The paper formulates several critical observations about the official (established by Polish legal statutes) rules of evaluation scientific publications. These rules are based on so-called parameterization, that is, ascription of points to papers and books published by scientists. The criticism is made from the point of view of the humanities and social sciences, although some assertions have a more general character. Although the author shares the opinion that evaluation of science is necessary, he points out several disadvantages of parameterization. One of them consists in the fact that the ascription of points to publications uses a relatively weak scale, that is, based on the ordering relation. This feature essentially limits the range of mathematical operations admissible to results of measuring in such a way. This situation results in conventional character of parameterization, contrary to expectations that it successfully display the real value of scientific achievements. The paper illustrates this problem by concrete examples as well as shows that parameterization makes possible pathological cases stemming from political and world-view preferences.