Urban Insurgency in the Twenty-First Century: Smaller Militaries and Increased Conflict in Cities

Remy Maduit | Authors published

Book in English | Livre en français | Synopsis du livre | Articles in English | Articles en français

Urban Insurgency in the Twenty-First Century:
Smaller Militaries and Increased Conflict in Cities

Anthony King is a Professor and Chair of War Studies at the University of Warwick, UK.

In the past two decades, urban insurgencies have become a major concern for both scholars and practitioners. The longest and most intense civil conflicts have taken place in cities, not in the field. Up to now, this proliferation of urban insurgency has been attributed to two central factors: demography and asymmetry. It is widely assumed that because half the world’s human population lives in cities, in often appalling conditions, the conflict has inevitably migrated to these environments. [i] In addition, dense urban areas—especially ones with rapidly growing slums—now offer insurgents the best opportunities for evading the technological superiority of state forces. [ii] As a result, many are convinced that, in the next two decades, insurgents will increasingly operate not just in cities, but in megacities of 10 million inhabitants or more.

These explanations are sensible. Rapid urbanization has precipitated discontent, and urban areas have offered cover for insurgents. However, this article explains the rise of urban insurgency in the past two decades. Rather than focusing on the civilian population of cities, the insurgents, or advanced technology, it concentrates on state forces themselves and, in particular, on their sheer size. Force size may seem banal and unimportant, but in fact, it has typically played a very significant role in the character of military operations. [iii] The article addresses this lacuna in the insurgency literature, exploring the connection between the size of state forces and the prevalence of urban insurgency. It claims that the contraction of military forces in the last few decades has played a major role in facilitating the rise of urban insurgencies. In the twentieth century, when states possessed massive armies, they could dominate urban areas through the force of numbers. However, as forces have downsized, states have struggled to control urban areas, allowing insurgents to operate inside cities with the freedom that was once denied them. In the twenty-first century, cities have become excellent environments in which to operate—not only because they have become so big but also because security forces are now so much smaller than they once were. The contraction of state forces has amplified the effects of demography and asymmetry. Untroubled by the constant presence of security forces, insurgents have been able to maintain large home bases inside urban areas, enabling them to exploit their increasingly advanced technology asymmetrically. [iv]

To show the connection between declining force size and the rise of urban insurgency, we organized the article into two parts. The first part will show that, contrary to some contemporary presumptions, urban insurgencies were very common in the twentieth century. However, because of the sheer mass of state security forces, civic uprisings were typically short-lived; insurgencies were recurrently driven out of the city early in the campaign. An examination of insurgency literature at the time shows that insurgent leaders themselves fully recognized this problem of security force numbers and, therefore, often actively shunned cities. This section of the article concludes with an analysis of the British Army’s suppression of the Irish Republican Army in Belfast in 1972 as an empirical example of the effects of mass on the urban insurgency. In the second part, the article turns to examine the twenty-first century, exploring force ratios in Iraq and Syria to show how the relative lack of state forces allowed insurgencies to hold urban terrain in both theatres—in stark contrast to the previous era.

Force size and urban insurgency

Demography and asymmetry are key factors in explaining urban warfare in the twenty-first century. Sometimes, these variables provide a very good explanation of the phenomenon. The battle of Homs in 2012, at the start of the Syrian civil war, for instance, would be an almost ideal case study; here, the protests of the disfranchised descended into violent conflict.[v] However, it is questionable whether demography and asymmetry always provide sufficient explanations for the contemporary urban insurgency. The urban population has indeed increased dramatically since the 1960s. Yet urban areas were largely in the past too. For instance, some of the most famous insurgencies of the twentieth century took place in countries with many large towns and cities. In 1957, Algiers had a population of 900,000, with some 70,000 in the Casbah; Nairobi had about 100,000 inhabitants in 1954, Aden about 200,000 in 1966, Saigon 2 million in the 1960s, Rio de Janeiro 3 million in 1950, and Belfast over 300,000 in 1970. Indeed, some insurgent leaders, such as Abraham Guillén, believed that urban areas were easily big enough to offer excellent opportunities for guerrilla operations. [vi] It might be thought that the demographics of these cities could sustain an urban insurgency. Yet successful urban insurgencies were rare in the twentieth century. It is a conundrum that they were not more common or more successful. If demography alone were so decisive, one would expect that insurgents would have exploited cities more in the twentieth century than they did.

There are problems with the asymmetric argument, too. Many scholars have argued that the urban environment is optimal for concealment and protection. It is often very difficult for security forces to infiltrate hostile urban terrain. For instance, the favelas in Brazil are almost impenetrable to the security forces, since each comprises a dense maze of streets. [vii] Wars in Syria and against the Islamic State in Iraq and Syria (ISIS) show how insurgents have been able to build formidable urban strongholds against state forces. It is not true that states always struggle to defeat insurgents in urban areas. During the battle of Sadr City, for instance, US forces could track Shi’a fighters with almost complete fidelity; in a three-week battle, they killed 600, mainly by drone strikes. [viii] Precisely because they were operating within the confines of a narrow urban area, the Shi’a militias were vulnerable. Cities are not always the self-evidently best sanctuary for insurgents.

