Н. В. Головко The paper aims to look for the arguments against the tenet that the inference to the best explanation that connects the scientific realism and the success of scientific knowledge, and that is coined as the Main argument for the scientific realism, is circular for it contains the affirming the consequent fallacy (L. Laudan, B. Van Fraassen, etc.) A special case of the «Main argument» is considered - the inference to the existence of unobservables on the basis of the causal explanation of the phenomenon. Such an argument is the subject of the so-called Hitchcock Dilemma, which requires that for the inference to be sound it must be accompanied by an independent argument in favor of the existence of a «cause». The adoption of a special kind of ontology - in our case, it is one of the possible interpretations of D.Ross's rainforest realism, makes it possible to construct the required «independent argument». A pattern is not only defined by the projection relation, first of all, in order to play a functionally significant and successful role in explaining phenomena within a given area, but it also «must contain information about another pattern» (J. Ladyman), which is the exact the «independent condition» we are looking for to prove the soundness of the «Main argument».
Научная статья 
К. Хичкок