Copyright © 1999 Massachusetts Institute of Technology
Although the term "semiotics" has come into common usage only in this present century, the discipline of semiotics itself is much older and may be traced back to Greek philosophy. The first formulations of the notion of the so-called semiotic triangle may be found already in Aristotle, where three elements are seen as constitutive of signs: pragmata ("the things to which the sign refers"), the expressive element ("that which is in the voice"), and thoughts ("that which is in the mind"). This third element alludes to a level that we today would call the level of mental representations and, as such, is that aspect of the sign that is directly concerned with cognition. We could therefore say that semiotics is constitutively connected with cognition right from its historical roots.
In modern times, there seem to be two main directions in semiotic research projects: interpretative semiotics inspired by the work of Charles Sanders Peirce (1931-58), and structural semiotics, sometimes referred to as "semiology" following the terminology introduced by the Swiss linguist Ferdinand de SAUSSURE (1906 - 11). Their concerns with the issue of cognition and their possible connections with related works on contemporary cognitive science map onto different areas of interest, and I will discuss them separately.
It is certainly within Peircean semiotics that the most straightforward link between semiotic processes and processes of cognition can be found, so much so that this direction in semiotics is often referred to as cognitive semiotics. According to Peirce, KNOWLEDGE ACQUISITION and thought are never immediate, direct processes but are always mediated through signs, or interpretants, which are more developed signs and which thus allow the subject to know more than she knew before, in an endless process of interpretation known as unlimited semiosis. Interpretants, which are the central element in the sign process, are first of all mind-internal signs, that is to say mental representations. In this way, thought, signs, and cognition become one and the same thing. "All thought is in signs" (Collected Papers 5: 252), but because each interpretant sign adds something new to the process of thinking and knowing, cognition is strongly characterized as an inferential process.
Peirce's philosophy has strongly influenced Umberto Eco's semiotic theory of signs and interpretational processes (Eco 1976, 1984). Eco stresses in his theories the Peircean notion of semiosis as an inherently inferential process. Signs are not regulated by some kind of equivalence rule between expression and content but are always considered as inferential devices, even in cases where the inferential relation between the two has became so stabilized as to appear to be a purely automatized correlation. In this way, Eco, not unlike Peirce, considers that sign processes will always, and necessarily, imply the factual occurrence of some form of internal reasoning and inferencing process, so making the sign process that is the object of study of his semiotics functionally indistinguishable from cognition.
The particular type of thought process involving inferences based on the interpretation of signs is referred to by Peirce as "abduction," which he distinguishes from INDUCTION and deduction on the basis of their fundamental logical forms. Abductive inference is considered as constituting the core of cognition itself, because it is the only type of inferential process that actually contributes to an increase in our knowledge of the object and, without deriving logical truths, only possible ones.
Abductive reasoning has recently received growing attention in cognitive circles, especially in relation to modeling in Artificial Intelligence (AI), where abduction is seen as a theory-forming or interpretative inference. In their textbook on AI, Charniak and McDermott (1985) claim that everyday reasoning as well as medical diagnosis, story understanding, and vision are all abductive processes. Many AI systems are presently being developed around various forms of abductive inference in order to model different areas of cognition (see Josephson and Josephson 1994), from perception to natural language understanding and even METAPHOR (Hobbs 1992) and translation (Hobbs and Kameyama 1990). In particular, abduction has been considered in relation to interpretation by both AI researchers (e.g., Hobbs et al. 1993) and by semioticians (Eco 1979, 1990). According to Eco (1979), no text can make all its premises explicit; in this sense, a text can be seen as a lazy machine asking the reader to fill in a whole series of gaps. In order to fully understand a text, the reader has to make a series of abductions (known as "inferential walks") on the basis of both her general knowledge of the world and specific textual scripts. Such an approach appears highly consistent with most of the cognitive work done in the field of READING comprehension of text (see among others Schank 1982; van Dijk and Kintsch 1983).
Turning now to the structuralist semiotic perspective developed along the lines of Saussure and Hjelmslev, one could say that this latter approach is far less interested in issues related to cognition. However, there is at least one area that brings to light some interesting similarities with work in cognitive science, and this is the study of narrative structure. Since the seminal work of Vladimir Propp (1928), structural semiotics has focused on the analysis of the structural properties of any kind of narrative text, developing a highly complex and articulated model to account for the different levels of structural organization (see Greimas 1970, 1983). This line of research could be compared to work done in the cognitive area on story grammar (Rumelhart 1975, 1980; Thorndyke 1977), which also aims to individuate an underlying structure in stories and to define the nature of concepts such as state and event, despite the fact that these two traditions seem, unfortunately, to ignore each other.
