Abstract
The article tries to answer the following question: what is the most promising epistemological strategy if my objective is the construction of a theory which gives me the opportunity to decrease the risk of getting to what is actually absolute, that is, to irreversible negative actions (irreversible as a theory might not be, but as an action often is)? The answer proposed is a form of epistemological dualism which means that I metaphysically believe (that is, I programmatically and systematically believe, without certainly knowing it) that the epistemological relationship between any theory and any reality is dualistic. More specifically, I metaphysically believe that the epistemological relationship between any theory and any reality is not saturated: in any theory there is an ideal error, because there is no theory which is totally saturated by reality, and any reality can actualize the ideal error, because there is no reality which is totally saturated by theory