One technique I have used for dealing with concepts that have many, many meanings, is to:
1) deal with subsets of the concept
2) give each of them a different name.
So, instead of arguing about "souls" one might discuss anima, numus, etc.
With chi, there are a number of sub-meanings, each of which can be discussed separately without having to get into an argument about "you have the wrong definition for chi" or without having to accept the presuppositions most definitions bring with them.
I am reminded of Tai Fan Chen's comment about Chi (before he pretty much left the martial arts to sell herbs). He felt it was interesting, but irrelevant to most training. (Note, in this context, calling "chi" by the name "chi" is appropriate even if we are discussing "energy" [??still not defined enough] or biomagnetic/bioelectric fields or similar things because I am side tracking from "what is it" to "is it relevant for most people").
Definitional modes are especially important in areas where most definitions are developed from a combination of oral tradition (my instructor said), myths (other students said) and personal experience (I felt) and thus are in many ways personal gnostic moments.
Chi is one of those areas, since most definitions are created from a combination of experiences and traditions.
In comparison, consider discussions of "auras" -- and the question of "well, what kind of texture did that aura have?" interjected into a conversation about colors of auras (subsuming the question of whether or not auras exist, whether or not they are related to "chi" and other definitions, etc.).
And realize that this discussion is different than just changing the word choice (i.e. saying Nishiyama has great personal force, instead of saying "Nishiyama's chi was so great I could feel it across the room when he knocked Brad Webb down from a distance" -- using the term force instead of chi doesn't add much or change the ambiguity).