Now, as ever, citizens fight and die in battle to protect their homeland. But exactly what is the homeland? Why is this land my land, why is that land not part of my homeland? My project seeks to explain and empirically test why some territory is considered homeland while other territory is not.

The project builds on core insights from the work of the political geographer Robert Sack, who recognized that there are two ways to define group membership. He distinguished categorization by type, for example language or skin color, from categorization by area, whereby inclusion in a specified area bestows group membership. A clear and unambiguous spatial definition of group membership increases the likelihood that individuals are willing to sacrifice on behalf of the group. Such a definition makes it easy to detect shirkers - people who do not abide by group norms - and thereby makes it less likely that people will shirk in the first place. Moreover, a strict definition also ties the hands of the leader. By refusing to defend homeland territory, or asking for sacrifices to obtain non-homeland territory, the leader signals he may be acting out of self-interest rather than on behalf of the group. This makes it less likely that individuals would be willing to sacrifice for the goals set by the leader, and more likely the group will seek to replace the leader. Thus, because it benefits both leaders and followers, such a spatial definition of group membership promotes strong group identity and powerful group norms.

In practice, school atlases, geography textbooks, wall maps and the constant repetition of the 'map as logo' coordinate expectations, and create common knowledge about the group's spatial boundaries: i.e., the homeland. Maps and geographic education thus create or strengthen a group identity with its concomitant norm to defend the group and its members as defined by the homeland. Group members (citizens) will defend group members and homeland territory; they are less willing to sacrifice for non-homeland territory. Support for claims to non-homeland territory unravels the coordination equilibrium, leaving everybody worse off. In short, my theory explains why people are willing to fight and die for some areas and not others on the basis of their coordination on a common map-based definition of the homeland.

The project empirically assesses this theory with two complementary empirical strategies. First, I am collecting, geo-referencing, and tracing a time-series of maps from school atlases and geography textbooks, from all South American countries. For each country, I aim to have at least one map for every five year period between 1870 and the present. This allows me to trace conceptions of the homeland and how they changed and when
leaders from different countries lay claim to the same territory. According to my theory, international disputes should arise only over areas included by both countries on their respective homeland maps. Second, surveys will test whether these geography textbooks and school atlases actually create lasting conceptions of the geography of the homeland. I examine two recent cases where a country drastically changed the boundaries of its homeland: Ecuador (1995-6) and Venezuela (1966-7). The surveys will examine whether people who graduated after the major change, i.e., 1997 in Ecuador and 1967 in Venezuela, have a different geographic and spatial conception of their homeland than people who graduated before the major change.

This research will help us better understand the origins of international territorial disputes, the deadliest form of international conflict. In addition, it will throw new light on the enduring questions of the development and construction of nationalism and national identities. Finally, the project will generate the first data set to trace how homelands and borders change over time.