Thinking Through Things: Analytic or Heuristic?

In my thesis, I am struggling to develop a clear articulation of what I am trying to do with the objects that I'm studying. I have been resistant to identifying my approach with a single theory, though I can cite several texts, concepts and scholars who have inspired my approach, or who inform what I am doing. As I unpack the contexts surrounding certain items now in museum collections, I do wonder -- am I implicitly using the objects as portals into a time and place, or are they somehow more central? Is there something about using objects as sources that gives my analysis something extra, that could not be gleaned through documentary evidence?

 In the introduction to Thinking Through Things: Theorizing Artefacts Ethnographically (2007), the editors Amiria Henare, Martin Holbraad and Sari Wastell describe an emerging way of approaching material culture they have found in their discipline, anthropology. Now, there's a lot going on in this intro. And I'm only on page 9.

 One part that has stuck with me is how they differentiate between two ways of approaching things:
1) Thing-as-Analytic 
2) Thing-as-Heuristic

Skipping over their discussion on why the word "thing" is more useful than "object" or "material culture" (basically, as they say, it carries "less theoretical baggage" than other terms), I will jump to how these two approaches differ.

Thing-as-Analytic - This is when a thing is treated as a point of entry into understanding the internal logic of its users, or the culture from which it came. So you try to understand why, say, wampum became so valued in Onkwehonwe communities back in the day. Could it have had something to do with its shiny surface, which lent it an aura of power or value? We might look into how shiny things, the colours white and purple, or shells, carry value in Onkwehonwe oral tradition, and then re-construct how it might be that people came to believe that wampum was valuable.

Thing-as-Heuristic - This is when we try to understand the thing on its own terms, and then build a theory from that. It is almost backwards from the other way. We accept the valuation that wampum is powerful and valuable, and then go from there. From this starting point, we then try to build categories that make sense, rather than parsing the thing and its significance in ways that fit with categories or theories that we might already believe, or like to use.

 Or, in their own words, on page 7: "So the distinction between 'things-as-analytics' versus 'things-as-heuristics' points toward the absolute productivity of non-definition -- towards a new impulse within anthropology to move beyond the development of ever more nuanced filters through which to pass phenomena, through to engagements with things as conduits for concept production."