Of course, an urban environment is not just defined by its size. As Wirth emphasized, density is very important here. A population of 3,000 people spread over a large area is not urban, as Afghanistan shows. Urbanization requires density. Again, as with the figure of 3,000, there is some arbitrariness about the density required before a settlement can be defined as urban. On the basis of a variety of national demographics, the UN and the World Bank employ a figure of 400 people per square kilometre. Consequently, a settlement can be defined as urban if it consists of 3,000 people, at a population density of 400 people per square kilometre, and, therefore, concentrated in an area of no more than 7.5 square kilometres. Military operations become urban when forces begin to fight in densely populated areas that consist of at least 7.5 square kilometres of continuous human habitation and associated structures. It might, perhaps, be useful to round up the figures for mnemonic purposes, suggesting that urban warfare begins in settlements of 3,000, at a density of 500 inhabitants per square kilometre and, therefore, covering 6 square kilometres. Of course, an urban area can expand far beyond these figures, as recent battles of shown. They have been regularly fought in cities with populations of hundreds of thousands or millions, extending over hundreds of square kilometres. Yet, the figures of 3,000 people at a density of no fewer than 400 per square kilometre provide a useful threshold for urban warfare.
As the Western campaign in Afghanistan has shown, troops can certainly operate in settlements below this size and density. These actions may certainly feel urban to the platoons and companies conducting them. Below the level of 3,000, forces can be involved in ‘fighting in built-up areas’ and the immediate tactical problems are often exclusively urban for them. Squads and platoons can be engaged in urban combat. However, although individual combat can certainly be urban, the operation – the battle – is simply not urban. The congested, multidimensional challenges of urban operations do not pertain below a dense population of 3,000.
It is possible to set a numerical threshold for urban operations but it is also useful to identify a spectrum of conflict along which urban warfare is located. In On War, Clausewitz famously proposed that ‘war is merely the continuation of policy by other means’.3 War is an act of organized violence aimed at achieving a political goal:
War is nothing but a duel on a larger scale. Countless duels go to make up war, but a picture of it as a whole can be formed by imagining a pair of wrestlers. Each tries through physical force to compel the other to do his will; his immediate aim is to throw his opponent in order to make him incapable of further resistance. War is thus an act of force to compel our enemy to do our will.4
Defined in this way, it becomes easy to define urban warfare when the fighting is intense. When the armies of two states attack each other in a densely populated settlement of more than 3,000 inhabitants, it is self-evidently urban warfare. In the twenty-first century, states have rarely fought each other within cities; the norm, rather, has been for state actors to fight nonstate insurgents. However, even here, the combat has become intense enough to describe the subsequent engagements as urban warfare without difficulty; Mogadishu, Grozny, Mosul, Aleppo and Marawi were all manifestly high-intensity battles.
As with the density and size of settlements, the problem of definition comes at the lower end of the spectrum, when violence is infrequent and small-scale. Urban violence takes many forms. Protestors, terrorists and criminals are all potentially violent; they can all compromise civic order. Yet, it would be wrong to describe lawlessness and criminality as warfare, disturbing though they might be to authorities or citizens. When exactly can protest, crime and terrorism, as forms of urban violence, be defined as urban warfare?
There are some useful statistics here that are typically employed by academics and analysts to categorize conflict and war. While a conflict is defined as active if either protagonist suffers 25 battle deaths in a year, war requires at least 1,000 or more combat-related deaths in a year. It is an arbitrary but useful number. Consequently, although the line is blurred, urban warfare begins when 1,000 combatants or civilians are killed in politically motivated attacks by recognizable, antagonistic groupings inside in the city. Gang warfare, terrorism and public protests are certainly not as obvious as a form of urban warfare as interstate conflict, but there is a threshold at which criminality or political violence can begin to be defined as war. In Brazil today, one in five citizens lives in a favela (slum) on the peripheries of major cities; in these areas, every year, there are 200 murders per 100,000 people, which is comparable to the death rate in Iraq in 2003 or the Balkans in the 1990s. The Brazilian state has contributed significantly to these figures: 676 people were killed by the security forces alone in 2019. The conflicts in these areas are not wars in a traditional sense. Yet, it would also seem incorrect to dismiss them as mere criminality.
Yet, while numbers are important, they are not sufficient to define urban conflict as war. Political motivation is, of course, very important here. In a large city, it might be possible that 1,000 citizens could be killed in quite random, unconnected murders. Public order would be severely compromised in this situation, but it would not be war. Crime descends into warfare when the objectives of criminals become explicitly political – when criminals are not simply robbing civilians or disputing drugs deals, but are actively contesting terrain and the application of the law against the state, compelling their enemies to do their will. Similarly, at a certain point, terrorist attacks escalate into a serious challenge to the state when they start to kill enough people.
While urban warfare is most obvious in a battle like Mosul, there is no reason to limit the analysis of contemporary warfare in cities to high-intensity conflict. On the contrary, to have a full appreciation of the diversity and complexity of urban warfare in the twenty-first century, it is pertinent to include lower levels of conflict in the analysis. Gang warfare, terrorism and protests have often taken place alongside genuine urban warfare; they are part of the phenomenon. Although Brazilian favelas might typically be relatively peaceful for most of their inhabitants, not only do they have very high murder rates, but they are periodically the scene of intense gun battles between rival gangs, and between gangs and the police. ‘In 2018, this violent territoriality was reflected in 83 instances of gunfights lasting two hours or more.’5 At such times, favelas are rightly described as warzones. The actions of police and gangs in these clashes provide useful additional evidence about the character of urban warfare today, and should not be ignored simply because no official war has been declared or recognized between them.
Demography and Asymmetry
It is now universally acknowledged that the principle reason for the rise of urban warfare is demographic. There are simply so many people now living in cities that conflict has necessarily converged on urban areas. In 1960, the world population was 3.5 billion, of whom 0.5 billion lived in cities. In 2020, the world population was 7 billion, of whom 3.3 billion lived in cities.6 As early as 1996, commentators began to notice this urban turn. Russell Glenn, one of the first urban experts, rightly noted the military implications of this demographic revolution: ‘Demographics ensure that cities will become future battlegrounds.’7 At the same time, a prominent US Army officer, Ralph Peters, made the same point: ‘We declare that only fools fight in cities and shut our eyes against the future. But in the next century, in an uncontrollably urbanizing world, we will not be able to avoid