Neither does the proliferation of advanced weaponry and information communications technology alone explain why insurgents have thrived in cities in the past two decades. Better-equipped insurgents have certainly been able to hold urban areas against state forces more easily. Yet insurgents have been able to use complex weapons only when they already control secure areas in which they can store, maintain, and operate them. In Mosul, for instance, ISIS became a formidably equipped opponent because it converted many of the city’s facilities for war production. Hezbollah could defend southern Lebanon in 2006 because it controlled and fortified towns and villages from which it could use its sophisticated weaponry. [ix] The question is not so much why insurgents have more advanced weaponry now, but why urban insurgents today have been able to create no-go urban enclaves more easily than their counterparts in the twentieth century.

In his recent monograph on non-state warfare, Stephen Biddle has suggested an answer. [x] In an important early section of the book, he discusses force-to-space ratios to argue that as insurgents ‘exclude state forces from base areas’, they can organize themselves for more complex military operations. In short, as insurgents have seized and held terrain, they have been able to exploit heavier and better weapons. Once insurgents possess advanced weaponry, they have often been able to hold urban territory more effectively. However, for Biddle, the possession of terrain precedes the possession of advanced weaponry: ‘A mid-century non-state actor who tried to take and hold ground against a state opponent would typically have been crushed by the weight of numbers’. [xi] Because state forces were so massive, insurgents could not hope to hold urban terrain against them. Here, Biddle explicitly recognizes the relationship between force mass and insurgency. Massive state forces could swamp areas, generating force densities in contested neighborhoods that made it impossible for insurgencies to operate freely.

This observation has profound implications for urban operations today because, in a process that began in the 1970s and sped up after the Cold War, state forces have contracted, sometimes radically. [xii] As state forces have declined in size, insurgents have been increasingly able to seize and hold urban terrain. As a result, they have been able to mount complex military operations inside cities with advanced weaponry. This article explores whether, in the past two decades, urban insurgency has become more common and serious substantially because state forces have contracted. By highlighting force size, instead of just re-emphasizing demography and asymmetry, it may be possible to reach a more comprehensive understanding of why urban insurgency has become the norm in these past two decades, and why it is likely to remain so in the coming decade.

Twentieth-century urban insurgencies

Kilcullen and many others claim that insurgents have come out of the mountains in the past 20 years. Urban insurgency was a constant feature of the twentieth century. Throughout the Cold War, Britain, France, and the United States conducted a series of counter-insurgency campaigns in Africa, the Middle East, and south-east Asia, most notably in Palestine, Malaya, Kenya, Cyprus, Algeria, Oman, Yemen, Vietnam, and Northern Ireland. Of course, the vast bulk of the operations mounted during these campaigns were in the field; most British, French, and American troops fought in arid mountains, dense jungles, or swamps. In the past ten years, there has been a major academic revision of work on these conflicts, as scholars have refuted earlier, sanitized accounts of these campaigns. [xiii] Yet it would be quite wrong to think that these campaigns lacked an urban dimension. In almost every single case, the conflict involved important urban fighting. A brief history of some of these campaigns proves the point.

Before the Second World War, the British Army had suppressed an Arab revolt in Palestine.[xiv] Almost immediately afterward, it was drawn back into a second conflict in the Holy Land; this time, against the Jewish struggle for independence.[xv] The war reached its climax at the battle of Jaffa, which Jewish militias had seized from Palestinians. The British forces’ inability to regain the city from Jewish insurgents played a significant role in their withdrawal.[xvi] Palestine was not exceptional. British campaigns in Malaya, Kenya, Cyprus, and Yemen also involved important urban fighting. Malaya, for instance, is rightly remembered for its jungle fighting; this is where most communist guerrillas were eventually located. The struggle became one for the countryside and villages. Yet initially, after the Second World War, the Communist Party of Malaya infiltrated the towns and cities of the country. Between 1948 and 1951, overwhelming force robustly expelled from these populated areas them. During this period, the colonial authorities employed 40,000 British and Commonwealth troops, 67,000 police, and 250,000 Home Guard against 8,000 insurgents in a country the size of England. These forces could resettle the population and dominate urban areas through sheer force of numbers. Only then, after 1951, when the crisis had been averted, did the insurgency become an only rural phenomenon, with most of the fighting taking place in the jungle. [xvii] Only after 1951 did it become the classic rural insurgency as it is now remembered.

A similar pattern is observable in Kenya. The Mau Mau originated as an insurgent group in the early 1950s among the Kikuyu, Meru, and Embu tribes in the White Highlands. The central dispute, which precipitated the uprising, was about land distribution. [xviii] Yet, while emerging in the White Highlands, the Mau Mau quickly infiltrated Nairobi where, by 1954, they had many cells operating in the city with important local support. Recognizing the danger of a general insurrection against the colonial authorities, General ‘Bobbie’ Erskine, Commander-in-Chief East Africa, mounted a major operation to clear the city of insurgents. On 24 April 1954, 20,000 British and Commonwealth troops, supported by police and Home Guard units, flooded the city as part of Operation Anvil. [xix] The Mau Mau continued to fight for four more years in the forests of the highlands. In Cyprus, too, Colonel Grivas’s Ethniki Organosis Kyprion Agoniston (EOKA, National Organization of Cypriot Greek Fighters) campaign from 1954 to 1958 involved a very significant urban element. [xx] The conflict in Cyprus was not self-evidently an urban one, but Grivas used his urban cells to plant bombs and assassinate military personnel in Nicosia in concert with rural groups. [xxi] In Aden, a similar dynamic was observable in the Crater. [xxii] In each of these cases—Malaya, Kenya, Cyprus, and Aden—urban guerrillas were an important part of the insurgency. However, mostly, they were driven out of urban areas by the sheer number of British and colonial troops.

