In this chapter and in the next we discuss what we believe is the main difference between physics and the social sciences, namely the high level of noise which spoils and often altogether forbids the measurement of many social variables. This might seem a fairly unconventional view. Usually many other differences are underscored, for instance the role of individual freedom, the diversity and complexity of human behavior and so on. More specifically it is often stated that (i) social systems are inherently more complicated in the sense that societies comprise many different kinds of agents and interactions; (ii) physical phenomena can be observed in the laboratory through experiments which can be repeated whereas social observations, it is argued, cannot be repeated. We do not mean that these differences do not exist but we believe that they are not as fundamental as may be the impression at first sight.

The first point can be readily dealt with by noting that the level of complexity depends on how much detail one wants to take into account. Seen at the molecular level, the dissolution of a piece of sugar in a cup of tea is a phenomenon of horrendous complexity in the sense that it involves many kinds of molecules and a great diversity of interactions. In contrast, for many social phenomena the precise nature of the agents is unimportant. For instance, the field of demography focuses on the numbers of birth and deaths, a perspective in which most other social features become irrelevant.