The City as a System: Future Conflict and Urban Resilience

By David J. Kilcullen

The City as a System

This era’s unprecedented urbanization is concentrated in the least developed areas of Asia, Latin America and Africa.  The data shows that coastal cities are about to be swamped by a human tide that will force them to absorb—in less than 40 years—almost the entire increase in population absorbed by the whole planet, in all of recorded human history up to 1960. And virtually all this urbanization will happen in the world’s least developed areas, by definition the poorest equipped to handle it—a recipe for conflict, crises in health, education and governance, and food, energy and water scarcity.

Rapid urbanization creates economic, social and governance challenges while simultaneously straining city infrastructure, making the most vulnerable cities less able to meet these challenges. The implications for future conflict are profound, with more people fighting over scarcer resources in crowded, under-serviced and under-governed urban areas.

Consider, for example, the interaction between climate change and coastal urbanization. Rural environmental degradation prompts migration to coastal cities, putting more people in low-lying regions, where the slightest sea level rise can cause major disruption. Indeed, the Asian Development Bank estimated in 2011 that drought, desertification and soil salinity, exacerbated by climate change, will prompt millions of rural people across the Asia-Pacific region to migrate to cities over coming decades:

Geography, compounded by high levels of poverty and population density has rendered Asia and the Pacific especially vulnerable to the impacts of climate change.  The region is home to more than 4 billion people and some of the fastest growing cities in the world. By 2020, 13 of the world’s 25 megacities, most of them situated in coastal areas, will be in Asia and the Pacific. Climate change will likely exacerbate existing pressures on key resources associated with growth, urbanization and industrialization.[i]

The food security effects are equally severe, as pollution from coastal urbanization imperils fish stocks, and peri-urban areas surround city cores whose infrastructure is scaled for populations far smaller than they now support. This newly settled peri-urban land was once used for farms, market gardens and orchards, but as cities expand into this space, the distance between the city core and its food sources increases significantly. Food must now be produced further away and transported over ever-greater distances, increasing transportation and refrigeration costs, raising fuel usage and carbon emissions, exacerbating traffic problems, and creating “food deserts” in urban areas. Likewise, many cities are running out of water, a problem that will only increase as populations swell, and as urbanization covers rainfall catchment areas, pushing cities further from their water sources.

The growing size and complexity of cities also strains the infrastructure of governance—police, district administrators, courts, hospitals, schools, and maintenance services. In particular, government presence can be extremely limited in peri-urban areas, allowing the emergence safe havens for criminal networks or non-state armed groups, or creating a vacuum filled by local youth, who do not lack for grievances arising from their new urban circumstances or from their home villages. Even in developed cities like Paris and London, rioting, youth unrest and crime in peri-urban districts reached significant levels on several occasions over the past decade—and in underdeveloped regions the problem is even worse.[ii]

It is useful to think of urban migration from the rural hinterland—driven by rural environmental degradation, energy poverty or conflict—as the supply side of a population-flow system whose demand side includes the problems of urban overstretch, crime, scarcity and conflict just mentioned. The city is a system which, in turn, nests within a larger national and global system, with coastal cities functioning as an exchange mechanism that connects rural hinterlands with urban populations, and with international networks. We can represent this graphically, along with the key trends we have been discussing, as follows: NIC Blog – Kilcullen – City as a system

In this model, the coastal city is the center of a larger system, with rural factors in the city’s hinterland—including environmental degradation, poor rural infrastructure, and rural conflict—prompting rapid urbanization. This creates ad hoc peri-urban settlements where slums and shantytowns displace land formerly used to provide food and other services to the city, and cover the rainfall catchment area for the city’s water supply. The city’s growth puts its infrastructure under stress, so that both the old urban core and the new peri-urban areas experience weak governance, crime, urban poverty, unemployment and conflict. Shortages of food, fuel, electricity and water exacerbate these problems. In turn, the city’s connectedness allows its population to tap into licit and illicit activities offshore, and to connect with global networks, including diaspora populations, an interaction that affects both local and international conflict dynamics.

The data suggest that this is the environment in which future conflict will occur. To recap, this is not a futuristic prediction, but rather a projection of trends that are evident now, and an assessment of their effects on cities as they exist today. There will certainly be outliers—wars in desert and mountain terrain, and state-on-state “conventional” war—but this analysis does suggest the system parameters within which future conflict is likely to occur, and the majority such conflict seems likely to be urban, networked and littoral.


This analysis of the future conflict environment suggests that the recent U.S. shift away from protracted stabilization and counterinsurgency operations, toward conventional conflict, conflict prevention and military-to-military assistance, is somewhat unrealistic. American policy-makers, not for the first time, have expressed a strong preference for avoiding messy conflicts like those of the past decade. Nonetheless, as we have seen, policy-makers’ preferences seem to have little effect on the frequency of overseas interventions, a historical pattern which suggests that the United States is likely to undertake one major stabilization or counterinsurgency operation (on the scale of Iraq, Afghanistan or Vietnam) every generation, and smaller operations every (on the scale of Bosnia or Kosovo) every five to ten years, for the foreseeable future.

