The attack on content-externalism on the grounds that it is incompatible with content-transparency has been much debated over the decades. The first line of argumentation claims that content-externalism undermines the a priori authoritative self-knowledge or the authority of the first-person. According to content-externalism, the content of mental states is determined, in part, by the non-intentional relations that the subject has with her/his environment. Given this, the content of the subject's mental state could change without her/his knowledge under the assumption that the subject undergoes undetectable environmental changes, say, from Earth to Twin-Earth and vice-versa. Thus, if content-externalism is right, one can never authoritatively know a priori whether one is now thinking about water or twater.According to Burge, however, content-externalism is not a threat to cogito-like thoughts, that is, second-order contemporaneous thoughts about one's first-order thoughts. Without the need for any empirical enquiry or any epistemic self-justification, the introspective second-order thought "Inherits," so to speak, the content of the first-order thought by embedding it. If some Earthian's first-order thought "water quenches thirst" refers to H 2 O rather than to XYZ, his introspective second-order thought "I know that I am thinking that water quenches thirst" also refers to H 2 O (instead of XYZ) because the first-order thought is embedded in the introspective second-order thought.
AbstractIn this paper I take side on externalist incompatibilism. However, I intend to radicalize the position. First, based on my criticism of Burge's anaphoric proposal, I argue that there is no reasoning-transparency: the reasoner is blind to the reasoning he is performing. Second, assuming a global form of content-externalism, I argue that "in the head" are only logical and formal abilities. That is what I call "bite the bullet and swallow it too." K E Y W O R D S content-externalism, content-transparency, dangerous reasonings, rationality failure, safe reasoning | 193 de SÁ PeReIRA Yet, even if Burge's self-verification account of cogito-like thoughts is convincing, two additional key questions still remain open. The first question is whether under content-externalism the content can be preserved in slowswitching cases (introduced by Burge in the 1988 literature). Suppose that content-transparency fails and I am an accidental tourist that travels from Earth to Twin-Earth when I remember that Pavarotti once floated on Lake Taupo. What is the content of my memory? The second is a direct development of the first. Suppose now that the original content is preserved and I remember that the Earthian Pavarotti floated on the Earthian Lake Taupo. Yet, assuming assimilation to the dialect of twin Earthians, whenever I think from now on of Pavarotti I refer to Twin-Pavarotti rather than to the Earthian Pavarotti. The problem is that if the content is preserved I might from this infer that Twin-Pavarotti got wet, thereby making my inference in...