Out of the Mountains

by David Kilcullen

  At the unit level, an organization that operates like this doesn’t have a command post that can be found and killed. Ever since J. F. C. Fuller’s Plan 1919, modern maneuver tactics have centered not on fighting and defeating each and every enemy combat unit but rather on finding and destroying the enemy’s command node. Writing at the bloody climax of World War I in 1918, Fuller argued that “the first method may be compared to a succession of slight wounds which will eventually cause [an enemy] to bleed to death; the second—a shot through the brain.”71 In a similar vein, Colonel John Warden’s “five rings” model of targeting analysis for airpower seeks to achieve physical paralysis by finding and targeting key nodes (centers of gravity) in an enemy system.72 But a fully decentralized swarm system like the one these Somali fighters employed has no brain, no central command node that can be killed. The swarm’s command system is distributed, rule-based, emergent, and thus embedded in the system itself, not tied to any one person, vehicle, or physical location. This suggests the uncomfortable possibility that even if TF Ranger had succeeded in killing General Aidid, the loss of its commander would have had a negligible effect on his organization’s ability to function.

  The Somali approach is also a very different solution to the same problem that led Lashkar-e-Taiba to adopt its remote-node command model for the Mumbai attack: where LeT made its command node invulnerable by putting it in another country and relying on Internet and satellite connectivity to connect the operations room to the assault teams on the ground, the Somali militia made their command node invulnerable by not having one at all. When I asked the SNA soldiers how their tactics differed from those of Shabaab and the various local militias, they laughed. “They are us,” they said with a shrug, pointing out that many of them—like many Shabaab fighters—had previously served with militias of one kind or another before joining the SNA.

  Long-Term Flows

  As well as these swarm tactics (which we’ll return to in Chapter 5), we’ve already noted the way that temporal rhythm and spatial logic affected the Mogadishu battle over the term of the city’s daily flow cycle. But there is a lower-frequency cycle also, a longer-term metabolic flow that shapes the urban environment in a place such as Mogadishu. This is the pattern of population movement, urbanization, and littoralization, occurring over decades, and it was this pattern that gave the city its structure, both in 1993 and today. This is obvious if we note that—in common with other organisms—the history of an urban organism is physically recorded in its structure, just as scar tissue, a lost digit, a callus, or a growth in a biological organism is a permanent structural manifestation of that organism’s past. Mogadishu today, like any other city or organism, embodies a physical record of its history.

  The Somali writer Nuruddin Farah, writing in 1998, brilliantly captured the long-term flows that have shaped Mogadishu. “If Mogadishu occupies an ambiguous space in our minds and hearts,” Farah wrote, “it is because ours is a land with an overwhelming majority of pastoralists, who are possessed of a deep urbophobia. Maybe this is why most Somalis do not seem unduly perturbed by the fate of the capital: a city broken into segments, each of them ruthlessly controlled by an alliance of militias.”73 Farah identified several waves of urbanization and coastal settlement that over decades drew in population from the hinterland, expanded the city, changed its social and political character, and created accretions of new peripheral settlements around the older coastal core:

  Before independence, huge numbers of Somalis, who could best be described as semi-pastoralists, moved to Mogadishu; many of them joined the civil service, the army and the police. It was as if they were out to do away with the ancient cosmopolitan minority known as “Xamari,” Xamar being the local name for the city. Within a short time, a second influx of people, this time more unequivocally pastoralist, arrived from far-flung corners to swell the ranks of the semi-pastoralists, by now city-dwellers. In this way, the demography of the city changed. Neither of these groups was welcomed by a third—those pastoralists who had always got their livelihood from the land on which Mogadishu was sited (natives, as it were, of the city). They were an influential sector of the population in the run-up to independence, throwing in their lot with the colonialists in the hope not only of recovering lost ground but of inheriting total political power. Once a much broader coalition of nationalists had taken control of the country, these “nativists” resorted to threats, suggesting that the recent migrants quit Mogadishu. “Flag independence” dawned in 1960 with widespread jubilation drowning the sound of these ominous threats. It was another thirty years before they were carried out.74

  These tensions, which arose from long-term flows of population, goods, and money, and a struggle for control of economic resources and nodes within Mogadishu—including, particularly, the port and the livestock and trading markets—manifested themselves physically in a patchwork of informal urban settlements, and socially in a pattern of fragmented territorial control across the city, with each group dominating its own area and the clans coexisting in an uneasy, shifting pattern of temporary alliances of convenience.

