Facts, events and their identity conditions

At one time I believed the first two identity statements to be false but I now believe all three to be true (the first, pretend-true, of course). Part of the aim of this paper is to explain and justify all three judgments. In a word, I shall be partly concerned with identity conditions for events. If I spend a lot of time on identity conditions for facts it's just because believe the former are correctly explicated by the latter. Davidson3 has offered the following identity conditions for events: Events are identical if and only if they have exactly the same causes and effects. (p. 231) He goes on to add that although the definition may seem to have an air of circularity about it, it is not a formal circularity, since "no identities appear on the right hand side." But this observation misses the point of identity conditions. The usual criterion for classes is: Two first order classes are identical if and only if every individual which is a member of one class is a member of the other. The right hand side is logically equivalent to: Every individual which is a member of one class is identical with some individual which is a member of the other. This is a bit fancy but would be an unobjectionable statement of identity conditions for classes. The reason is that although it contains the identity sign, what is in question is the identity of members, not of classes themselves. Contrast this with the following: Two first-order classes are identical if every subclass of one class is a sub-class of the other. This is true