Alternatively the GNOWSYS kernel can also be viewed in the form of three semantically significant levels: metatype, type and token levels. All the two type level constructors help a knowledge engineer construct (specify) a model (the structure of the cognitive system), The epistemic value of this modeling type layer is consistency, since these are conceptual in nature. All the token level constructors help store the data about a given situation. The epistemic value of this token layer is truth. Though contradictory propositions can be stored in the system at the ground layer, while building the structures additionally and explicitly specified semantic constraints prevent contradiction. The connection between the levels is made by the inference layer, which does the validity checking, to deduce consequences that are not fed into the knowledge base explicitly, to hypothesize either abductively or by other ampliative means. This will add new inferred elements, both propositions and concepts, to the system. The inference layer is optionally kept out of the system, for it should be possible to conduct semantic discourse about an instantiated structure. The top metatype level can serve the purpose of expressing and storing what is now popularly known as upper ontologies and the relations among them[17,18].