Do We Need Mechanisms in the Social Sciences?

Citation: Reiss, Julian (2007) Do We Need Mechanisms in the Social Sciences?. Philosophy of the Social Sciences 37: 163-184 (RSS)
Internet Archive Scholar (search for fulltext): Do We Need Mechanisms in the Social Sciences?
Tagged:

Summary

Reiss discussed some problems of causal mechanism, which is called new mechanist perspective in this paper, and stated that causal mechanism is not the best methodology to give explanation in social science because it ignored non-explanatory aims of social science. First, he mentioned that mechanist perspective is based on three fundamental arguments: 1. theoretical explanation is the aim of social science, 2. the subject of explanation is empirical phenomena, 3. Empirical phenomena is explained in terms of the causal mechanisms.

However, Reiss pointed out that the aim of social science is not just theoretical explanation, but the non-explanatory goals, such as description, prediction and control. For example, rate of unemployment or inflation rate is descriptive measurement for specific phenomena, but it plays an important role for understanding the economy of a state and improving the situation. In addition, predicting or controlling social phenomena is the main goal in social science. Based on the aims of description and prediction in social science, Reiss argued that causal mechanistic model cannot predict social phenomenon well. For example, the econometric model explains events based on data-generating structure, but the structure is easy to break. This means that in real world in addition to causal factor, there are non-causal variables which may also influence the outcome.

Moreover, causal mechanistic model cannot control social phenomenon, because the relationship between causal mechanism and outcome is unstable in social science. For instance, a policy was introduced in order to control or change the target. The real control involves in the stable relationship between the policy variable and the target rather than the mechanism of the relationship. In conclusion, Reiss stated that in social science, methodology and the philosophy of social science should focuses more on the non-explanatory aims, such as measurement, prediction, and policy evaluation, instead of theoretical explanation.