Contested Boundaries: Explaining Where Ethno-Racial Diversity Provokes Neighbourhood Conflict


Theoretical background and objectives

This project further refines our knowledge about negative diversity effects by taking a spatial approach, which asks where diversity provokes conflicts. As such, its research question lies at the intersection of the topics of clusters 4 and 5. The project develops the “contested boundary hypothesis”, which states that neighbourhood conflicts occur along fuzzy boundaries between ethnically homogenous neighbourhoods. These transitional areas are contested and conflict-prone because they are subject to competing ethnic territorial claims. Clear-cut boundaries between groups, by contrast, are less conflictual, because they tend to be generally accepted. This implies the need for a methodological approach that does not treat neighbourhoods as independent spatial islands, but focuses instead on where they are located in the urban structure (see similarly project 4.2 above). To this end the project draws on publicly available data on 4.7 million phone calls to the New York City central service number (311), which are then linked to geo-coded data from the US Census and the American Community Survey. We explain variation in the rate of phone calls that concerned complaints about neighbours across 30.000 census blocks as a function of so-called “edge intensity”, which we code with the help of „edge detection“ algorithms, which have been developed in the field of image processing. The results clearly confirm that the rate of neighbourhood conflicts is higher where boundaries between local ethnic communities are fuzzy.