Externalism seems to be closely linked to the "Fido"-Fido theory advocated by Bertrand Russell, according to which the content of a representation is none other than the entity represented (the reference). The opposite view, that of Frege, distinguishes two dimensions of content: meaning and reference - meaning being the way in which the reference is represented or conceived. According to Frege, meaning determines reference: reference is the entity that corresponds to the subject's representation of it, i.e., the entity that actually possesses the properties listed in the subject's "mental file". This descriptivist thesis is precisely what externalism rejects.
However, we must dissociate the meaning/reference distinction from Frege's descriptivist interpretation of it. In the " Fregean cases ", much discussed in the previous year's lecture, the same subject has two distinct files on a given individual, without realizing that it is one and the same individual. Frégé's distinction between meaning (the mode of presentation) and reference helps to account for the fact that a rational subject is prepared in such cases to attribute contradictory properties to the same object. This is a strong argument in favor of Frege's two-level semantics, as opposed to Russell's monostratal semantics. We can maintain, however, with Russell, that in the most fundamental cases what the subject is thinking about or talking about is not determined by the content of his mental representation, i.e., by the properties listed in the file: reference is established by direct contact with the object (what Russell calls acquaintance). Contrary to Russell's view, however, this idea of direct reference is compatible with the Fregéan distinction between meaning and reference, for even when an object is given directly in experience, it is given in a certain way, and it is possible not to realize that an object given in one way is the same object given to us in another way.