The British postwar counter-insurgency experience was anything but unique. The Algerian war of independence also involved very significant urban operations. [xxiii] They fought the key battle of the war in Algiers between Brigadier-General Jacques Massu and his 10th Parachute Division and the Front de Libération Nationale (FLN). The battle began in January 1957 when, following a series of bomb attacks, 4,600 paratroopers were deployed on the streets. [xxiv] Early in that month, the paratroopers forcibly broke a general strike that the FLN had organized, levering open the shutters of shops and physically dragging shopkeepers to their stores. They followed this with an intense ten-month operation in which the paratroopers dominated the Casbah, arresting thousands of Algerian suspects, 3,000 of whom ‘disappeared’. [xxv] On 8 October, the French forces finally cornered and killed the final FLN terrorist leader, Ali la Pointe. The battle of Algiers was over; but the struggle continued until 1962 out in the mountains and fields, where guerrilla bands fought French troops. As in Kenya and Malaya, colonial forces had driven the insurgents out of the city early in the conflict, although the methods they used would later lose them in the war.

Vietnam in the 1960s was an overwhelmingly rural country. It was all but inevitable that most of the war would be fought in the mountains, jungles, and paddy fields, not least because the popular base of the insurgency lay in the south’s peasantry. Despite its political and military ineptitude, the US-backed South Vietnam regime dominated urban centers at the very end of the war. However, there was one notable but telling exception to this pattern: the Tet Offensive and, specifically, the battle of Hué in February and March 1968. The battle of Hué was, in fact, a conventional fight between the American and South Vietnamese forces and the North Vietnamese Army. The Communist Party deployed the 5th Regiment of the North Vietnamese Army (NVA) into the city. Assisted by local Vietcong, this regiment infiltrated successfully and seized control of the Citadel on 31 January. [xxvi] There followed a bitter two-week battle, in which a US Marine regiment and an Army of the Republic of Vietnam battalion eventually retook the Citadel, eliminating the NVA force. [xxvii] The battle of Hué was interesting because, like the conflicts in Algiers, Kuala Lumpur, and Nairobi, it demonstrated that in the twentieth century it was extremely difficult for an insurgent force, even one the size of the Vietcong, to control urban areas.

It is possible to identify a pattern across twentieth-century insurgencies. In most campaigns, there was very significant urban fighting. Insurgents frequently operated in cities and towns, and some terrorist cells were highly successful. In each case, the rural insurgency predominated; the urban guerrilla was, at most, a supporting element of the uprising. The regime was strongest in towns and cities; the security forces dominated there. It was, therefore, hard for insurgents to gain a foothold in urban areas, or to mount successful operations from them.

Why did guerrillas avoid cities in the twentieth century?

Insurgent literature

During the twentieth century, there was a rich literature on the insurgency. Its primary aim was to educate potential revolutionaries. However, that literature also provides a very good answer to the question of why urban insurgency was so difficult in the twentieth century. Most of this literature asserted that insurgency was a rural phenomenon not because of demographics, but because state forces were so massive. For instance, the central principle of Mao’s insurgency doctrine was dispersal; the effective guerrilla exploited space to extend opponents, enabling them to be struck at their weakest point. Mao learned this principle from bitter experience. When he attempted to concentrate pre-emptively against superior state forces in 1934, he was defeated badly and forced into the Long March. As a result, faced with massive state forces, he preferred the field. His doctrine presumed a very large, remote theatre of operations that even superior regime forces could not hope to cover.

Che Guevara’s approach to insurgency was similar. He, too, strongly advocated the exploitation of an inaccessible hinterland as a requirement for success. Against those who advocated urban insurrection, he claimed: ‘The situation in the open country is not so difficult, where the armed guerrillas can support the local people and where there are places beyond the reach of repressive forces.’ [xxviii] While they could inundate towns and cities, security forces lacked the mass to penetrate every part of the countryside. Indeed, for this very reason, Fidel Castro himself had declared: ‘The city is the cemetery of revolutionaries and resources.’ [xxix] Castro noted bitterly that whenever guerrilla leaders had gone to the city to attend political meetings or engage in military action, they had been apprehended.

Mao and Guevara were certain that, in the twentieth century, insurgencies had, of necessity, to be peasant movements. It was the only way to avoid state suppression. The other prominent insurgent theorists of the time adopted a different view. Carlos Marighella was, perhaps, the most important of them, with his famous manual written specifically for the urban guerrilla. Marighella’s Minimanual focused on the critical role which the urban guerrilla might play in a revolutionary war. Precisely because urban guerrillas operated in cities, at the heart of government, the effects of their terrorism were amplified. They could ‘threaten the triangle in which they maintained the Brazilian state system and North American domination in Brazil, Rio, Sao Paulo, and Belo Horizonte’.[xxx] The city was a more demanding environment than the countryside in terms of the human population, the physical terrain, and the security forces. Urban guerrillas had to be more adaptable, flexible, and cunning than armed peasants if they were to stage their attacks, bank robberies, and assassinations effectively. Marighella identified particular characteristics of the successful urban guerrilla, who had to be not only ‘a good tactician and a good marksman’ but also ‘a person of great astuteness’, who displayed ‘moral superiority’. [xxxi] He or she must also be “familiar with the avenues, streets, alleys, in and outs, and corners of the urban centers, its paths, and shortcuts, empty lots, its underground passages, its pipes, and sewer system; the urban guerrilla safely crosses through the irregular and difficult terrain unfamiliar to the police”. [xxxii]

