Mountainous Terrain and Civil Wars: Geospatial Analysis of Conflict Dynamics in the Post-Soviet Caucasus

Abstract

Existing research on the relationship between mountainous terrain and conflict has generally been implemented using crude metrics capturing the actions and motivations of armed groups, both insurgent and government. We provide a more geographically nuanced investigation of two specific propositions relating mountainous terrain to violent conflict activity. Our study covers five wars in the Caucasus region: the second North Caucasus war in Chechnya and neighboring republics (1999–2012); Islamist and Russian government conflict in the same area (2002–2012); fighting between Armenians and Azerbaijanis in Nagorno-Karabakh (1990–2012); and battles between Georgia and separatists in South Ossetia (1991–2012) and Abkhazia (1992–2012). Our analysis of insurgent and government violence reciprocity illustrates some expected patterns of what we call the operational costs of context. By varying the dimensions for our units of analysis—the context within which violent interactions take place—however, we arrive at differing conclusions. Our research represents a meaningful and transparent engagement with the influences of the well-known and understudied modifiable areal unit problem (MAUP) in geographically sensitive analysis.

山区和冲突间的关係之研究, 一般透过运用粗糙的指标, 同时捕捉作为反抗和政府组织的武装团体的行动及意图。我们为连结山地与暴力冲突活动的两个主张, 提供地理上更为细緻的探讨。我们的研究涵盖高加索地区的五场战役: 在车臣及周围的共和国发生的第二次北高加索战役 (1999 年至 2012 年); 在同一地区中的伊斯兰与俄罗斯政府间的冲突 (2002 年至 2012 年), 亚美尼亚人与阿塞拜疆人在纳戈尔诺-卡拉巴赫发生的战争 (1991 年至 2012 年), 以及格鲁吉亚和分离主义者在南奥赛梯的战役 (1992 年至 2012 年) 。我们对于反抗和政府暴力互动的分析, 描绘出我们称之为脉络操作成本的若干预期模式。但透过多样化分析单元的各个面向——暴力互动所发生的脉络——我们却得到了不同的结论。我们的研究, 呈现对具有地理敏感度的分析中为人所熟知且未被充分研究的可调整地区单元问题 (MAUP) 进行有意义且透明的涉入。

La investigación existente sobre las relaciones entre terreno montañoso y conflicto ha sido implementada, en general, con el uso de métricas crudas para captar las acciones y motivaciones de los grupos armados, tanto de insurgentes como de los gobiernos. Lo que nosotros entregamos es una investigación de matices más geográficos sobre dos proposiciones específicas que relacionan el terreno montañoso con las actividades del conflicto violento. Nuestro estudio cubre cinco guerras en la región del Cáucaso: la segunda guerra del Norte del Cáucaso en Chechenia y las repúblicas vecinas (1999–2012); el conflicto islamista con el gobierno ruso en la misma área (2002–2012); la lucha entre armenios y azerbaiyanos en Nagorno-Karabakh (1990–2012); y las batallas entre Georgia y los separatistas en Osetia del Sur (1991–2012) y Abkhazia (1992–2012). Nuestro análisis de la reciprocidad en violencia de insurgencia y gobierno ilustra algunos de los patrones esperados de lo que nosotros denominamos costos operacionales del contexto. Sin embargo, variando las dimensiones de nuestras unidades de análisis—-el contexto dentro del cual tienen lugar las interacciones violentas—-llegamos a diferentes conclusiones. Nuestra investigación representa un compromiso significativo y transparente con las influencias del bien conocido como poco estudiado problema de la unidad areal modificable (MAUP, acrónimo en inglés) en análisis geográficamente sensible.

Key Words:

• Caucasus
• conflict
• modifiable areal unit problem
• political geography
• spatial analysis

关键词：:

• 高加索
• 冲突
• 可调整地区单元问题
• 政治地理学
• 空间分析。

Palabras clave:

• Cáucaso
• conflicto
• problema de la unidad areal modificable
• geografía política
• análisis espacial

Acknowledgments

We acknowledge the work of many undergraduate research assistants who coded and georeferenced the events data. We thank two reviewers for comments that forced us to improve this article. Nancy Thorwardson prepared Figure 1 for publication. Supporting information documents and replication materials are available online through the corresponding author's Web site.

Funding

The authors thank the National Science Foundation's Human and Social Dynamics program (grant numbers 0433927 and 0827016 to Principal Investigator John O'Loughlin) for the financial support that made possible both field work in the North and South Caucasus between 2005 and 2012 and the violent events data collection.

Supplemental Material

Supplemental analysis for this article can be accessed on the publisher's Web site at http://dx.doi.org/10.1080/24694452.2016.1243038. Table S1 presents descriptive statistics for our data sets at alternative spatial resolutions (10 km² and 50 km²). Figure S1 shows the temporal smoothing term for month ID that corresponds with the spatial smoothing term of our main model presented in Figure 7. Table S2 shows estimates for a Poisson functional form of our model of conflict event count outcomes. In Table S3 we present the results for a binary version of the outcome variable in a logistic regression estimate.

Note

Notes

1. We adopt the term rayon to refer to the county-scale units in the North Caucasus as well as the subnational units in the South Caucasus, which are variously termed in these countries (Georgia = municipality; Armenia = marz; and Azerbaijan = rayon).

Additional information

Notes on contributors

Andrew M. Linke

ANDREW M. LINKE is an Assistant Professor in the Department of Geography at the University of Utah, Salt Lake City, UT 84105. E-mail: [email protected]. His research interests include conflict and violence, political geography, and quantitative methods and geographic information systems.

Frank D. W. Witmer

FRANK D. W. WITMER is an Assistant Professor in the Department of Computer Science and Engineering at the University of Alaska Anchorage, Anchorage, AK 99508. E-mail: [email protected]. His research uses remote sensing data and spatial statistical methods to quantify human activity, including violent conflict.

Edward C. Holland

EDWARD C. HOLLAND is an Assistant Professor in the Department of Geosciences at the University of Arkansas, Fayetteville, AR 72701. E-mail: [email protected]. His research interests include political and cultural change for Russia's national minorities since the end of communism.

John O'Loughlin

JOHN O'LOUGHLIN is College Professor of Distinction, Professor of Geography, and Faculty Research Associate (Program on International Development) in the Institute of Behavioral Science at the University of Colorado at Boulder, Boulder, CO 80309. E-mail: [email protected]. His research interests include political geography, nationalism, and the former Soviet Union.