When Great Transformations Ebb Out: How to Study Discontinuance of Diffusion

Great transformations refer to political changes that change the political conditions in fundamental ways, as regime changes, political conflicts and political crises. As previous research has shown, these great transformations tend to spread between societies. One mechanism that provides the spread of great transformations is diffusion, which means that ideas are spread between societies. Diffusion is thus seen as an explanation to why great transformations arise. Great transformations tend, however, to discontinue, which means that the ideas that the great transformations are built upon are not spread further. Even though great transformations have had significant effects on the global development, the discontinuance of great transformations has not been studied before. This paper aims hence to develop conditions that enable the study of the discontinuance of one of our great transformations: democratization.The democratization under the last century is one of the greatest transformations in contemporary politics. The number of democracies has increased significantly, as autocratic states have democratized. Previous studies have explained this development as the result of waves of democratization, where the breakdown of an autocratic regime increases the probability that autocratic regimes in neighbouring states will also fall. This spread of democratic ideas is the essence of democratic diffusion, and democratic diffusion is therefore able to explain one of our greatest transformations. The diffusion of democracy has, however, discontinued unexpectedly under the waves of democratization. As mentioned before, the discontinuance of great transformations has not been studied in previous studies, which also means that the discontinuance of democratic diffusion has not been explained. One reason why explanations to the discontinuance of democratic diffusion have not been examined before is due to a lack of concepts and measurements that would make it possible to empirically study the discontinuance of democratic diffusion, and by that enable the explanation of the discontinuance of democratic diffusion. As a first step to explain why democratic diffusion discontinues, this paper presents how the discontinuance of democratic diffusion can be conceptualized and measured.The aim with this paper is to develop conceptual and methodological conditions that makes it possible to study the discontinuance of democratic diffusion. A first objective is to define and specify the concept of the discontinuance of democratic diffusion. The paper specifies how democratic diffusion is assumed to discontinue based on conceptual discussions about diffusion. Based on the general model of diffusion by Rogers (1995:5), which defines diffusion as “the process by which an innovation is communicated through certain channels over time among the members of a social system”, four process components are identified as necessary for diffusion: communication channels, actors, openness, and time. The paper presents a model that explains how these necessary components for diffusion are related to the discontinuance of democratic diffusion. The second objective is to develop methods that identify the discontinuance of democratic diffusion. This paper presents three methods that identifies cases of discontinuance of democratic diffusion in different ways. These methods are developed based on how democratic diffusion has been measured in previous studies (Brinks & Coppedge 2006; Gleditsch & Ward 2006; O’Loughlin et al. 1998; Wejnert 2005; 2014). The first method identifies discontinuance of democratic diffusion as deviation from the political regimes in the region. When a state in a region with a majority of democratic states has an autocratic regime, it is regarded as a case of discontinuance of democratic diffusion. The second method identifies cases of discontinuance of democratic diffusion when the cases have autocratic regimes, but are, according to regression analyses, expected to have democratic regimes as an effect of the share of democracies in the region. The third method is also based on the outcome from regression analyses, but includes control factors. This method identifies cases of discontinuance of democratic diffusion when democratic diffusion according to the regression analyses is expected to have a decisive effect on the probability for democratic regime in a state, but the state nevertheless has an autocratic regime. In this paper, these methods are further developed, and used in empirical analyses to illustrate the possibilities to identify where (states) and when (years) the democratic diffusion has discontinued, which enables comparative studies of the discontinuance of democratic diffusion. The paper will therefore present a comparative picture of where and when one of our greatest transformations of the last century has stopped.