How States Create Ethnic Cleavages
Prominent theories of ethnic politics expect the initiatives of political entrepreneurs to determine whether an ethnic identity—religious, tribal, regional, linguistic, or racial—becomes an element of party competition. According to these theories, entrepreneurs strategically construct ethnic-based coalitions of voters, organize around ethnic identities in response to electoral rules or institutional change, or use pre-existing organizations to reach and organize voters into ethnic blocs. However, the direction of causality, I argue in my book project, can also be reversed. The actions of entrepreneurs, institutional design, and organizational capacity could be shaped in response to an emerging ethnic-based coalition of voters.
In my book project, I advance a new theory of ethnic cleavage formation. I argue that voters and not political entrepreneurs determine whether an ethnic cleavage emerges. Ethnic identities begin to predict voting behavior when government policies link ethnic identities to costs and benefits. Because voters cannot easily change their ethnic identityto avoid the negative implications of the policies, they begin to coordinate their voting behavior around the aggrieved identity, therefore prioritizing that identity over other identities and interests. Political entrepreneurs, who face uncertaintyin electoral competition about the implications of their actions and the actions of other parties have strong incentives to organize around the emerging ethnic coalition. By organizing around a viable voter coalition, entrepreneurs can expect to win legislative seats and therefore increase the chances that the party survives until the next elections. Moreover, parties that do not have access to state resources and cannot credibly commit to rewarding voters after the elections can benefit from mobilizing voters based on non-material promises like protecting their interests and rights.
The methodological challenge in testing the argument that voters organize around ethnicity in response to government policies is that government policies and group-based policy preferences often affect each other. This mutual influence is clearly evidenced in many theories of ethnic politics: political entrepreneurs and parties mobilize voters around ethnic identities based on the expectation that the party will deliver goods to the members of the ethnic group. Given the mutual influence between policy preferences and government policies, evaluating the impact of policies on preferences requires finding cases of government policies that were not designed to reinforce existing cleavages or strategically intended to motivate voters to coordinate around a shared ethnic identity. This identification strategy excludes cases in which ethnicity was already a dimension of party competition or elites in power or entrepreneurs in society instrumentally drove policy changes to benefit their goals. Cases that meet the selection criteria are notably rare given the legacies of colonial rule on the political relevance of ethnic divisions in contemporary democracies.
The book project draws on evidence from several cases of plausibly exogenous government policies. I test the theory in a mixed-methods research design that treats ethnic cleavage formation as a binary outcome in the qualitative analysis and as a continuous variable in the statistical analysis. In the analysis of each case study, I rely on detailed historical research in multiple languages to establish a benchmark against which to compare the voting behavior of the aggrieved voters before and after the policies were initiated. Through process-tracing, I establish the causal chain in the process of cleavage formation and show that voters are responding to government policies and not to the initiatives of entrepreneurs. I also gain insights from cases in which entrepreneurs failed to craft new ethnic cleavages and treat them as counter-factual scenarios in which voters were not aggrieved by government policies and an ethnic electoral cleavage did not emerge. Where election returns are available, I demonstrate in statistical analyses of new datasets that the cleavages emerged most sharply in districts that bore, or expected to bear, the highest cost from the policies.
The Religious Cleavage in Prussia
Among the several case studies in the book project is the formation of a religious cleavage in Prussia in 1852 when decrees that were meant to target a small religious order affected all Catholics. In a second Prussian case, external circumstances brought liberal delegates after 1866 to a position of power and allowed them to persecute Catholics. Religion was a salient ethnic identity in Prussia, as in many other parts of Western Europe. Catholics were concentrated in the eastern and western provinces of Prussia and were an ethnic minority that composed roughly a third of the population in 1849. Despite the distinctiveness of the Catholic identity, Catholic communities differed in their cultural practices and ideological outlooks. The Catholics in the East were ethnically Polish and agrarian while the Catholics in the western provinces were comparatively more liberal, especially in the cities, and experienced greater economic development.
