Out of the Mountains

by David Kilcullen

  Once they landed in Mumbai, the terrorists also exploited the connected, networked nature of the urban environment. They used Skype, cellphones, and satellite phones to connect with their handlers in Pakistan, who in turn monitored Twitter, news blogs, international and local satellite news, and cable television in real time, which allowed them to control the attacks and react as the Indian response developed.

  The importance of the Karachi control node is obvious if we look at the role of Abu Dera Ismail Khan, the team leader, who died early in the operation, in a diversionary attack a long way from the main targets. If Khan had been running the operation in a classical military command-and-control manner, it would have made no sense for him to lead a secondary attack of this kind. His place would have been with the main team at the main objective: he would have given the job of leading the secondary attack to a trusted subordinate. That he was assigned to a diversionary objective—albeit one requiring considerable on-the-fly decision making—underlines the continuous and intimate control that the Karachi operations room exercised over the teams at the main objectives. Meanwhile, the assault pairs themselves seem to have operated autonomously, in a “flat” structure with no hierarchy among teams, each directly responsive to the command node in Karachi. The Mumbai attack was thus, in effect, directed by remote control, making the connectivity between the assault teams and the remote command center a critical element in the operation.

  Likewise, the attack team’s focus on foreigners seems to have been calculated by LeT controllers to maximize international attention, creating an extremely high level of news coverage—and resulting in an unusually large number of foreigners (including citizens of twenty-two countries) being killed.42 This may have been, in part, the classic terrorist tactic of maximizing publicity, but it may also have been an operational requirement: since the raiders’ command-and-control methodology relied on the Karachi operations room monitoring Twitter and Internet feeds in order to control the assault teams, the raiders needed to do something in order to create large-scale Twitter and Internet traffic, so as to generate a sufficient online signature to close the feedback loop with their command node.

  The raiders mostly didn’t target specific individuals—the killing of Commissioner Karkare, the Mumbai counterterrorist chief, seems to have been a pure accident, while the importance of each target group seems to have been determined either (in the case of the diversionary attacks) by its disruptive effect on the city or (in the case of the main attacks) by its media value. The attackers seem to have deliberately drawn out the operation over as many days as possible, hardening and consolidating their positions as soon as they entered the main target sites, and preparing for a lengthy siege. Their goal seems to have been to maximize the raid’s disruptive impact and increase the effect of terror and urban dislocation by shutting Mumbai down for as long as possible. The attacks on transportation and public health infrastructure (the taxis, railway station, and hospital) also seem calculated to maximize disruption within the urban flow of Mumbai and slow the Indian response.

  This response was affected by problems in coordination among the city authorities of Mumbai (including local police, fire brigade, ambulance, and hospital services), the government of Maharashtra State, and the Indian central government in New Delhi. In order to use national-level assets such as the MARCO and NSG teams, the state government had to approve their deployment and agree to cede control over the incident sites to central government organizations, a process that took almost six hours to negotiate, delaying the national response; in the meantime, the local police were severely outgunned, while the Mumbai antiterrorism squad was reeling from the loss of its commander early in the raid.

  The attacks didn’t involve weapons of mass destruction or particularly high-tech equipment. As in most irregular conflicts, the raiders used small arms (rifles and pistols), improvised explosive devices, and grenades; they didn’t even use rocket-propelled grenades. Small arms, however, because they involve intimate contact between attackers and victims, because their use implies the presence of an enemy on the spot, and because gun battles tend to last longer than bombings, can have a greater terror effect than a bombing or hostage situation. In an urban environment, where firefights tend to be fleeting and to occur at short ranges among small numbers of combatants, the terrorists’ ability to operate in a distributed swarm of autonomous small teams, with low signature and high mobility (due to their light weapons and combat loads), was a key tactical advantage.

  Likewise, the raiders used no unusually sophisticated or specialized communications devices: they employed commercially available phones and off-the-shelf GPS devices, and pulled much of their reconnaissance data from open-source, online tools such as Google Earth. They did, however, display an excellent standard of preparation and reconnaissance, and extremely good skills in sea movement, coastal infiltration, and small-boat handling, techniques that are obviously optimized for littoral environments. They clearly understood the urban-littoral dynamics of Mumbai—the systems logic of the way the city worked—and used this knowledge to their advantage. In this respect, assistance from state sponsors (or perhaps, nonstate sponsors who somehow managed to gain excellent access to military-grade equipment, training, intelligence, facilities, and weapons) was a key factor in the raid’s success.