While Marighella gave the urban guerrilla a distinctively prominent role—far more than Guevara or Mao—even he did not believe that the urban guerrilla alone could bring down the government or liberate Latin America from US hegemony. This becomes clear only at the very end of the manual. There he confirms that while the urban terrorist might bring militarists and dictatorships to ‘the brink’ the rural insurrection would always prove decisive. Rural guerrilla warfare is the ‘backbone of the revolution’; ‘from this backbone will come the marrow of the revolutionary army of national liberation’. [xxxiii] In the end, Marighella does not contradict Mao or Guevara; he does not suggest that the urban terrorist can replace the traditional guerrilla. Rather, especially in the specific metropolitan conditions of Brazil, he saw an opportunity for urban terrorists to catalyze the revolution. He does not discuss the reasons for this, but they are implied in his text: urban guerrillas could not mount a revolution on their own because state security forces were too massive and powerful in the city. Indeed, the urban guerrilla had to be cunning and covert, precisely because of the potency of the security forces in urban areas. Ironically, Marighella’s career proved the point. He was shot dead in a police ambush in São Paulo on 4 November 1969, and his colleagues in the Action for National Liberation were arrested.

Abraham Guillén represents the extreme of this trend: he genuinely believed that an urban insurgency could work. Guillén was originally Spanish but, having been imprisoned by Franco, in 1948 he emigrated to Argentina where, in 1962, he became the leader of a radical Uruguayan terrorist group, the Tupamaros. Based on his experiences in Uruguay, Guillén believed that the urban guerrilla could be the prime revolutionary force: ‘It is absurd to engage in mountain guerrilla warfare when revolution can be decided in a few hours or days in the cities.’ [xxxiv] Guillén certainly believed that cities in the 1960s were sufficiently large and discontented to serve as the bases for insurgencies. Yet, he also recognized the difficulties of operating in the metropolises of Latin America in the 1950s and 1960s. Guillén did not advocate mass urban uprising, for it would be easily crushed: ‘If revolutionary war does not spread to over one district, then it involves a linear battle that lends itself to defeat before the massive use of tanks, light artillery, cavalry, and infantry.’ [xxxv]

Urban guerrillas should form themselves into tiny, highly professional cells capable of moving about the city at will, never compromising themselves by staying in one location or with one population too long. These skilled urban terrorists could mount attacks across the city, demonstrating the weakness and corruption of the government, so that the people might ‘arise en masse like an enraged lion’. [xxxvi] For Guillén, the city offered a unique political advantage. As long as they did not tether themselves to a location, the guerrillas could remain anonymous, evading detection and arrest. In this situation, urban guerrillas overcame the weakness of the traditional insurgency. They could mobilize the citizenry without exposing themselves or the people to government repression.

It is an ingenious idea. Yet it has never succeeded. Indeed, Guillén recognized the great vulnerability of the urban guerrilla. Since state forces were strongest in towns and cities, Guillén fully accepted that a territorial strategy must fail, because the security forces would always have enough troops to counter such an offensive. As Guillén himself notes, they would simply deploy tanks against the guerrilla strongholds. The crushing of the great urban insurrections in history illustrates the point very well; in 1830, 1848, and 1871, massive state forces easily suppressed the Parisian insurrectionists. [xxxvii] For all his advocacy of the urban as not just the primary theatre of insurrection, but the only necessary one, Guillén ultimately admitted how difficult it was to mount and sustain a revolutionary campaign in the urban domain in the twentieth century. States’ security forces were so many that it was simply impossible for insurgents to operate effectively in this environment, so close to the seat of government, its police stations, barracks, courts, and prisons.

The battle of Belfast

The cases of Malaya, Kenya, Aden, Algeria, and Vietnam show how massive state forces crushed nascent urban insurgencies in line with insurgent literature. However, to affirm the correlation between force numbers and urban insurrection suggested in that literature, it is useful to consider one historical example in depth. It would be possible to consider more closely one example cited above: Nairobi, Algiers, Aden, or Nicosia, for instance. However, as an alternative example, Northern Ireland provides rich, additional empirical evidence to support the argument. The ‘Troubles’ is an important case because it was the longest counter-insurgency campaign (1969–96) in which any western force was involved in the postwar period. Although casualties were light in comparison with Vietnam or Algeria, over 3,000 security force personnel and civilians were killed during the Troubles, and both Belfast and Derry/Londonderry were seriously damaged in the course of the conflict. It was a major campaign worthy of attention.

