A Relational Theory and Nested Methods

Defining Nation Building

As with most other concepts in the social sciences, there is no consensus about the definition of "nation building." Most scholars agree, however, that nation building entails national identification: citizens begin to see themselves as members of a national community and feel loyal to conationals, above and beyond their attachment to an ethnic group, a tribe, a village community, or a religion. But here the consensus ends.

One strand of thinking emerged from Stein Rokkan's influential work (see Flora et al. 1999). In this view, democracy and the welfare state are important tools, but also important consequences, of nation building. This broad understanding can still be found today, for example, among American foreign policymakers who focus on the democratization aspect (Dobbins 2003–2004; Fukuyama 2004) or among political economists who highlight the education, welfare, and infrastructure policies that support nation building (Miguel 2004).

Another strand of thinking sees nation building as a matter of power relations between citizens and the state. This is the perspective I assume in this book. It emerged first in the writings of Reinhard Bendix, for whom "the central fact of nation-building is the orderly exercise of a nationwide, public authority. ... Some subordination of private to public interests and private to public decision," he continues, "is therefore sine qua non of a political community. Implicitly more often than explicitly, the members of a political community consent to that subordination in an exchange for certain public rights" (1964: 18–19) — hence the title of his often cited book, Nation-Building andCitizenship. René Lemarchand pursued this exchange theoretic argument more explicitly. He suggested

new perspectives from which to look at processes of nation-building in Africa: Viewed from the micropolitical perspective of traditional patterns of interaction among groups and individuals, nation-building becomes not so much an architectonic, voluntaristic model divorced from the environmental materials available; it becomes, rather, a matter of how best to extend to the national level the discrete vertical solidarities in existence at the local or regional levels. (1972: 68)

Similar to this book, in other words, Lemarchand understood nation building as a process of political inclusion through establishing encompassing exchange relationships between the state and its citizenry. National identification, in turn, will follow from such relationships as citizens will no longer define themselves primarily as members of a guild, a city, a village, a tribe, or an ethnic group and more as members of the imagined community of the nation, to cite Anderson's (1991) proverbial formula.

On both sides of the coin of nation building — political integration and identification — countries differ considerably from each other. In some places such as France, individuals have ceased paying much attention to their regional, local, guild, or ethnic identities and think of themselves primarily as nationals. Belgians, by contrast, think of themselves foremost as Walloons or Flemish, rather than Belgian. The same goes for the political integration aspect of nation building, as Figure 1.2 will illustrate. In some countries, large ethnic groups remain outside the alliance and support networks stretching from the seats of government down to the villages of the hinterland. In Syria, the Assad clan and their fellow Alawi have held a firm grip on all high-level government and military positions over the past several decades. Alawi are also dramatically overrepresented in lower-level government positions compared to Sunni or Kurds (Mazur 2015). In other societies, more inclusive configurations of power have emerged and most citizens are integrated into the web of alliances and support centered on the national government. Examples include Switzerland, Malaysia, and Burkina Faso — all ethnically heterogeneous countries.

To illustrate, Figure 1.1 shows an inclusionary and an exclusionary configuration of power. Nodes represent political actors (organizations or individuals), lines describe exchange relationships, and actors higher up in the graph wield more political power, with those at the top representing national government. Two clarifications follow from that. First, the same ethnic demographics characterize both countries represented in Figure 1.1: they are composed of an ethnic majority (the gray dots) and a minority (the white dots). This illustrates that ethnic diversity and ethnic inclusion are different concepts. Figure 1.2 shows, with data that will be more fully explained and explored in subsequent chapters, that they also need to be distinguished from each other empirically. The y-axis reports the population share of the ethnic communities not represented at central-level government. It is a rough measurement of the extent to which nation building has succeeded (when the share is low) or failed (if it is high). The x-axis represents the ethnic diversity of a country — measured as the likelihood that two randomly chosen individuals share the same ethnic background: 1 in perfectly homogeneous countries and 0 if every individual belongs to a different ethnic group. As is clear from the figure, nation building and ethnic diversity might be related to each other — as will be argued in detail — but they are conceptually distinct.

Second, political inclusion and nation building also need to be distinguished from democratization, in contrast, for example, to Dobbins (2003–2004) and much of the policymaking literature. Access to state power can also be organized, for example, through ethnic patronage networks within a one-party regime (as in Burkina Faso; Rothchild 1986). Fully democratic countries can be exclusionary (e.g., the United States until the civil rights reforms) and very undemocratic countries more inclusive (such as the Ivory Coast during the presidency of Houphouët-Boigny; see Rothchild 1986). This is shown in Figure 1.3, where the x-axis now depicts the degree to which the state is governed as a democracy. The scale ranges from -10 for total autocracies to +10 for perfect democracies....