SI: A Note on Error Typologies and the Possibility of Gaining insight in Mental Processes

Abstract

Forming the theory of SI would probably be one of the most difficult and daring projects in the history of the human sciences. The SI process is so versatile that the number of variables involved is presumably infinite and, what is worse, some variable types may belong to incompatible logical classes, e.g. 'encyclopedic knowledge' does not belong to the same logical category as 'linguistic knowledge', but no one would deny that both are crucially involved in SI. Hence, a unified theory of SI is not likely to be viable and a "modular approach" to the field seems to be advisable. By "modular approach" I mean the methodological necessity of splitting up an area into smaller units which can be studied in relative isolation from other areas of the same field, even though from a superficial point of view the modules may seem to form an inextricable whole. Thus, "module" is not an explicated term as yet, and the question "What modules are involved in SI is only answerable in terms of a theory which covers as much of the SI process as is possible at any given time. What I would like to suggest, however, is that formal aspects of language, syntax in particular, may play a more important role in SI than is commonly assumed, i.e. I would like to propose that any adequate theory of SI must include a formal module". The general view seems to be that formal aspects of language can safely be either neglected in SI research or, at best, given relatively minor attention due to certain outdated ideas about language relationships. An example is given in sect. II. SI may be looked at from (at least) two radically different angles ; the first concerns the conditions of possibility of SI, or, more plainly, questions like "What condi- tions must be fulfilled in order for SI to be at all possible ?". The second, equally important, angle focusses on applied and possible strategies in SI, e.g. techniques employed when interpreting from verb final languages into verb second or verb initial languages. Of course, it is not inconceivable that the problems studied under each heading are somehow related, perhaps even closely so. But that is another empirical issue still to be settled.