It might be thought that since Northern Ireland is part of the United Kingdom, with all the political, administrative, and intelligence advantages that were conferred on the British state, the requirement for massive security forces would be much less than in the colonial campaigns of the 1950s and 1960s. Indeed, Catholics comprised only a third of the population of Belfast in 1970. Northern Ireland is empirically pertinent in other ways, too. Although the British Army was a large force of 190,000 soldiers in 1969,[xxxviii] it was fully professional; conscription had been abolished in 1960. With better-trained volunteers, they might think that it would have been able to operate more efficiently in Belfast than it had in the colonial struggles of the 1950s. Sheer size might have been less noteworthy. Yet, despite all these offsetting political and military conditions, the struggle over Belfast between August 1969 and July 1972 exemplifies the dynamic identified in insurgent literature very well. Numbers played a crucial role.

The history of the Troubles is well known. In the 1960s faced with systematic discrimination in jobs, housing, and the law, the Catholic community engaged in marches and protests for civil rights. [xxxix] These marches escalated into open confrontations between the two sectarian communities. In July 1969, the Protestant marching season precipitated a new level of violence in both Derry/Londonderry and Belfast. By August, there were major riots along Falls Road in Belfast, in which Protestants and Catholics each sought to drive the other out of their neighborhoods. Many houses, especially Catholic ones, were burned out. [xl] The Royal Ulster Constabulary (RUC) subsequently lost control of these riots and, exhausted, called for military support. On 14 August 1969, the British government ordered the British Army onto the streets.

Initially, the British Army brought order to Belfast. The troops successfully protected the Catholic areas from the Protestants and limited the predations of the RUC. Although the IRA was always opposed to their deployment, most Catholics were originally supportive of the British Army. However, by the summer of 1970, the situation had changed entirely. [xli] By January 1971, only 7,742 troops had been deployed to the province; [xlii] overstretched, the troops often resorted to excessive violence. The army was no longer seen as a neutral party—still less an ally—but as part of the oppressive security forces of the Protestant-dominated regime in Ulster. [xliii]

As a result, from 1970 both wings of the IRA, the Officials, and the Provisionals, created no-go areas in Belfast and Derry/Londonderry. By 1972, the northern end of Falls Road was a notable Catholic ghetto. The ghettoes had an important military function. When the security forces tried to enter these areas, Republican terrorists had the perfect opportunity to attack troops or provoke British soldiers into retaliating in ways that would alienate Catholics. These areas also allowed the Provisional IRA, in particular, to mount a systematic attack on Belfast’s commercial and business district. From March 1971, the Provisionals bombed the city center to inflict economic damage on Ulster to show that the province was ungovernable. On 9 July 1972, ‘Bloody Friday’, the Provisionals detonated 20 devices in Belfast city center, killing nine people and wounding 130.

Because of Bloody Friday, the British government ordered the army to bring Londonderry and Belfast under control. The no-go areas in Belfast (and Derry/Londonderry) were to be eliminated. Operation Motorman was the result. At 0400 on the night of 30–31 July, the operation began. [xliv] Troops flooded the two cities. [xlv] Royal Engineers’ armored vehicles (effectively armored bulldozers which were mistaken for tanks) removed illegal barricades, while 12,000 extra troops were deployed into the province. Fifteen additional battalions reinforced the garrison already in Northern Ireland; four battalions moved into West Belfast itself to support the seven battalions already there. [xlvi] There were 22,000 British troops in the province, supported by a further 6,000 regular and reserve RUC police officers and 844 personnel from the newly formed Ulster Defence Regiment. [xlvii] In Belfast, this force cleared the Falls Road and built a network of patrol bases throughout Republican areas. At a stroke, the security situation in Belfast had been transformed. Although riots were common and patrolling was dangerous throughout the 1970s, the nascent enclaves in the Catholic and Protestant areas were eliminated and the security forces took back control of the streets permanently. In place of informally erected barricades, a new network of patrol bases and observation points was the principal security architecture within the Catholic areas after Motorman. Only the allowed ‘Peace Walls’ between the two communities remained. [xlviii]

Belfast affirms the insurgency literature of the postwar period. In response to Motorman, the Provisional IRA had no choice but to change its approach. They continued to plant bombs in Belfast and Derry/Londonderry; but, more significantly, they mounted a terrorist campaign on the UK mainland, clear in the bombings in Guildford, Birmingham, and, later, the City of London, Canary Wharf, and Manchester. Meanwhile, the Provisionals’ insurgency moved south to the border, to the bandit country of South Armagh. There, with solid local support and a porous boundary, Provisional IRA cells could fight a war of attrition against the security forces. In the 1970s, the Troubles imitated the geography of Algeria, Kenya, and Malaya. Having lost a battle in the city, the Provisionals continued to challenge the British in the countryside—just as Guevara, Mao, and Marighella had recommended. The Troubles show that in the twentieth century, while urban guerrillas often played an important role, they were always extremely vulnerable to counter-action by massive state forces, which could normally drive them out of cities.