  What does all this say about the future environment? First, I should make it clear that Mumbai represents only one kind of threat that will exist in the urban, networked littoral of the future; we’ll take a detailed look at others later in this chapter. That said, Mumbai exemplifies the higher end of the threat spectrum, that of state proxies using irregular (sometimes called “asymmetric”) methods to temporarily disrupt an urban target, rather than to control an urban population over a long period of time. Crime researchers John P. Sullivan and Adam Elkus describe this as an evolved twenty-first-century form of “urban siege”:

  There are several methods that terrorists and criminal insurgents use to besiege cities from within—pure terror and systems disruption, although the two are often combined together. Both methods are sustained means of besieging a city with a campaign of protracted urban violence. Pure terror is a form of social systems disruption. It is a spasm of violence intended to demonstrate to the public that the authorities cannot help them, and that they are helpless against the power of the gun. . . . While the success of the Mumbai terrorists came in large part from the tactical and operational inadequacy of Indian law enforcement response, it is easy to imagine a small group of terrorists creating multiple centers of disorder at the same time within a major American city in same manner. An equally terrifying scenario is a Beslan-type siege in school centers with multiple active shooters. Paramilitary terrorists of this kind would aim for maximum violence, target hardening, and area denial—capabilities that many SWAT units would be hard-pressed to counter.43

  To my mind, Mumbai represents the current state of the art in urban littoral terrorism. The attack has served as the model for at least two planned copycat raids on major coastal cities in Southeast Asia and Europe, and its level of technical difficulty alone was enough to raise LeT’s stature as a regional terrorist organization.44 In this context it’s worth noting that guerrillas and terrorists can gain strategic advantage just by demonstrating skill, daring, and tactical competence: the “style points” they acquire, and the shock value of showing they’re a force to be reckoned with, can outweigh tactical failures.

  But Mumbai was far from a tactical failure: on the contrary, the attack showed that a nonstate armed group can carry out an appallingly effective seaborne raid on a major coastal city, over a three-day period, in several dispersed locations—the type of operation traditionally associated with high-tier special operations forces such as the U.S. Navy SEALs or the Royal Marines’ Special Boat Service. Indeed, Mumbai was a further demonstration of a long-standing trend, sometimes called the democratization of technology, in which nonstate a
rmed groups are fielding highly lethal capabilities that were once the sole preserve of nation-states. As a combat-experienced officer with an understanding of urban riverine operations in Iraq said to me, the Mumbai terrorists’ callous disregard for human life was deeply horrifying, but “any maritime special operator in the world would be proud to pull off such a complex operation.”45

  This was far from the first major seaborne terrorist attack—it wasn’t even the first such attack in India. But the Mumbai raiders showed an extraordinary ability to exploit transnational littoral networks and both legitimate and illicit traffic patterns, inserting themselves into a coastal fishing fleet to cover their approach to the target. Their actions blurred the distinction between crime and war: both the Indian ship captain and local inhabitants initially mistook them for smugglers, and their opponents for much of the raid were police, not soldiers. They exploited Mumbai’s complex patterns of coastal urbanization by landing from the sea close to the urban core but choosing landing places in slum settlements with limited government presence. Obviously enough, this approach wouldn’t have worked without significant help from active or retired members of the Pakistani military, so Mumbai is rightly seen as a hybrid state/nonstate attack. Equally obviously, though, the attack could only have occurred in a highly networked, urban, littoral environment—precisely the environment that’s becoming the global norm.

  II. Mogadishu: Things Fall Apart

  Along with terrorism and proxy warfare, the urban, coastal, connected environment of the future will harbor more diffuse threats—what we might call “threats without enemies,” which, by definition, aren’t amenable to military or law enforcement responses.46 These may arise not from the presence of armed groups per se but from complex interactions among criminal and military actors, domestic and international networks, city populations and governments, and the urban organism and its external environment.

  Richard Norton’s idea of the “feral city,” quoted at the start of this chapter, is relevant here. A decade ago, in an influential article in the Naval War College Review, Norton defined a feral city as “a metropolis with a population of more than a million people, in a state the government of which has lost the ability to maintain the rule of law within the city’s boundaries yet remains a functioning actor in the greater international system.”47 This kind of city, Norton points out, has no essential services or social safety net. Human security becomes a matter of individual initiative—conflict entrepreneurs and community militias emerge, Mad Max style. And yet feral cities don’t just sink into utter chaos and collapse—they remain connected to international flows of people, information, and money. Nonstate groups step up to control key areas and functions, commerce continues (albeit with much corruption and violence), a black market economy flourishes, and massive levels of disease and pollution may be present, yet “even under these conditions, these cities continue to grow, and the majority of occupants do not voluntarily leave.”48 In urban metabolism terms, these are cities whose flows have overwhelmed the carrying capacity of their internal systems: the problem is not collapse (a lack of flow, as it were) but rather a superabundance of uncontrolled flows, and the toxic by-products arising from the city’s failure to absorb and metabolize them.

  Like the notion of urban metabolism, the idea of the feral city is drawn from a concept in biology. A feral animal is a domesticated one that has regressed to the wild, adapting to wilderness conditions and reacquiring (perhaps over generations) some characteristics of the original untamed species. The same thing happens with agricultural crops.49 Feral populations may be coarser, rangier, and fiercer than their domesticated counterparts.50 Norton applies this biological metaphor to cities that keep on functioning, after a fashion, even as they regress to the wild in the absence of government authority following a state collapse or during a war or natural disaster.

