Abstract
Passionate debates often occur in the scientific community concerning the relevance of new theories which compete with classical descriptions. To discriminate between them, a set of epistemological tools may be used, the most familiar being Ockham's razor. This criterion is used as a procedure for selecting between theories with different degrees of complexity. Given an observation, it selects the simplest non-falsified theory which accounts for it. It has been quite widely used, in contexts from general epistemological considerations to specific examples such as the study of turbulence. Using the example of fractal tools for the study of turbulence, I aim to demonstrate that the criterion of Ockham's razor is very stringent and may lead to the rejection of potentially fruitful new approaches. Hence I propose a new epistemological criterion, which does not reject theories solely on the grounds of their complexity. It is founded on so called PaRDeS hermeneutics, developed by Jewish exegetists for the study of the Bible and other religious texts such as the Talmud. This approach, involving four levels of understanding of increasing complexity, allows the definition of a new criterion of ‘scientific worthiness’ for descriptive/explicative theories. I conclude by discussing how Ockham's razor and the PaRDeS method relate to the notion of hypercredulity.