I draw on new datasets constructed from biographical handbooks and electoral returns from Prussian state elections to document the changes to the political relevance of the Catholic identity in Prussia from 1848 to 1873. I show that these patterns of Catholic alignment and dealignment cannot be explained as the outcome of changes to mobilizational capacity or elite initiatives. Ahead of the elections to the Frankfurt Assembly in 1848 and the Prussian lower house in 1849, the Catholic Church attempted to organize Catholics to vote for its candidates but many Catholic voters joined Protestants and voted for liberal candidates. In the following years, Catholics did not vote together whenever Catholic entrepreneurs attempted to organize Catholics (marked in gray columns in the figure below) or when Catholic associations were particularly encompassing and broad (1848-9 and during the 1860s). Although Catholics in Prussia underwent a cultural revival starting in the early 1850s that improved the moral authority of the parish priest, Catholics did not vote together unless aggrieved by government policies. Neither did the exceptionally passionate rhetoric of priests and laymen in 1848, 1861, and 1870 or the formation of a Catholic cleavage in the neighboring state Baden in 1864 unite the Catholic vote. I find that what explains the patterns of Catholic alignment and dealignment was whether government policies aggrieved Catholics because of their ethnic identity. When the state was neutral to Catholics, they voted based on other identities and interests.
The Linguistic Cleavage in Belgium
In another case study, I examine how the industrialization of present-day Flanders during the 19th century challenged the status-quo in which French was the dominant language of the business and political elites after the Belgian independence in 1830. The rapid economic development in Flanders in the second half of the 19th century led to the growth of a new social class of technicians, clerks, teachers, lawyers, and doctors, who wanted access to administrative and other high-status positions, thus creating a new demand to expand the use of Flemish to business and commerce. Before the industrial revolution in Flanders, the handful of upwardly-mobile middle-class Flemings were integrated into the francophone culture and did not make demands to create new opportunities for Flemish speakers. The majority of Flemings were disenfranchised manual laborers who lived and worked in Flemish-majority districts and did not require French for everyday interactions. Because industrialization escalated the rate of social and economic change, gradual assimilation into the francophone culture was no longer feasible, and language became a barrier to upward social mobility for a growing number of Flemish speakers.
The demand that Flemish will gain an equal status to French had the potential to take a more central place in Belgian politics when the franchise was expanded in 1894 and the size of the electorate increased tenfold. Based on the legislative debates of the Belgian lower house, I constructed a dataset in which each observation is an MP that was elected between 1882 to 1919 in present-day Flanders and Brussels. I match this dataset to information about the social and demographic profile of Belgian constituencies that I calculate based on the Belgian industrial census of 1895 and the population census of 1900.
Because the electoral reform of 1893 did not alter the boundaries of the Belgian constituencies or their district magnitude, I am able to conduct meaningful comparisons between the behavior of MPs before and after the expansion of the franchise in 1894, when the majority of the MPs began to take their oath of office in Flemish. I find that MPs were likely to take the oath of office in Flemish or switch the language of the oath from French to Flemish in constituencies that had a larger share of newly enfranchised voters from the middle and lower classes.
Implications for Understanding Party System Formation and Change
The book project contributes to a longstanding and persistent literature about the formation of ethnic cleavages. It develops and tests a new theory that voters and not entrepreneurs determine whether ethnic cleavages emerge and that voters are responding to government policies. The project also expands the growing literature about the formation of political parties in Western Europe. While scholarship on the early stages of European mass democratization is largely elite-centered, the study shows how theory and evidence from developing democracies and the increasing availability of historical data can be used to study the social bases of political parties and the role of voters in forming electoral coalitions. Second, it expands the literature about new parties. The project challenges the conventional wisdom that parties create new policies and shows that policies can also create new parties. Finally, this study also shows that government policies are crucial in determining the structure of party systems. While conventional theories expect social divisions and electoral rules to produce the main dimensions of party competition, they are insufficient in explaining why contemporary party systems change even when electoral rules are static and the social structure changes very gradually. Given the recent formation of new identity-based electoral cleavages in consolidated party systems, the lesson that government policies can become focal points for new coalitions of voters is especially timely.