  Feral animals and plants do actually infest cities during and after conflict or disaster and are prevalent on the fringes of larger built-up areas.51 In Australian slang, the term ferals is also sometimes applied to humans who live in shanty settlements and reject the country’s metropolitan culture. (Australia, like most developed countries, is heavily urbanized and, because of the inhospitable terrain and climate of its interior, it also has a very high degree of littoralization: 89 percent of Australia’s population lives in cities and 82 percent lives within fifty miles of the sea.)52 There’s a radical, antiurban streak in “feral” subculture, and even though members may come from inner-city or middle-class areas, they favor a radical chic that makes a fetish of grassroots resistance to “the Man.”53

  A variant of this culture emerged in Britain during the August 2011 riots, in which marginalized city dwellers turned antiurban violence against the very cities where they lived. The destruction led Daily Mail columnist Richard Littlejohn to describe the rioters, who looted shops and vandalized symbols of authority and prosperity, as a “wolfpack of feral inner-city waifs and strays.”54 The notion of a lack of governability—the exact kind of thing Haussmann was trying to prevent in Paris, or the London authorities sought to address in their plan for the 2012 Olympics—is important here. It manifests in diffuse and apparently random patterns of crime and violence, and in self-marginalization by city dwellers who see themselves as victims of social injustice and economic exclusion, standing apart from the mainstream—in the city, but not of it—yet still maintain a high level of connectedness both with other members in their group and with the ebb and flow of the city itself.

  The London riots also suggest that the idea of a peripheral settlement or population (which we’ve so far been using mainly in a spatial sense, meaning people or districts that are located on the edge of town) can be broadened to include people who are marginalized or excluded in an economic, political, or cultural sense, even if they live in the physical center of a city. In this reading, which is of course extremely familiar to anthropologists or urban sociologists, the “urban core” of a city isn’t just the older, more central, downtown part of its built environment but also its economically, politically and culturally dominant terrain, the part of the city system that accumulates value at the expense of a periphery.55 In fact, better-off populations in many countries have self-segregated, moving farther out of city centers to gated communities or simply to islands of prosperity in the suburbs, abandoning the city’s center.56 Urban peripheries, in this sense, aren’t merely places on the physical outskirts of a city. Rather, they’re areas that are dominated, marginalized, exploited, victimized, or excluded by the core—wherever they happen to be physically located. The “ferality” of the 2011 London rioters thus wasn’t that of a core population attacking its own city but that of a marginalized population attacking a city it saw as someone else’s.57

  It’s worth pointing out here that the conditions Norton describes can exist at multiple levels within an urban fractal pattern, meaning that one level of an urban system—a few districts within a city, a few neighborhoods within a district, a few blocks within a neighborhood, or a few houses within a street—can become feral even while the broader system remains within limits, or can slip out of equilibrium even as the higher-level system is getting more stable. Conversely, ferality can bleed from one level of an urban system into another, such as when a city’s broader equilibrium is compromised by war or natural disaster, when a parent city or district loses the ability to integrate its component parts, when whole periurban districts effectively secede from the larger city (part of a broader process that some political geographers call “internal secession”), or when—as in the case of San Pedro Sula, discussed in Chapter 1—an urban metabolism loses the carrying capacity to process the by-products of the city’s flows.58 The city or district may not collapse, and as we will see, it may be far from anarchic or ungoverned, but as it slips from state control and “goes feral,” a series of overlapping threats emerges both for local residents and for the broader urban, national, and global systems
that surround it.

  Obviously enough, the very term city embodies peace, order, and tameness. English words that connote domesticity, peace, tranquility, development, and order, and which we use every day—words such as politeness, civilization, citizen, civility, civilian, urbane, and of course police—all derive from Latin and Greek words for the city (polis, urbs, civis). When Aristotle called man a political animal, he was referring to the predominantly urban habitat of our species: humans, he was arguing, are by their very nature “city-dwelling animals.”59 This idea of the city as the culmination of human development (literally, civilization) is so deeply embedded in our thinking that the notion of a feral city, moving backward in time and downward in social order, regressing to the warlike chaos of the wild—not a noncity but an anticity, a perversion of the natural way of things—can be deeply shocking.

  This, I suspect, is what lies at the root of our modern fascination with world-ending societal collapse, a scenario beloved of survivalists and cinematographers. Think of the box-office success of movies such as I Am Legend, Mad Max, World War Z, or 28 Days Later—or, indeed, the seductive appeal of any number of zombie apocalypse or dystopian sci-fi novels, comics, videogames, and television shows—all of which tap into a deep anxiety that underlies contemporary urbanized society. Urban theorists such as Stephen Graham argue that this anxiety is actually a direct result of the very processes of population growth, urbanization, and technological modernization that we’re examining here, so attacks on a city’s complex system-of-systems can be seen as a form of forced de-modernization.60 Graham quotes the architecture critic Martin Pawley, who wrote that “fear of the dislocation of urban services on a massive scale” is now “endemic in the populations of all great cities.”61

  The Battle of Mogadishu