Emilio Caceres

It is commonly assumed that the concept of mechanism is a keytool for the scientific understanding of observable phenomena. However, there is no single definition of mechanism in the current philosophy of science. In fact, philosophers have developed several characterizations of what seemed to be a clear intuitive concept for scientists. In this paper, I will analyze these philosophical conceptions of mechanism, highlighting their problematic aspects and proposing a new mechanistic approach based on the idea that the pertinent levels of organization for a mechanistic explanation can be identified with intervals of quasi-decompositionality. I argue that this approach allows us to consider that activities are directly derived from the entities’ structure. Consequently, a mechanistic explanation implies an arbitrary but not capricious choice of an organizational level. According to this approach, inter-level causation is merely apparent and there is no place for emergent properties.

The notion of emergence has accompanied philosophy of science since the late XIX century, claiming that in some systems there are properties in certain levels that cannot be deduced from properties of their components as seen in more fundamental levels. Throughout the XX century, emergence has been characterized by four pillars: unpredictability, novelty, restriction and downward causation. These four pillars have been related to the assumption of a hierarchical order of reality in different levels of organization. In this paper, we show that it is possible to explain the nature of the emergent properties through the nearly-decomposability criterion introduced by Herbert Simon.