Predictable Policing: Predictive Crime Mapping and Geographies of Policing and Race

Abstract

This article draws on critical geographic engagements with policing and race and geographic information systems (GIS) to investigate the implications that predictive crime mapping has for racialized modes of urban policing. Focusing on the Chicago Police Department (CPD), it analyzes collaborations between geographic information scientists, crime experts, and police who have recently begun integrating temporal data into GIS-based maps to predict when and where future crimes will occur. The article builds the case that predictive crime mapping further entrenches and legitimizes racialized policing as it (1) rearticulates police data sets as scientifically valid and (2) correlates those data with other geocoded information to create new rationalizations for controlling racialized districts through differential policing practices. The article uses a mixed-methods approach that includes analysis of open-ended interviews with computer scientists involved with the CPD's Predictive Analytics Group and city technical documents to explain the recursive relation between GIS-based knowledge production and racialized policing. The article casts into relief the central role that the production of geographic information plays in current modes of racialized policing and how this contributes to the ongoing racial differentiation of urban geographies.

本文运用警备和种族的地理涉入, 以及批判地理信息系统 (GIS), 探讨预测犯罪製图对于城市警备的种族化模式之意涵。本文聚焦芝加哥警察局 (CPD), 分析地理信息科学家, 犯罪专家, 以及警察之间的合作, 他们最近开始将时间数据整合进根据 GIS 的地图, 以预测未来的犯罪将在何时何地发生。本文将论证, 预测性的犯罪製图进一步确立并正当化种族化的警备, 因其 (1) 再度表达警察数据集是具有科学根据的, 以及 (2) 将上述数据连结其他进行地理编码的信息, 为透过差异化的警备行为来控制种族化的地区创造了崭新的合理化。本文运用混合方法取径, 包含对参与芝加哥警察局预测分析团体的计算机科学家的开放式访谈之分析, 以及城市的技术文件, 以解释根据地理信息系统的知识生产与种族化的警备之间的递归关係。本文清晰化地理信息生产在当前的种族化警备中的核心角色, 及其如何导致城市地理的持续种族差异化。

El artículo se basa en enfrentamientos geográficos críticos con la vigilancia policial y la raza, y en los sistemas de información geográfica (SIG) críticos para investigar las implicaciones que tiene el mapeo predictivo del crimen para los modos racializados de la vigilancia policial urbana. Centrando la atención en el Departamento de Policía de Chicago (CPD), el artículo analiza las colaboraciones entre científicos de información geográfica, expertos en criminalística y la policía que hace poco empezaron a integrar datos temporales en mapas basados en SIG para predecir cuándo y dónde ocurrirían los crímenes futuros. El artículo construye el caso de que el mapeo predictivo del crimen atrinchera aun más y legitima la vigilancia policial racializada por cuanto ese mapeo (1) rearticula conjuntos de datos policiales como científicamente válidos, y (2) correlaciona esos datos con otra información geocodificada para generar nuevas racionalizaciones que ayuden a controlar distritos racializados por medio de prácticas de vigilancia policial diferenciales. El artículo usa un enfoque de métodos mixtos que incluye análisis de entrevistas de final abierto con científicos de computación involucrados con el Grupo de Análisis Predictivo del CPD y documentos técnicos de la ciudad para explicar la relación recursiva entre la producción de conocimiento basada en SIG y la vigilancia policial racializada. El artículo destaca el papel central que juega la producción de información geográfica en los modos corrientes de vigilancia policial racializada y cómo eso contribuye a la actual diferenciación racial de las geografías urbanas.

Key Words:

• Chicago
• GIS
• police
• predictive policing
• race

关键词：:

• 芝加哥, 地理信息系统, 警察, 预测性警备, 种族。

Palabras clave:

• Chicago
• SIG
• policía
• vigilancia policial predictiva
• raza

Acknowledgments

The author is grateful to Phillip Kalantzis-Cope for bringing predictive policing technologies to his attention in 2014. He would also like to thank Mei-Po Kwan, Sara McLafferty, and James Pascaleff for their help with the conceptual and technical issues explored in this article. Finally, he thanks Angel L. Veles and Shawn Fields for their research assistance.

Notes

1. In 2006, the Memphis Police Department teamed with University of Memphis criminologists to create its Blue CRUSH system, which uses IBM predictive analytics to narrow patrol areas to a few square blocks of prospective crime (Vlahos 2012). That year, the Los Angeles Police Department began working with UCLA researchers to develop a model that simulates future crime diffusion and dissolution patterns (Pittman 2010). The Santa Cruz Police Department uses a mapping system that shows the probability of property crimes in 500 × 500 foot areas by the hour. The software uses an algorithm to predict the spread of property crimes based on a method used by seismologists to forecast earthquake aftershocks, and was credited for bringing property and violent crime to their lowest levels since 1973 (Thompson 2011).

2. Herbert's (1996, 2005; Herbert and Brown 2006) work, which is examined later, provides a notable exception.

3. This argument is derived from Turk (1969).

4. These materials were located using search phrases including “predictive analytics group,” “CrimeScan,” “geographic information systems,” and “CLEARmap,” which stands for Citizens Law Enforcement and Reporting map. It is the CPD's current GIS crime mapping application.

5. Manning's (2011) book, The Technology of Policing, explores the impact (or lack thereof) that GIS crime analysis and mapping has on the internal, day-to-day operations of police. This article, by contrast, explores how these technologies are used by police officials to legitimize such operations.

6. Data are from 1990 (U.S. Census 2000).

7. Black Belt neighborhoods are at least 80 percent black community areas, including West Garfield Park, East Garfield Park, North Lawndale, Near West Side, Near South Side, Douglas, Oakland, Grand Boulevard, Washington Park, and Englewood.

8. Statistics were not kept for latinx inmates in 1995.

Additional information

Notes on contributors

Brian Jordan Jefferson

BRIAN JORDAN JEFFERSON is an Assistant Professor in the Department of Geography and GIScience at the University of Illinois Urbana–Champaign, Champaign, IL 61820. E-mail: [email protected]. He researches interfaces between geographic knowledge production, political economy, race, and carceral governance.