In this dissertation I draw upon Donald Davidson's theories of language and the mental in order to argue that it is impossible for an agent to hold ethical beliefs that are, when taken as a whole, radically different than one's own. This argument starts with Davidson's theory that beliefs, desires and other mental objects are theoretical entities that are attributed to speakers in order to make sense of their intentional actions. These attributions must be made in accordance with the principle of charity and, as such, it is impossible for an agent to have a set of beliefs and desires that are radically different from those of the interpreter. The trick is to show that there are limitations on how much disagreement is possible with a subset of the speaker's beliefs.

In order to do this I stress the importance of holism within Davidson's theory of meaning. One result of his holism is that a belief gets its content from its connections to other beliefs in the belief net. Hence, changes in one belief create a ripple effect which changes the contents of related beliefs. Because of this effect, in order for two people to have the same belief they must have similar beliefs throughout their belief nets local to that belief. When combined with an argument about the interrelated nature of ethical beliefs we can conclude that in order for a speaker to hold any ethical belief they must share a number of ethical beliefs with the interpreter.

What this does not illuminate is how much agreement is necessary. In order to get a handle on this I turned to the arguments presented in Philippa Foot's “Moral Beliefs” and Bernard Williams' “The Truth in Relativism.” I argue that the eccentric that Foot presents is an example of a speaker who attempts to present a theory of ethics that is simply too different to be considered an ethical theory. Finally, I consider the distinction Williams' makes between real confrontation and notational confrontation as a possible guideline for where real ethical disagreement begins to break apart.