Finally, another line of research that also should be mentioned here in respect to the relationship between semiotics and cognition is that represented by some recent developments in what is usually referred to as dynamic semiotics or semio-cognitive morphodynamics (Petitot 1985, 1992; Brandt 1994, 1995). The basic hypothesis of these works is that there exist syntactico-semantic infrastructures of topological and dynamic nature that constitute universals underlying language, perception, and action. Such a line of thought is consistent with the basic tenets of much work being done in COGNITIVE LINGUISTICS today (see, for instance, Langacker 1987). Dynamic semiotics aims, however, not only to individuate the schematic structures that underlie meaning in different types of systems but also to model them through a qualitative mathematics based on the catastrophe theory developed by the French mathematician René Thom (1975), where states, events, acts, and processes are understood formally in terms of force topologies and where form is dynamically represented as based on opposing forces. Dynamic models can therefore be of help in reformulating the findings of structural semantics in dynamic terms, in a way that is very close to approaches used in more recent work in cognitive semantics, where the role and function of image schema and force-dynamic schema have been investigated (see Talmy 1988; Lakoff and Johnson 1980).
Brandt, P. A. (1994). Dynamique du sens. Aarhus: Aarhus University Press.
Brandt, P. A. (1995). Morphologies of Meaning. Aarhus: Aarhus University Press.
Charniak, E., and D. McDermott. (1985). Introduction to Artificial Intelligence. Reading, MA: Addison-Wesley.
Eco, U. (1976). A Theory of Semiotics. Bloomington, IN: Indiana University Press.
Eco, U. (1979). The Role of the Reader. Bloomington, IN: Indiana University Press.
Eco, U. (1984). Semiotics and the Philosophy of Language. Bloomington, IN: Indiana University Press.
Eco, U. (1990). The Limits of Interpretation. Bloomington, IN: Indiana University Press.
Greimas, A. (1970). Du Sens. Paris: Seuil.
Greimas, A. (1983). Du Sens 2. Paris: Seuil.
Hobbs, J. (1992). Metaphor and abduction. In A. Ortony, J. Slack, and O. Stock, Eds., Communication from an Artificial Intelligence Perspective: Theoretical and Applied Issues. New York: Springer, pp. 35-58.
Hobbs, J., and M. Kameyama. (1990). Translation by abduction. In H. Karlgren, Ed., Proceedings of the Thirteenth International Conference on Computational Linguistics. Helsinki, pp. 155-161.
Hobbs, J., M. Stickel, D. Appelt, and P. Martin. (1993). Interpretation as abduction. Artificial Intelligence Journal 63(1-2): 69 - 142.
Josephson, J. R., and S. G. Josephson, Eds. (1994). Abductive Inference: Computation, Philosophy, Technology. Cambridge: Cambridge University Press.
Lakoff, G., and M. Johnson. (1980). Metaphors We Live By. Chicago: University of Chicago Press.
Langacker, R. (1987). Foundations of Cognitive Grammar. Stanford, CA: Stanford University Press.
Peirce, C. S. (1931-58). Collected Papers. Cambridge, MA: Harvard University Press.
Petitot, J. (1985). Morphogenèse du Sens: Pour un Schématisme de la Structure. Paris: P. U. F.
Petitot, J. (1992). Physique du Sens. Paris: Editions du CNRS.
Propp, V. (1928). Morfologija skazki. Leningrad.
Rumelhart, D. E. (1975). Notes on a schema for stories. In D. G. Bobrow and A. Collins, Eds., Representation and Understanding: Studies in Cognitive Science. New York: Academic Press.
Rumelhart, D. E. (1980). On evaluating story grammars. Cognitive Science 4:313-316.
Saussure, F. de. (1906-11). Cours de linguistique générale. Lausanne-Paris: Payot.
Schank, R. (1982). Reading and Understanding. Hillsdale, NJ: Erlbaum.
Talmy, L. (1988). Force dynamics in language and cognition. Cognitive Science 12:49-100.
Thom, R. (1975). Structural Stability and Morphogenesis: An Outline of a General Theory of Models. Amsterdam: Benjamins.
Thorndyke, P. W. (1977). Cognitive structures in comprehension and memory of narrative discourse. Cognitive Psychology 9:77-110.
Van Dijk, T. A., and W. Kintsch. (1983). Strategies in Discourse Comprehension. New York: Academic Press .
Copyright © 1999 Massachusetts Institute of Technology