From system thinking perspective, addiction can be described pretty well with a fairly simple model of a few feedback loops. Regardless of the exact feedback st

From system thinking perspective, addiction can be described pretty well with a fairly simple model of a few feedback loops. Regardless of the exact feedback structure, the mechanisms are both social and bilogical, human biology will get hurt in the process and the entire thing is likely to behave chaotically. This is why you can argue it's a disease (because you get biological causes and effects) and that it's not (because AA works by providing a balancing loop designed to be tolerant to substance abuse). And this is why you probably won't get to a clean medical diagnosis - a very similar set of factors can lead to a wide range of outcomes. Unless addiction is seen as a systemic issue, I can't really see how it could be better understood beyond describing less and less significant factors in more and more detail.