Boghossian-Style Incompatibilism (Part 2)

In my previous post I outline a reductio against compatibilism. However, I believe the compatibilist has good reason reject (R4) of the reductio. Underlying (R4) is the implicit assumption that a subject can only have privileged access to the fact that she is thinking some thought θ if she is able to distinguish it, from some other thought θ*, without consulting her environment. However, the compatibilist has good (independent) grounds to reject this assumption, and with it, (R4). For example, Falvey and Owens [1994] distinguish between ‘introspective knowledge of content’ and ‘introspective knowledge of comparative content’:

(*) An individual knows the contents of his occurrent thoughts and beliefs authoritatively and directly (that is, without relying on inferences from observation of his environment). Call this kind of knowledge introspective knowledge of content.

(**) With respect to any two of his thoughts or beliefs, and individual can know authoritatively and directly (that is, without relying on inferences from his observed environment) whether or not they have the same content. Call this kind of knowledge introspective knowledge of comparative content. (pp. 109-110. Italics theirs)

Clearly, C-externailism is incompatible with (**). However, Falvey and Owens maintain that (**) does not coincide with our everyday attributions of self-knowledge. To see this, they invite us to consider a well-known debate between Benson Mates and Alonzo Church. Mates and Church disagree on how the relationship between the following two propositions should be understood:

(i) Nobody doubts that whoever believes that Mary is a physician believes that Mary is a physician.

(ii) Nobody doubts that whoever believes that Mary is a physician believes that Mary is a doctor.

Mates [1952] takes himself to be expressing two different thoughts when he utters (i) and (ii), while Church [1954] believes that the thought he expresses by (i) and (ii) are the same. Each believes that the thought he expresses when he utters either of these sentences is the thought captured by the sentence in the public language, English. Moreover, both are proficient in English, and are familiar with all of the terms expressed in (i) and (ii). Presumably, each of them knows perfectly well the thought each sentence expresses. However, one of them must obviously be wrong. Which of the two is in error is irrelevant for our present purposes. The take-home point is that it would seem highly implausible to suppose that whoever is wrong is guilty of some sort of introspective failure.

This example intimates that one can, in a very ordinary sense, know that one is thinking the thought θ and not know whether or not one is thinking the thought θ* in thinking this very thought. Moreover, introspection is of no assistance to Mates and Churchland in attempting to settle this disagreement. Falvey and Owens [1994] summarises the upshot of the Mates-Churchland dispute:

To resolve the dispute between Mates and Church one does not need better inner eyes; one needs additional information about the world we live in, the nature of our linguistic practice, the semantic theories that best represent that practice, and so on. (p. 113)

Some of this information would admittedly be logico-philosophical in nature. However, it is also clear that no consensus could be reached between the two independent of a serious empirical investigation of linguistic practice. Thus, the fact that we have a priori knowledge of our thought contents does not entail that we must be able to discriminate between them without recourse to empirical observation. We may therefore consider (**) implausible, along with (R4).