  Nuruddin Farah’s analysis here echoes the Palestinian historian Hanna Batatu’s comment on the urban-rural dynamic between Damascus and the rural hinterland of Syria, suggesting a widespread pattern of conflict between population groups that have traditionally dominated cities and the former peasants or rural dwellers arriving as migrants from the countryside. Batatu identifies this as a cyclical flow, “a phenomenon that repeats itself: rural people, driven by economic distress or lack of security, move into the main cities, settle in the outlying districts, enter before long into relations or forge common links with elements of the urban poor, who are themselves often earlier migrants from the countryside, and together they challenge the old established classes.”75

  Batatu’s notion echoes an old and very influential idea that came out of the coastal cities of North Africa in the fourteenth century—a theory of the circulation of elites put forward by the great Tunisian scholar Ibn Khaldun. As Malise Ruthven points out, Khaldun’s theory, sometimes called Khaldunian circulation, is based on the idea that “‘leadership exists through superiority, and superiority only through asabiyya—social cohesion or group feeling. In desert conditions, the social solidarity of the tribe is vital to its survival. If and when the tribes decide to unite, their cohesion puts the city-folk at their mercy. Inspired by religion, they conquer the towns, which are incapable of defending themselves, and become the rulers until such time as, corrupted by luxury and the loss of their group cohesion, they are in turn replaced by a new nomadic dynasty.”76

  This same cyclic flow seems to have been present in Mogadishu’s evolution. Indeed, Farah’s and Batatu’s analyses turn on its head one common interpretation of Somali history: namely, the idea that the intergroup competition, corruption, winner-take-all abuse of defeated opponents, and clan-based violence that Mogadishu experienced after the fall of the Barre regime in 1991 was primarily a symptom of state collapse. The popular notion is that this chaos emerged after Barre’s rule fell apart under the pressure of war, drought, and economic collapse. On the contrary, in Farah’s telling, it was the pattern of fragmented urbanization (producing marginalized garrison communities with patron-client connections to political leaders) and rapid population growth (with the resultant lack of resilience and carrying capacity in the city’s metabolism) that produced the violence and instability that eventually destroyed the state. In this version of events, Mogadishu didn’t become a feral city because the state collapsed; rather, the state collapsed because the city was already feral. Mogadishu’s very structure created a political and social space for the city’s own destruction at the hands of “a cast of borderline characters posing as city-folk leading armed communities of marginalised nomads. . . . The savageries visited on the city’s residents [were] masterminded by urbophobics already installed in Mogadishu, which for hundreds of years has lain under the en
vious gaze of people who hated and feared it because they felt excluded from its power politics.”77

  I just used the term “garrison community” in reference to Mogadishu. This expression is widely used in the Caribbean, on the other side of the world from Somalia, to describe the informal systems of security and order that have emerged in marginalized urban settlements in Jamaica. One district of Kingston—the coastal slum known as Tivoli Gardens—exemplifies the threats and challenges of yet another part of the spectrum that will affect the urbanized, coastal, connected environment of the future.

  III. Kingston: Garrison Communities and Nested Networks

  6:32 p.m., Monday May 24, 2010

  U.S. Embassy, Kingston, Jamaica

  As the sun set on a long day, Isaiah Parnell, chargé d’affaires at the U.S. embassy in Kingston, sent an Immediate cable to Secretary of State Hillary Clinton in Washington, D.C., with copies to the Central Intelligence Agency, Drug Enforcement Administration, Department of Justice, Special Operations Command, Southern Command, chairman of the Joint Chiefs of Staff, and U.S. embassies in Ottawa and London. His cable read, in part:

  At midday on May 24 the Jamaican Defence Force (JDF) launched an all-out assault on the heavily-defended Tivoli Gardens “Garrison” stronghold controlled by Christopher “Dudas” Coke, the alleged overlord of the “Shower Posse” international crime syndicate who is wanted to face extradition to the USA on drugs and weapons trafficking charges. . . . The JDF fired mortars and then used bulldozers to break through heavy barricades which Coke’s supporters had erected to block entry to the fortified enclave. As of 6:00 p.m. May 24, heavy fighting continued in Tivoli Gardens, and a fire was burning out of control in the adjacent Coronation Market. The JDF plans to continue operations through the night. Large numbers of women and children have fled the area. . . .

  Elsewhere in the metropolitan area, armed gangs attacked police stations, overturned vehicles, and erected roadblocks. The Hannah Town Jamaica Constabulary Force (JCF) station was destroyed by fire, and police were pinned down by gunfire at their Denham Town station. Armed gang members also surrounded and threatened to overrun the central police compound and central lock-up. In response to the escalating civil unrest, Prime Minister (PM) Bruce Golding declared a limited state of emergency, which took effect at 6:00 p.m. May 23 and is expected to last a month.78

  We’ve already seen how the Mumbai raiders nested within licit flows and illicit networks in the complex littoral of Karachi and Mumbai, and how Somali militias nested in the maze of a feral city, using adaptive swarming tactics to confront an American force that had confidently ignored the spatial and temporal flow of Mogadishu. A different example of this nested-network phenomenon—and one that illuminates another part of the urban threat spectrum we have been describing—is the pattern of criminal control within marginalized urban settlements such as Tivoli Gardens. Local nonstate armed groups may gain control of these districts and use their broader affiliations—both with offshore networks and with leaders at the city or national level—to nest within larger networks for protection.

