When Human Behavior Enters the Equation

ABSTRACT

In political science, there is a profound tension between those who argue that the prediction of human behavior is possible and those who disagree. On the one hand, experimental methods may be able to detect behavioral patterns. On the other hand, events, the evolution of institutions, and individual behavior itself may in many cases be too historically contingent to be reliably predicted. Methodological pluralism, and a greater degree of openness to mere historical understanding, would therefore seem to be called for.

Keywords:

• behavioral economics
• contingency
• nonlinearity
• prediction
• scientific pluralism
• thick data

Notes

1 See also Taleb [2007] 2010 for an interesting essay on so-called black swan events. These are highly improbable events (outliers) that have a tremendous impact on the way the world is turning. Fully in line with Streeck and Morgenthau, Taleb notes that “the inability to predict outliers implies the inability to predict the course of history, given the share of these events in the dynamics of events” (ibid., xxiv).

2 For a critical discussion of Andersen’s remarks, see Kaspersen 2014.

3 For an interesting and general critique of this approach, see Henrich et al. (2010).

4 What Madsbjerg is concerned with goes beyond the notion of “ecological validity” in the social science methodology literature, which mainly stresses that experimental settings ought to reflect real-life conditions of the participants as closely as possible (Bracht and Glass 1968). This would merely lead the researcher’s attention towards field experiments and so-called natural experiments (Shadish et. al 2001; Gerber et al. 2008)

5 As the physicist Neil deGrasse Tyson once said, “In science, when human behavior enters the equation, things go nonlinear. That’s why physics is easy and sociology is hard” (quoted in Madsbjerg 2017, x).