Abstract
What makes a concept what it is? What distinguishes, for example, my concept of a tiger from my concept of a cat? Is it what the concept refers to, or is it the subject's conception of it? According to the referentialist thesis, the first answer is the right one. Of course, these two things - reference and conception - normally go hand in hand. My concept of tiger refers to tigers AND, for this very reason, its content includes the characteristics of tigers: the objective differences, at the level of the concept's reference, between tigers and cats are reflected in the conceptions associated respectively with these two concepts. But while it's true that the two things - reference and conception - normally go hand in hand, they may not, insofar as the conception may be wrong, or even massively wrong. This is the case in the example of robot cats mentioned in the first lecture. This type of example suggests that the concept is individualized by its reference, determined by environmental factors, and not by the conception of the subject deploying the concept.
The advantage of the referentialist viewpoint, according to which a concept is individualized by its reference, is that it guarantees the stability of the concept despite the variability of its content. We illustrate this stability with a well-known example from the work of Tyler Burge: the example of the concept of arthritis, which an ignorant patient inherits from his doctor despite the radical difference in their respective conceptions.
The referentialist thesis implies that there cannot be two distinct concepts referring to the same thing (since referring to the same thing would automatically make these concepts the same concept). Yet the possibility of a plurality of distinct concept-types referring to the same thing is one of Gottlob Frege's well-known assertions. To account for the examples Frege invokes (the "Fregean cases" I've discussed in detail in my 2019-2021 lectures), we can try to relativize the thesis that there can't be two distinct concept-types referring to the same thing. This thesis would only apply to "atomic" concepts, which would be well individualized by their reference, unlike compound concepts, which are individualized by their constituents and the way they are combined. But the referentialist thesis cannot be maintained, even in this limited form: it is indeed easy to produce examples of distinct atomic concepts referring to the same thing. The existence of such cases shows that concepts cannot be individualized by reference alone: what Frege calls the mode of presentation also counts.
