AJS. 2016 Jul;122(1):125-161. doi: 10.1086/686942.
Concerns about neighborhood erosion and conflict in ethnically di- verse settings occupy scholars, policy makers, and pundits alike; but the empirical evidence is inconclusive. This article proposes the contested boundaries hypothesis as a refined contextual explanation focused on poorly defined boundaries between ethnic and racial groups. The authors argue that neighborhood conflict is more likely to occur at fuzzy boundaries defined as interstitial or transitional areas sandwiched between two homogeneous communities. Edge detection algorithms from computer vision and image processing allow them to identify these boundaries. Data from 4.7 million time- and geo-coded 311 service requests from New York City support their argument: complaints about neighbors making noise, drinking in public, or blocking the driveway are more frequent at fuzzy boundaries rather than crisp, polarized borders. By focusing on the broader sociospatial structure, the contested boundaries hypothesis overcomes the "aspatial" treatment of neighborhoods as isolated areas in research on ethnic diversity.
人们普遍关注在种族多元化背景下邻里关系的侵蚀和冲突问题，学者、政策制定者和专家们对此也各抒己见，但实证证据尚无定论。本文提出了有争议的边界假说，作为一种精细化的情境解释，该假说重点关注族裔群体之间定义不明确的边界。作者认为，邻里冲突更有可能发生在界定模糊的边界，这些边界被定义为两个同质社区之间的中间或过渡区域。计算机视觉和图像处理中的边缘检测算法使他们能够识别这些边界。来自纽约市 470 万条时间和地理编码的 311 服务请求的数据支持了他们的观点：有关邻居大声喧哗、在公共场所饮酒或堵住车道的投诉在模糊边界更为频繁，而不是在清晰、两极分化的边界。通过关注更广泛的社会空间结构，有争议的边界假说克服了将邻里关系作为研究族裔多样性时孤立区域的“非空间”处理。