  As Parnell’s Emergency Action Committee was meeting at the embassy on the night of May 24, Kingston’s waterfront was burning. Dozens of small groups of fighters from Coke’s Shower Posse and from neighboring allied groups were swarming toward the scene of the action—establishing barricades, ambushing police and military vehicles, and creating blocking positions to deny the government advance. Flatbed trucks belonging to Coke’s construction company (whose business was built on government contracts gained through his relationships with city politicians) had hauled in building materials to construct the barricades and fortified positions that were now under attack by the government’s own forces. Police stations, cars, and houses were burning, and a heavy firefight between police, army troops, and gang members with military-grade weapons (including AK-47 assault rifles, machine guns, and .50-caliber heavy sniper rifles) was raging throughout Tivoli Gardens and the surrounding settlements. Coke’s supporters had taken over the Kingston Public Hospital, violence had spilled into half a dozen districts across the city, and roads and airports were closed, cutting Kingston off from the outside world. Schools and businesses were shuttered and would stay closed for weeks. Kingston’s hospitals were treating dozens of injured civilians, many of whom would later die from gunshot wounds. The Jamaican government was mortaring, bulldozing, and assaulting its own capital, and the city was pushing back.

  According to the U.S. embassy cable, Kingston had become a war zone in the course of enforcing a United States extradition request against a single international drug trafficker. Coke’s network operated in New York, Toronto, London, and farther afield. Parnell had sent copies of his cable to the CIA and DEA because United States agencies were intimately involved in this operation: as Parnell’s team was composing the message, a Department of Homeland Security surveillance aircraft was flying over Tivoli Gardens, recording live video of the attack.79 JDF major Wayne Robinson’s master’s thesis, completed in 2008 at the United States Marine Corps Command and Staff College at Q uantico, Virginia, explored the application of American counterinsurgency tactics from Afghanistan to counter Jamaican organized crime: it became a key source for the JDF operation, which planners conceptualized as urban counterinsurgency.80 DEA advisers, U.S. and Canadian Special Operations Forces, and U.S. surveillance drones had all helped prepare Jamaican forces for the operation or were supporting it in real time.81

  This assistance came with strong international pressure on a Jamaican government that was extremely reluctant to comply with the American extradition request. Prime Minister Golding had already delayed action for more than nine months, claiming that the evidence against Coke had been obtained illegally through unauthorized U.S. surveillance of Coke’s electronic communications. Golding, leader of the center-right Jamaican Labor Party (JLP), represented Tivoli Gardens in parliament and allegedly maintained a long-standing and close relationship with Coke and, before him, with his father and brother. The Shower Posse kept the peace, regulated criminal activity, and mobilized the district’s residents to support the JLP in elections, making this a supersafe JLP constituency. In turn, JLP politicians such as Golding ensured that the district received lucrative government contracts and public services.

  In the event, Christopher Coke escaped arrest during the invasion of his district, known as Operation Garden Parish, but the military occupation of Tivoli Gardens, under a national state of emergency, went on for weeks. It left parts of the city in ruins, disrupted Kingston’s port, railway, and airport (all located close to Tivoli Gardens and all—especially the port—influenced by Coke’s network), led to more than five hundred arrests, displaced thousands of local inhabitants, killed at least seventy-three civilians and six police and military personnel, and injured many more.82 The upheaval cost Golding his position and contributed to the JLP’s landslide December 2011 election defeat at the hands of its archrival, the left-wing People’s National Party (PNP). Christopher Coke was eventually captured a month after the start of Operation Garden Parish. Police found him hiding in the trunk of a car while attempting to flee the area, which had been cordoned off and subjected to weeks of strict curfews, searches, and police and military saturation patrols. Coke was extradited under heavy guard, tried in New York on weapons and drugs charges, found guilty, and on June 8, 2012, sentenced in federal court to twenty-three years in jail.83

  But to frame this series of events solely as a law enforcement action to arrest an international drug trafficker is entirely to misunderstand what happened in Tivoli Gardens throughout the summer of 2010. Likewise, to characterize the Shower Posse solely as the U.S. embassy cable did—as an “international criminal syndicate”—is to describe only a small part of the group’s role. The Shower Posse was (and is) both local and transnational, a nonstate armed group that nests within a marginalized and poor but tightly knit local community in Kingston, yet is conne
cted both to the Jamaican government and to a far broader international network. It was and is as much a communitarian militia, social welfare organization, grassroots political mobilization tool, dispute resolution and mediation mechanism, and local informal justice enforcement system as it is an extortion racket or a transnational drug trafficking organization. Drug trafficking doesn’t define what an organization like Coke’s group is; it’s just one of the things the group does. To grasp this deeper background, we first have to understand the origins of Tivoli Gardens and the other garrison districts of Kingston.

  What Goes Around Comes Around

  In the words of one of Coke’s henchmen at his subsequent trial, garrison district is the Jamaican term for an urban or periurban “neighborhood whose members are armed by the leader of the community, and also a neighborhood that is loyal to and affiliated with one of the major Jamaican political parties . . . in the case of Tivoli Gardens, the Jamaica Labour Party (JLP).”84 Tivoli Gardens is the oldest of Kingston’s garrisons, and its history shows how Jamaican clientelism, political populism, gang violence, and international connectedness have shaped (indeed, in large measure created) the very urban landscape and flow of the city, and have in turn been influenced by that landscape.