Force size and the twenty-first-century insurgency

The insurgents of the postwar period emphasized the difficulty of mounting an urban insurgency; there were simply too many security forces. Counter-insurgents of the same era also fully acknowledged that force ratios played an important role in their operations. To interdict insurgents, it was vital to have enough troops to dominate the terrain physically, maintaining constant contact with the civilian population. For instance, Robert Thompson and David Galula, the pre-eminent British and French counter-insurgents of the 1950s and 1960s, were very conscious of combat ratios. Thompson and Galula had served in Malaya and Algeria, respectively; Thompson had joined the British Advisory Mission in Vietnam. They had both written classic texts on counter-insurgency based on their experiences. Thompson thought it impossible to put a precise figure on the optimum ratio of counter-insurgents to insurgents but suggested based on experience in Malaya and Vietnam that approximately 20 counter-insurgents were required for every insurgent. [xlix] Galula claimed that a ‘ratio of the force of ten or twenty to one between counter-insurgent and insurgent was not uncommon’. [l] Roger Trinquier, who had served in Algiers in 1957 running counter-terrorist squads, similarly emphasized the importance of numbers; he noted that in Algiers, 300,000 French soldiers had fought 30,000 insurgents. [li] Counter-insurgents in this period often worked off a rough force ratio of one soldier to every 50 civilians in the intense period of a conflict as the minimum requirement for success. In 1957, the French deployed an extra 4,600 paratroopers into the Casbah of 70,000 civilians to reinforce the already massive security presence in the city—a force ratio of 1:15. During Operation Motorman, the British state deployed approximately 6,600 combat troops into Belfast, alongside about 2,000 RUC officers, to control a hostile Catholic population of 100,000: one soldier to every 15 Catholics. [lii] With these numbers, both the French and British states were able to drive a nascent urban insurgency out into the countryside.

Force ratios have been inverted in the twenty-first century. For western states, the main reason for this is obvious. From the early nineteenth century until the end of the twentieth, western states relied on conscription to generate mass forces when they required them. Today, almost all western states have abandoned national service in favor of all-volunteer professional militaries. [liii] This transition has improved the tactical performance of the armed forces, but it has also ensured that professional forces are much smaller, as they are more expensive to maintain. Western armed forces are now about a third or even a quarter of the size that they were in the Cold War. The reduction of military forces has played a very important role in recent military campaigns because, in stark contrast to the postwar era, western states cannot hope to inundate cities with troops; they lack the numbers. This is a major problem. Although advanced surveillance technology has certainly helped, urban operations still require a substantial and permanent physical presence on the streets to maintain public order, reassure the civilian population, gather intelligence, and restrict insurgent freedom of movement, as they did in the postwar period. It is noticeable that, despite all the technological advantages of American forces, the central tenet of the famous US counter-insurgency manual, FM 3-24, was the necessity of deploying sufficient troops into urban areas: ‘The vast and necessary resources spent for security by the counterinsurgent always dwarf those of their opponents, and successful COIN [counter-insurgency] often requires a very high force ratio.’ [liv] So, while today’s western troops might be better trained and more disciplined than their predecessors in the 1950s and 1960s, they have confronted a situation that no amount of tactical skill or technology can resolve: they still need to dominate the streets. In recent campaigns, western security forces no longer outnumber insurgents. Insurgents have often outnumbered security forces. No matter how effective individual units have been in comparison with their twentieth-century, often conscripted predecessors in Kenya, Algeria, and Malaya, western forces have operated at a major disadvantage; they have lacked the numbers required to pacify urban areas.

Iraq

The Iraq War showed the problem clearly. For instance, in the cities of Iraq after the invasion, the British never came remotely near the force ratios they had enjoyed in the twentieth century. In Basra in the period 2004–2008, the British experience contrasted drastically with that in Belfast between 1969 and 1972. At the height of the insurgency in Basra, in 2006–2007, the British commanders of Multinational Division South-East had only a single British Army battalion, along with the Iraqi Army and (often unreliable) Iraqi police at their disposal to secure a city of 1.2 million. General Richard Shirreff, who commanded Basra and had also been a soldier on the streets of Belfast in the 1970s, offered an illuminating comparison between the two campaigns:

The single battalion commander responsible for a city of 1.3 million people told me he could put only 13 half platoons or multiples on the ground, less than 200 soldiers on the ground, in a city of 1.3 million. You compare that, for example, with what I recall, as a young platoon commander in West Belfast in the late 1970s when there was a brigade on the ground. [lv]

The problem was compounded by the fact that while there were some Iraqi Army units in the city—2,300 from the 10th Iraqi Division took part in Operation Sinbad in September 2006[lvi]—the Iraqi police had been infiltrated by the Shi’a militias and were unreliable. It was questionable whether they could count the police as part of the security forces at all. In late February 2007, they sent two Iraqi battalions to Baghdad. In radical contrast to Motorman, the British were, according to one commentator, operating with a force ratio of one soldier to 370 civilians in Basra in 2006–2007. [lvii] This figure excludes the Iraqi Army and Iraqi security forces, but it is close to the truth, and it was one reason the campaign has been the subject of severe US criticism. [lviii]

Until the surge in 2007, Americans also struggled with force densities, which fell far below the accepted ratio. They consequently struggled to control the streets. For instance, in 2006 in Baghdad, a city of 5–7 million inhabitants, [lix] the US Army deployed two brigade combat teams, of about 10,000 soldiers. These soldiers augmented an important Iraqi security force and Iraqi Army presence. For Operation Scales of Justice in March 2006, the Iraqi government committed to providing 26,000 Iraqi security force personnel (national and local police) and 11,000 Iraqi Army troops. Only 2,000 of the troops ever arrived; because of this drastic shortfall, along with the poor conduct of the Iraqi force, the operation failed. [lx] About 30,000 Iraqi and American security forces took part in the operation. In June, the coalition staged a follow-on operation, Operation Together Forward. That operation is a good data point for force densities in Baghdad in 2006, as it occurred at the height of the insurgency/civil war but before the surge. The participants were 13 Iraqi Army battalions, 25 Iraqi national police battalions, and 10 coalition battalions; this amounted to 21,000 Iraqi police, 13,000 Iraqi national police, 8,500 Iraqi Army soldiers, and 7,200 coalition troops: 49,700 in all. [lxi] The performance of the troops and police was again disappointing, with the police often acting as proxies for Shi’a militias. Yet even had the Iraqi security forces performed well, the force ratios in the city during this mini-surge were not propitious: about one member of a security force for every 100 inhabitants.

