressing (INDICATION OF SIGN PRODUCTION; level 2b). Expressing one’s state by indicating an expressing is a case of COMMUNICATION (SIGN PRODUCTION; level 2b including Communication Condition). On the level of communication, signals become IMPERATIVES, indicators ASSERTIVES, expressions EXPRESSIVES, and intention expressions COMMISSIVES (in the sense of John Searle).
Now, expressing is often taken as signaling to get help. When someone relies on this tendency and articulates an expression with an intention of having it interpreted as a signal, he performs INDIRECT COMMUNICATION. By uttering an expressive he produces an imperative.
It is demonstrated that all these sign concepts can be defined on the basis of the predicates BELIEVE, CAUSE and INTEND. This approach also allows an explication of the concepts of ACTION, INTERPRETATION and INDICATION of ACTION, and DECLARATION.
The given hierarchy of sign concepts sketched here is exhaustive, covering all possibilities of semiosis, but open to further differentiation. It is offered as a theoretical basis for the reconstruction of sign processes in Artificial Intelligence research.