Abstract
According to the classical view, the content of a concept is its associated definition. To master or possess a concept is to have at least tacit knowledge of the definition in question. So, of course, the content of a concept determines its extension (everything that satisfies the definition falls under the concept), but it also makes it possible to individualize the concept, i.e. to distinguish it from other concepts (if you change the definition, you automatically change the concept).
There are two variants of this theory, and both run into a problem. 1st variation : through the definitions associated with them, concepts refer to each other in a more or less circular fashion, like a dictionary that defines A in terms of B, B in terms of C, and C in terms of A. Problem: in this holistic version of the theory, it's not clear how we get out of the system of concepts (which operates as a closed circuit, so to speak) to anchor it in the reality that the concepts are supposed to represent. 2nd variation: Some concepts are defined in terms of other concepts, which can themselves be defined in terms of still other concepts, but at the end of the chain are concepts that are not themselves defined in terms of other concepts (and at whose level the system is anchored). Problem: since these concepts have no definition, classical design is inapplicable to them, and we are inevitably led to look for an alternative design.
In the conception that has replaced the classical one, and which has been elaborated in various forms by philosophers and psychologists (stereotype theory, prototype theory, exemplar theory...), the content of a concept is not a definition, i.e. a set of necessary and sufficient conditions for something to fall under the concept, but an often contingent and superficial representation of the reality to which the concept relates. In most cases, such a representation makes it possible to identify instances of the concept in the environment, but by exploiting contingent facts so that the representation, unlike a proper definition, does not deliver necessary and sufficient conditions for the concept's application.
The thesis that the content of a concept includes empirical knowledge about the things that fall under the concept implies the rejection of the thesis that the content of a concept determines its extension. Far from determining extension, it is the content of the concept that is to a large extent a function of the factual properties of the things that fall under the concept. In this respect, the content of concepts resembles the notions we associate with proper nouns. Drawing on this analogy with proper nouns, we can decide to include in the content of a concept not only the "stereotype" (the typical properties of the things that instantiate the concept) but also the totality of encyclopedic knowledge about these things. A concept is then conceived as a mental file including everything we know or think we know about the reality to which the concept refers.
