**2. Requirements for scientific explanations**

To answer the question about the relationship between the statistics producing research and the scientific explanation, it is necessary to first analyze how explanations are formulated in social sciences. The emergence of natural and social sciences is closely linked to a fundamental change in the historical development of thought that occurred at the beginning of the modern era. This change is particularly characterized by a break in the logic applied to understand and explain the phenomena of the world, or, in a given case, the world itself. Therefore, we must clarify what "explaining" means in the modern sense. From this surges every approach for society's theory construction as second-order observation. This radical transformation in the scheme or logic of thought, which may well be understood as a cognitive paradigm, should be understood as the basic condition for the development of any explanation or theory that claims scientific validity.

#### **2.1. The desistance of absolutes as explanation**

"The ancients" thought every phenomenon through until they recognized "one clear terminus" [6]. Nowadays, scientists agree that any hypothesis must desist from claiming an absolute origin that provides the ultimate argument. This is the result of the world's process of secularization that accompanied the three modernity revolutions: the revolutions in the natural sciences, in politics, and in economics.

From the epistemological point of view, this means that it is not possible to understand what exists as the product of an absolute origin and that all questions regarding its cause are answered by resorting to a single ultimate origin. Günter Dux explains this way of proceeding: "The absolute in premodern thinking, where spirit predominates, was absolute because it contained what was to come out of it as substance. The way of explaining consisted of attributing the explanation to the absolute to make it rise from there as an emanation." [7]. The explanatory potential of resorting to the absolute resided in the impossibility to question it. It was impossible to explain something within this logic, since the absolute origin subtracted itself from the question about its own cause. In this regard, Dux points out: "In a world that turned radically secular, in which… nothing that subtracts itself from a set of conditions can be found, the constructively achieved worlds – and their logics – must be explained through the conditions that made it be" [7]. Then, it's not that a certain explanation lost its convincing power along the history of thought, but the substitution of an explaining logic for another, of a paradigm for another. According to the old logic, explaining meant to part from the concrete object or phenomenon—natural or social—and to assign it an ultimate origin to claim it as its cause. This way of proceeding forced us to think of the cause as an origin analogous to what would emerge from it. Cause and effect were (partially) identical.

The attempt to find explanations through this process lost strength for the same question about the origin that could not be answered by resorting to an absolute origin anymore, not without falling into an endless return.

Accordingly, every form of explanation where the explainer is present in the initial explanation from which it emerges is unsatisfactory, for example, in the explanation of a social phenomenon where the said phenomenon is already embedded in the actors' wills, or in the explanation of specific social phenomena that are already in the own qualities of that society [4].

Modern thinking imposes then a first requirement for sociological research: it is not feasible to opt for a research strategy where the concept is defined beforehand to apply it then to the phenomenon being researched; we would rather have to work the other way, we must perform an observation and then specify how it developed.
