on dubious business.

W.V. Quine (1974, p. 187f) says that quantifying into intensional contexts is dubious business, and he offers the following as a case in point. Ralph suspects that the man wearing a green hat is a spy. Though they have never met, Ralph has also heard a great deal of good about a fellow by the name of Bernard. Unbeknownst to Ralph, the man in the green hat and Bernard are one and the same person. We ask Ralph about whether he thinks Bernard is a spy, and Ralph replies, “Bernard is no spy!” Thus, it appears that Ralph has contradictory beliefs: Ralph believes that Bernard is a spy and Ralph believes that Bernard is not a spy.

            Wait, what? This isn’t obvious. It seems to me that Ralph believes that someone who wears a green hat is a spy: (let ‘r’ abbreviate ‘Ralph’ and ‘(∃x)B_r(Sx • Gx)’ abbreviate ‘there is something such that Ralph believes it is a spy and it wears a green hat’. Thus:

(1)       (∃x)B_r(Sx • Gx)

Ralph also believes that Bernard (b) is not a spy:

(2)       (∃x)B_r(x = b • ~Sx)

(1) and (2) quantify into Ralph’s beliefs, but their conjunction does not imply that Ralph has contradictory beliefs.  

References

Quine, W.V. (1974) “Quantifiers and Propositional Attitudes”, reprinted in The Ways of Paradox and Other Essays (revised ed.), Cambridge: Harvard University Press.