Findings in Search of a Controversy and in Need of More Data

Dawn Brancati and Adrian Lucardi draw three conclusions from their statistical analysis of global patterns of democracy protests from 1989 to 2011. First, the cross-national diffusion of such protests is the exception, not the rule. Second, domestic factors, rather than international factors, such as diffusion, largely determine whether citizens mobilize in support of democratic change. Third, they argue that these findings call into serious question the received wisdom about the frequency of democratic diffusion.

We have two problems with their analysis. First, we are puzzled by their reading of the literature—a reading that allows them to play up the importance of their findings. To our knowledge, no one who has done serious qualitative or quantitative work on cross-national diffusion has argued that such processes are common or, by implication, easy, natural, or automatic, especially when, as with the case at hand, diffusion refers to the rapid spread among states of new and subversive political behaviors that directly threaten the tenure of authoritarian leaders. There is nothing surprising, therefore, about their conclusion that “in general,” democracy protests fail to diffuse among states that are, we must remember, committed above all to resisting pressures for democratic change. How could it be otherwise?

Just as unsurprising to analysts of diffusion is this article’s emphasis on the key role played by domestic factors in preventing (and sometimes supporting) the spread of democratic change. Just as virtually every empirical account and virtually every theory of cross-national diffusion place variations in domestic receptivity to change at the center of the analysis, so a major preoccupation of many empirical studies of diffusion is explaining its limits; that is, why the adoption of a particular innovation is so uneven across time and space (and usually turning to local conditions to provide the explanation). Witness, for example, the continuing debate among specialists in the Middle East and North Africa about the uneven geography of the Arab Spring; that is, where challenges to authoritarian rule erupted and where they did not; where these challenges succeeded in removing incumbent leaders and where they did not; and, finally, if successful in deposing political leaders, where a transition to democracy subsequently took place and where authoritarian rule remained in place (or, as in Egypt, returned after a brief vacation). The answers to all three questions, it is important to recognize, invariably point to differences in domestic contexts. The one exception is the key role that Saudi Arabia played in defending the regime in Bahrain from the protests taking place in Pearl Square. Even here, however, domestic factors have received a lot of attention.

Brancati and Lucardi might not have drawn a sharp contrast between their findings and the findings of other scholars in the field, if they had a better understanding of the literature on cross-national diffusion. One case in point is our book, Defeating Authoritarian Leaders in Postcommunist Countries (Cambridge University Press, 2011), which compares successful and unsuccessful electoral challenges to authoritarian rulers or their anointed successors in postcommunist Europe and Eurasia from 1998 to 2007. In our book, we present all the data we gathered (much of it qualitative) over many years to nail down the many innovations involved in these electoral challenges; to assess whether cross-national diffusion was in fact in play (and not, e.g., a purely domestic process in which similar conditions elicited similar responses); to identify the players, the ideas and the repertoires that shaped these challenges and their ability to move across state boundaries; and finally to answer the questions of why some states experienced challenges and others did not and why, in turn, some challenges succeeded in removing authoritarian leaders from office and others did not.

Like all scholars of diffusion, moreover, we were initially skeptical about whether it had taken place, precisely because of the formidable domestic obstacles that stand in the way of subversive politics taking their show on the road. On the other hand, we also recognized that, if we were ever to see diffusion at work and understand how the process proceeds, this region and these elections would be an excellent place to start. This is because of the legacies of communism and 1989, the similarities among the more authoritarian regimes that succeeded communism in this region, the interest of the United States and members of the European Union in promoting democratic change in the authoritarian frontiers of Europe, and, finally, the fact that there was variation in both whether and, if so, how oppositions challenged dictators at the polls.

As the discussion thus far implies, we also have some methodological concerns with the Brancati and Lucardi analysis. First, we question whether their statistical analysis of the spread of “democracy protests” can adequately, let alone fully and systematically, assess the role of cross-national diffusion. Innovative demands for democratic change in the context of authoritarian regimes have taken in fact many forms. For example, if we just look at the fall of communism in 1989–1991 and the color revolutions in same region from 1998 to 2007, we see a wide range of innovative behaviors—in particular, strikes, demonstrations, electoral mobilizations (followed by protests in some cases but not in others), roundtables between the ruling party and the opposition, and legal innovations, including legal challenges mounted by citizens, civil society groups, and even international institutions, along with changes in laws enacted by reform-minded parliaments.

These diverse behaviors, moreover, are best understood not as “the” innovation or, for that matter, freestanding ones but, rather, as the culmination of a series of innovative behaviors. To take one example: in the case of the color revolutions, electoral mobilizations came about as a result of a package of innovative actions including the decision to mount ambitious voter registration and turnout drives, forge an alliance between civil society groups and opposition parties, reduce the number of opposition parties and candidates running for office, conduct an energetic opposition campaign in not just the capital and major cities but also in smaller towns and the countryside, develop an alternative media, and carry out parallel tabulation of the vote to guard against stolen elections. These ideas about how to win elections were not just unprecedented; they also benefited from successful dress rehearsals in local elections and abroad.

Many of those who challenge dictatorship, moreover, are fully aware of what is happening in their neighborhood. They learn, for example, from their participation in transnational networks—a process that in Europe’s eastern half started long before the beginning of the data set assembled by Brancati and Lucardi. In addition, before the Internet, we must remember, there was radio. Radio Free Europe and the speeches of Gorbachev, for example, played a key role in the fall of communism. Opposition groups are also fully aware of local constraints and opportunities. As a result, they draw lessons from past interactions with the regime, pick and choose among available repertoires at home and abroad, avoid making themselves vulnerable by putting all their eggs in one basket (e.g., restricting themselves to “democracy protests”), and pay close attention to the issue of when to deploy which actions. A key part of this process is amending innovative repertoires that succeeded in challenging authoritarian states similar to and close to theirs in accordance with their reading of local conditions. Sometimes this process takes a long time, and sometimes it does not. Witness, for example, how quickly Egyptian protests followed the Tunisian precedent in late 2010 to early 2011 compared with the similarities between the recent and successful protests in Sudan and the protests that took place in Egypt seven years earlier.

This brief overview of how citizens challenge authoritarian regimes is a cautionary tale insofar as the methodology used in “Why Democracy Protests Do Not Diffuse” is concerned. One general implication is that a few allusions to qualitative data do not go very far in furthering our understanding of innovative political actions or whether and, if so, when and how these actions diffuse. If we restrict our attention to the quantitative portion of this article (which is its primary focus), we can argue that “democracy protests” are a small part of the story of innovative challenges to authoritarian rulers; they may not be in fact a very representative part of that story, and they are likely the result of a series of innovative actions that are hard to quantify, hard to trace, and, for these reasons, missing from the Brancati and Lucardi analysis.

In this way, our response comes full circle. Despite their claims to the contrary, Brancati and Lucardi do not make a major contribution to the literature on the diffusion of democracy. They misread this literature while supporting its major findings. At the same time, the support they provide is mixed because of limitations in the data they have collected and in their understanding of both innovation and diffusion.