The diffusion of protests (2)
Excerpts from the paper “Diffusion Models of Cycles of Protest as a Theory of Social Movements” n.d. by Pamela E. Oliver (University of Wisconsin) and Daniel J. Myers (University of Notre Dame), www.nd.edu/~dmyers/cbsm/vol3/olmy.pdf
This paper develops a theoretical framework for understanding social movements as interrelated sets of diffusion processes and explains why such a conception is broadly useful to scholars of social movements.
[…] We begin with the fundamental observation that in social movements, actions affect other actions: Actions are not just isolated, independent responses external economic or political conditions–rather, one action changes the likelihood of subsequent actions. That is, diffusion processes are involved. This inter-action influence has long been recognized. Tarrow’s work on cycles of protest (e.g. 1998) has long recognized these interrelations. McAdam’s work on “tactical diffusion” showed that the civil rights movement was not a steady stream, but a series of bursts of action each driven by a tactical innovation: bus boycotts, freedom rides, sit-ins, demonstrations, and riots (1983). Many scholars have also noted the many ways that protest actions cannot be understood in isolation, but rather need to be viewed as interactions with the police and other social control forces, particularly as the police learn more effective methods of repression over time. Protest actions obviously interact as well with social policy changes and political speech-making (what we often call “elite support”). And, of course, over time one social movement affects another, as tactics and frames diffuse and produce the effects that Meyer and Whittier (1994) call “movement spillover.” The civil rights demonstrations and marches of the early 1960s not only led to civil rights legislation, but indirectly fostered the increased militancy and anger of Blacks and the elite responsiveness which contributed to the wave of black urban riots. The Black movement, in turn, was a direct inspiration for activists who explicitly studied the histories and writings of Black movement activist, including for example the Chicanos who founded La Raza (García 1989) and early feminists (Evans 1980).
[…] In short, diffusion processes are critical to the evolution of social movements. Scholars are increasingly recognizing the theoretical importance of diffusion processes, and using diffusion language in discussing social movements. Until recently, however, these discussions have stayed at a fairly superficial level. The fact of the diffusion of action has been repeatedly demonstrated in quantitative data showing the dispersion of events across time or space, and in qualitative research documenting the direct connections between events. A wealth of new data has been and is being collected giving the time series of various kinds of violent and nonviolent events in a number of different nations (Hocke 1998; Jenkins and Eckert 1986; Kriesi et al. 1995; McAdam 1982; Olzak 1990; Olzak 1992; Olzak and Olivier 1994; Olzak, Shanahan and McEneaney 1996; Olzak, Shanahan and West 1994; Rucht, Koopmans and Neidhardt 1998; Rucht 1992). Careful analyses of these data are yielding great payoffs in our understanding of the dynamics of collective events and the interplay between different modes of action by different actors. The combination of these data and recent advances in the technology of modeling diffusion make it possible to give a much more detailed account of the mechanisms of diffusion and to integrate diffusion processes with the other processes known to be important in social movements.
Taking advantage of these data and technical advances requires reorientation of both social movement theory and traditional diffusion theory so that the two can be integrated. In this paper, we discuss the issues involved in integrating these theories, the steps that have been taken so far, and the tasks that remain. Although it is possible to imagine a full theoretical conception that is more complex than we are able to fully portray at present, we believe that the work accomplished so far indicates the tremendous advances that will be possible from completing the process of theoretical integration.
[…] The linchpin of the integration of social movement theory with diffusion concepts is to re-conceive the basic concept of a social movement. As we, among others, have written elsewhere, there has never been much clarity about just what kind of thing a social movement is. […] If we are to gain the advantages of diffusion theory, we need to give up the conception of a social movement as some kind of coherent entity, and instead conceive a social movement as a distribution of events across a population. We use the term “event” here in a general sense to encompass the actions of the various actors in a population, as well as their beliefs. In this sense, specific protest actions are events, but so is a resource flow from one group to another. It is also an event when a certain proportion of the population comes to hold a particular belief. Under this conception, a social movement peaks when there are a lot of protest actions happening involving a large proportion of the population “at risk” for participating.
[…] An emphasis on the diffusion of action as the core process in a social movement is central to studies of waves of conflict and cycles of protest. […] For scholars not used to thinking this way, the transition is difficult, but it is very important if we are to achieve a real understanding of the protest phenomenon. The transition perhaps can be compared to that in the study of evolutionary biology, where it is recognized that a species is not a distinct entity which can make choices about how to adapt to an environment, but a statistical distribution of traits across individual organisms. Species evolve when the distribution of characteristics within a breeding population changes. Social movements rise when the overall frequency of protest events rises in a population, they become violent when they ratio of violent events to non-violent events rises, and so forth.