Balancing researcher's inference and member's meanings

Earlier in the quarter we spent a week or two on the idea of pursuing member's meanings and recently we've read a bit about the idea of using ethnographic writing as a way of expressing ideas in participant's own voices. This week our readings addressed an issue that I'm dealing with in my own research -- how much (or little) can a researcher infer about member's meanings?

In particular, Katz's article "From how to why ..." struck a chord with me because it made explicit a feature of ethnographic writing that has been puzzling me for years. Generally we hear about ethnography as being descriptive -- telling us what people are doing and saying and maybe telling us how some aspect of social activity progresses. Rarely do ethnography methods texts delve into the idea of ethnographic research offering explanations -- telling why people are doing/saying whatever it is that they are doing/saying. However, when I read book-length ethnographies I see numerous ways in which the author is making the transition from description to explanation.

This week Becker warns us not to do too much inferring and challenges us to try to stick with what participants are really saying and doing, but at the same time Katz tells us that ethnographers do, in fact, make inferences from their data and those inferences are often attempts to craft explanations. My puzzle at this point is how to find a balance between two things that seem paradoxical to me. How do I craft explanations based on inferences without overstepping myself, and how do I make sure that I am representing member's meanings when, as Katz points out, the members themselves may not be able to fully explain their own words or actions?

~~Melissa