On the History of Statistics and Policy-Making

One of the main tasks of statistics is to collect data on an issue of interest, to make regional, factual or temporal comparisons based on these data and thus to prepare decisions. The way in which this task is carried out varies depending on the area of investigation. It essentially depends on whether the data collection can be done in controlled experiments, in which all influencing factors interfering with the intended comparison can be eliminated by appropriate experimental design, or whether the statistician only has the role of a chronicler who registers the data without being able to intervene in their development process. In demographic, social and economic science—to name but a few fields—it is almost always the latter case. If we restrict ourselves to official statistics (as a public good), their character becomes particularly apparent in the form of this registration. As already reflected in the term, (official) statistics are closely related to the state (it is about describing the state of the state) and are thus subject to the special features of state structures and state action. Official statistics is a tool to hold up a mirror to society—but also an instrument for those who can manipulate this mirror. For if the mirror does not show what it is supposed to show, different strategies can be chosen. Those in power have always been interested in controlling their understanding of reality. It is not for nothing that statistics got its name as the science of the state. The objective was and is the collection and provision of complete, comprehensive, consistent and timely information on the situation and the development of a state. Accordingly, official statistics has been and continues to be misused by many autocrats to exercise their power. From communist East Germany to present-day China, rulers who were or are interested in monitoring their populations usually used and still use different methods of collecting data than independent statistical offices in modern democracies. In any case, history shows that dictators often either have negligible interest in collecting sound statistics or have little ability to collect them accurately. This instrumental character of official statistics has often been its undoing, especially in the context of the censuses that have been conducted at regular intervals for a long time. As an instrument of power, these censuses have long served not only to count people but also to register their moral behaviour. Their history is often marked by mistrust. A recent example is the postponement of the census in the Federal Republic of Germany scheduled for 27 April 1983 due to an injunction by the Federal Constitutional Court, which was finally conducted on 25 May 1987.