Folk-psychological belief ascription and idiolectal translations

I wrote a paper on Quinean empathy qua idiolectal translation manual selection tool recently, and I think that I may be onto something, but Falk seemed to think differently. I submit to you some preliminary thoughts about this topic, and any feedback would be greatly appreciated.
I think that using empathy as a way to help select a (so-to-speak) idiolectal translation manual--being a "manual" that allows one to properly take words or phrases used in the lexical public language and effectively translate them into the idiolect of the original speaker--that will allow us to avoid what seems to be a common problem in the philosophy of language; namely the problem of identity when a speaker is unaware that two terms (or proper names) refer to the same object, makes contradictory statements (not knowing of the contradiction) and a third party seeks to report those beliefs having full knowledge of the co-reference and thus the third party is doomed to contradiction in a way that the original speaker is not. This should bring to mind Frege's "morning star-evening star," or for those of you in Falk's D&B class, his whole book.
Anyway, that's the problem... Here's what I think one can do:

I attempt to show that using empathy may provide us a sort of “back-door” into understanding the idiolect of another person. If we are able to detect idiosyncratic tendencies of another’s language (both verbal and non-verbal) then it may lend credence to our abilities to, so to speak, choose the correct idolectal translation guide and better understand another person. If this is possible, it will provide us a basis to maneuver around a seeming paradox in ascribing rational-but-false beliefs to another person; namely that we ultimately end in self-contradiction. This is a familiar problem to the philosophy of mind, but is posed decisively in Stephen Schiffer’s article, “A Problem for a Direct-Reference Theory of Belief Reports.” The confutation of this problem may not be, necessarily a solution, as much as a way to show that it really is not a problem after all.
I will not provide the entirety of the paper (as that would be a ludicrous abuse of this venue) but I will provide some key definitions to help you help me, I hope.
Empathy, as I shall use it, I mean to be the detection of idiosyncratic behavior (both verbal and non-verbal) and generating the ability to project ourselves into the position of the speaker-actor. Following Falk, I will refer specifically to empathy as a way to understand that attitude, as well as, behavior of another. Empathy, while not being entirely conscious at all times, is not an ability that we have with everyone, all the time. It is, however, imperative to learn a public language, and so is something that we have naturally.
(This is much like the definition outlined by Quine in Pursuit of Truth, §16 (pg. 43) where he states, “We judge what counts as witnessing the occasion… by projecting ourselves into the witness’s position.”) It is important to note, however, that I do not limit my use of empathy to observation sentences in terms of concrete objects; I extend this also to empathy of attitudes and behaviors.
Idiolect I take to be a fairly common term denoting a pseudo-public language that is personal to an individual. By ‘pseudo-public’ I mean that it is a personal reflection of a lexical public language that I happen to prescribe to but that is not exclusive to me. I borrow this notion from Alexander George who described idiolects as, “not something essentially private: you and I could have the same idiolect… An idiolect is idiosyncratic in being that about which a particular speaker at a particular time has some knowledge and not in being something about which only one person could have knowledge.”
In his paper, “Whose Language Is It Anyway? Some Notes on Idiolects,” George explains the idiolect in a way I find quite congenial. He says, “My idiolect is an object about which I have beliefs, in particular those beliefs I possess qua linguistic being. In this respect, one’s idiolect can be compared, for example, to the natural number series… These are abstract structures about which one can have beliefs, some true, some false.”
translation manual is a useful tool that everyone has at their disposal. What I mean by this is that with each person subscribing to an idiolect, there must be some way to get at how the idiolect applies to the public language. Think about this manual in the same way that one would think about one for a foreign language.
Alright. Go to town.