In Baghdad, Ramadi, and Fallujah, the American forces thickened up their force densities and increased the size of the Iraqi security forces dramatically as part of the surge. By the end of December 2006, Multinational Security Transition Command—Iraq had trained 187,800 police officers, many of whom they deployed to Baghdad; [lxii] this number increased dramatically in 2007–2008. By the mid-point of the surge in Iraq in 2007–2008, security forces of all types (coalition, Iraqi, Sons of Iraq, and private security contractors) outnumbered insurgents by ten to one or more and enjoyed a troop: population ratio of 1:50; roughly 600,000 security forces of all types operated amid an Iraqi population of 30 million. In Baghdad itself, under the Baghdad Security Plan, Operation Fardh al Qanoon, the United States deployed four brigade combat teams, three ground-holding brigades, and one Stryker brigade in reserve. The US military presence increased to about 20,000 troops in the city; two other brigades were deployed in belts outside it. In addition, by late 2007, Multinational Division–Baghdad commanded seven Iraqi Army brigades and eight national police brigades, amounting to approximately 44,000 Iraqi security personnel. [lxiii] It is very difficult to be precise about the total coalition security force presence in Baghdad in 2007 because the precise numbers for the Sunni militias are undocumented, but the total regular and irregular Iraqi force was perhaps 80,000–90,000. Only at this point did the coalition reach force ratios that approached the massive security presence in cities that France and Britain deployed in Algeria, Kenya, Malaya, Cyprus, and Northern Ireland.

The United States was much more dependent on local regular and irregular Iraqi personnel than French and British colonial powers were in the postwar period. Even in 2007, after the surge, the US was operating with a force ratio of one American soldier or marine for every 181 Iraqi citizens. In the battle of Sadr City, the ratio was even higher, although the coalition did not try to clear and hold the neighborhood. Then, 3,000 US soldiers—with, of course, many Iraqi allies—pacified an urban area of about 1.2 million: a ratio of about one American soldier to every 400 Shi’a citizens in that district. [lxiv]

Syria

Western states have not been alone in reducing their armed forces in the past 30 years. It has been a general trend. Western states have not been alone in experiencing unfavorable force ratios in urban areas. The wars in Syria, Yemen, and Libya have highlighted the problem of force ratios even more clearly. Each of these wars is politically complex; numbers and force ratios can not explain solely their origins and dynamics. However, the force ratios in these insurgencies are striking. The Syrian civil war offers a very significant, and perhaps the most pertinent, example of the issue. The origins of this war bear out the demographic arguments that most of the literature affirms. The conflict arose because of rapid urbanization and the grievances it generated. But while demographics may explain why the Syrian civil war started in towns and cities, demographics alone do not explain its subsequent development. The urban uprisings began in 2011. Yet, in 2021, after ten years of war, the conflict remained urbanized. All the major battles occurred in the towns and cities of Syria. The war started as an urban struggle and remained an urban struggle, with belligerents fighting for control of particular city sectors. It is necessary to explain this stasis.

Here, force ratios are instructive. Despite defeats by the Israelis in 1967, 1973, and 1982, the Syrian Army remained a large and well-equipped force. In 2013, in the early years of the civil war, it comprised 178,000 soldiers.[lxv] However, most of this force comprised conscripts drawn from the Sunni population, most of which were rebelling. The regime could trust only 65,000–80,000 of its soldiers, and only its elite troops—the Republican Guard, the 4th Armoured Division, and the special forces—were reliable. [lxvi] In addition, national defense forces also supported the Syrian regime, numbering some 50,000–60,000, and the sectarian militia, Shabbihah. Eventually, Hezbollah and Iran provided noteworthy additional forces. The Assad regime fielded about 150,000 combatants during the war. This was not a small force.

However, the uprising eventually involved every major Syrian city and most towns, mobilizing a large part of the Sunni population. Regime forces attempted to suppress armed resistance in Aleppo, Damascus, Homs, Raqqa, Idlib, Latakia, Hama, and Daraa, as well as some other smaller towns. Between 2011 and 2018, a force of 150,000 was divided between eight major cities and many smaller ones. In these locations, the regime employed considerable and indiscriminate force, but rarely enjoyed the combat supremacy necessary to defeat insurgents outright. Indeed, before the Russian intervention, in 2015, the regime gave up much of the country and focus on the ‘rump’ of Syria in the east from Suwaida in the south to the coast in the north. The regime could take back urban areas from the rebels only with the support of Russian air power, Iran’s Quds Force (under the command of Qasem Soleimani), and Hezbollah. Even then, the fight has been bitter and protracted. Like the Americans and British in Iraq, they confronted the Syrian armed forces with a severe shortage of troops. They could not drive their opponents out of the cities. As a result, long sieges have characterized this dreadful war. The battle of Aleppo was one of the most intense of the Syrian war, lasting from 2012 to 2016. Although eventually, the Syrian regime deployed about 30,000 regular and irregular troops into the city, at the start of the conflict force ratios were unfavorable to the regime. As the anti-regime forces seized large parts of the city in 2012, there were only 7,000 security force personnel in a city of 1.5 million: a force ratio of 1:214. They forced the Assad regime to cede large parts of the city.