While this pattern is likely to endure, the environment in which such interventions occur is shifting. The three megatrends of urbanization, littoralization and connectedness suggest that conflict is increasingly likely to occur in coastal cities, in underdeveloped regions of the Middle East, Africa, Latin America andAsia, and in highly networked, connected settings. Adversaries are likely to be non-state armed groups (whether criminal or military) or to adopt asymmetric methods, and even the most conventional hypothetical war scenarios turn out, when closely examined, to involve very significant “irregular” aspects.

Whether irregular or state-based, adversaries in the future conflict environment will exhibit hybrid characteristics (combining state and non-state, domestic and international, military and criminal and licit and illicit elements) and these adversaries will be able to nest within the complexity of urban environments, as well as within legitimate national and international systems.

The implications for the military are reasonably obvious, if difficult to act upon in the current fiscal environment. Capabilities such as Marine amphibious units and naval supply ships, as well as facilities for expeditionary logistics in urbanized coastal environments are fairly obvious requirements, as are rotary-wing or tilt-rotor aircraft, precise and discriminating weapons systems, and the ability to rapidly aggregate or disaggregate forces to operate in a distributed, small-unit mode while still being able to concentrate quickly to mass their effect against a major target. Combat and construction engineers, civil affairs units, intelligence systems capable of making sense of the clutter of urban areas, and constabulary (gendarmerie)-type and coast guard capabilities are also likely to be important. The ability operate for a long period in a city without drawing heavily on that city’s water, fuel, electricity or food supply will be important also.

The implications for civilian agencies of government are equally obvious—the ability to expand social services, city administration, and rule of law into peri-urban areas are clearly important, as are investments in infrastructure to guarantee supplies of fuel, electricity, food and water. Less obvious, but equally important, are investments in governance and infrastructure in rural areas, as well as efforts to mitigate the effects of rural environmental degradation that cause unchecked and rapid urban migration. Given the prevalence and increasing capability level of threat networks, policing capabilities will need to embody a creative combination of community policing, constabulary work, criminal investigation and special branch (or police intelligence) work, potentially in close collaboration with the military. And local city managers, district-level officials and ministry representatives may need capabilities that allow them to operate in higher-threat, opposed governance environments.

The implications for businesses, civil society and the public go well beyond either of these rather narrowly scoped considerations. In the first place, the environmental shifts outlined above represent far more than a future theory of conflict—indeed, they are a “theory of everything” in the sense that the three key megatrends identified here (urbanization, littoralization and connectedness) will affect every aspect of life on the planet in the next few decades, and will not solely affect conflict. Caerus Associates’ systems design approach for developing urban resiliency and mapping conflict[iii] and the IBM Smarter Cities project[iv] are two examples of businesses and civil society organizations taking a holistic approach to the city as a system, and thereby seeking to anticipate and address the full range of future issues that cities will confront.

More broadly, the city-as-a-system approach described earlier can be applied as a methodology to identify how complex problems that may appear unrelated—rural soil salinity, urban crime, piracy and diaspora-sponsored terrorism, for example—interact with each other in the context of a given city or threat network. Taking this approach may allow planners to identify emergent patterns within the complex adaptive system of a relevant city, make sense of the system logic, and thus begin to design tailored interventions. These would begin in a consciously experimental way, seeking to identify and reveal the complex interactions between different parts of systems, and among systems nested in larger systems, but would rapidly increase in effectiveness, as each experimental intervention would generate new data that would enhance the effectiveness of the next.

The future conflict environment is likely to be characterized by rapid population growth, increasing urbanization, accelerating littoralization, and greater connectedness. The future of conflict—like the future of most things on the planet—is likely to be urban, littoral and networked. Thinking of the city as a system seems to make the most sense as a way to understand this environment, and the irregular, hybrid and nested threats that we are likely to encounter within it. The implications for the military and for civilian government are fairly obvious, but in broader terms a city-as-system approach may also allow urban planners, city managers, businesses and communities themselves to understand their environment and develop tailored interventions to deal with it.

We are still likely to experience wars between nation-states, and conflict in remote areas such as mountains, jungles and deserts will still undoubtedly occur. But the trends are clear: more people than ever before in history will be competing for scarcer and scarcer resources, in poorly-governed areas that lack adequate infrastructure, and these areas will be more and more closely connected to the global system, so that conflict will have immediate wider implications. The future is hybrid and irregular conflict combining elements of crime, urban unrest, insurgency, terrorism and state-sponsored asymmetric warfare—more Mumbai, Mogadishu and Tivoli Gardens—and we had better start getting ready for it.

David J. Kilcullen is the Chief Executive Officer of Caerus Associates, and the author of The Accidental Guerrilla and Counterinsurgency. This post is excerpted from a forthcoming article in the Fletcher Forum on World Affairs.

[i] See Asian Development Bank, Climate-Induced Migration in Asia and the Pacific, September 2011, online at

[ii] Widespread rioting and civil unrest in outlying and peri-urban areas struckParis (and several other French cities) in 2005 and again in 2007 and 2010, while large-scale rioting and looting occurred in parts ofLondon in 2011.