Table 1 summarizes the predicament faced by many states today. As their security forces have contracted and, therefore, force densities have declined in comparison with the twentieth century, insurgents have been able to operate more freely in cities.

Table 1

Force sizes and ratios in the twentieth and twenty-first centuries, 1950–2020

Period	Campaign	Year	Urban population	Security forces	Force ratio	Average force ratio
1945–1990	Nairobi	1954	100,000	20,000	50:1	49:1
Algiers	1957	900,000	14,000/30,0001	64:1
Belfast2	1972	300,000	6,200	48:1
Belfast3	1972	300,000	8,600	35:1
2000–2020	Basra	2006	1,200,000	5,300	226:1	149:1
Baghdad4	2006	5,000,000	49,700	100:1
Baghdad5	2007	5,000,000	90,000	56:1
Aleppo6	2013	1,500,0007	7,000	214:1

¹To calculate the force ratio; I use the lower figure, derived from Alastair Horne, A savage war of peace: Algeria 1954–1962 (New York: New York Review of Books, 2011). Other scholars claim the figure is much higher (about 30,000 soldiers, police, and gendarmerie) (Martin Thomas, personal email, 25 Aug. 2020). In the higher figure, the ratio is 30:1.

²Pre-Operation Motorman.

³Post-Operation Motorman.

⁴Pre-surge: the figures are taken from Operation Together Forward in the last six months of the year.

⁵Post-surge.

⁶Operation Northern Storm, the regime’s first major offensive against the rebels, October–December 2013.

⁷Aleppo’s population declined sharply from 3 million in 2010, after the start of the war.

Conclusion: urban counter-insurgency in the future

Urban insurgency is not new. Guerrillas have not come out of the mountains in the past two decades; they were already in the city, and always have been. However, cities have become a major—even the prime—theatre for insurgency in the twenty-first century. Insurgents now typically situate their home bases inside cities; they dominate neighborhoods which they transform into no-go areas. The huge increases in the size of cities can certainly explain the rise of the urban insurgent in part, so that their vastness quite overwhelms the security forces. Slums are ideal home bases for urban gangs; such areas are almost impenetrable for the security forces.

Yet there is also an additional, often ignored, factor that helps explain why the urban guerrilla has become such an important factor in contemporary conflict: the contraction of state forces. State forces are tiny compared to those of the twentieth century. Most states today simply do not have the personnel to deploy one member of the security force for every 50 citizens to suppress an insurgency. So while states could drive insurgents from cities in the 1950s, 1960s, and 1970s, as events in Nairobi, Algiers, Aden, and Belfast showed, their reduced forces today can no longer dominate urban areas. They cannot patrol all the streets and alleys; and in this gap in the security apparatus, urban insurgents have proliferated in recent decades. Within cities, no-go areas have emerged which are very difficult for the security forces to penetrate or clear. This has allowed insurgents to buy and use more sophisticated, heavier weaponry and to engage in more ambitious military operations. In consequence, we see the appearance of the chronic inner-urban standoff, which has been such a feature of twenty-first-century urban conflict. Insurgents and state forces have been locked in an interminable struggle for supremacy inside the city itself.

The question for state forces, which are likely to have to deal with urban conflicts either at home or abroad in the next decade, is: how are they to offset the problem of their declining force sizes? How might states prevail in the urban environment even without mass? Many scholars believe that new equipment and weaponry will be decisive. For instance, the proliferation of disruptive security technologies deeply disturbed urban geographer Stephen Graham. He believes that the security forces are actively seeking to submit cities to total control. Security forces are obsessed with ‘technophilic desire and fetishistic urges for mastery and control, adjusted to the new imperatives of urban counter-insurgency warfare’. [lxvii] Graham is not alone in his concerns. In his recent bestseller, Paul Sharre expresses his fear that military forces are developing autonomous weapons systems that will identify and engage targets independently of human control:

Militaries around the globe are racing to deploy robots at sea, on the ground, and in the air—over ninety countries have drones patrolling their skies. These robots are increasingly autonomous and many are armed. They operate under human control for now, but what happens when a Predator drone has as much autonomy as a Google car? [lxviii]

He fears that ‘in future wars, machines may make life and death engagement decisions’. [lxix] These claims are exaggerated.

In the past two decades, disruptive informational and remote technologies have become commonplace in urban conflict. In particular, drones have proliferated: US forces employed them extensively in the battles of Fallujah, Sadr City, and Mosul, [lxx] while the Russian state employed them in the Donbas and Syria to great effect. [lxxi] States are also experimenting with nano-drones which might be deployed in swarms to monitor streets, buildings, or even rooms. Partly autonomous weapons have also appeared. These weapons have taken—and will take—many forms, but some prototypes are already operational. For instance, both the US and British armies have successfully trialed an autonomous vehicle. Quantum navigation and sensing systems are likely to be increasingly important in the coming decades. Downsized security forces are likely to rely more and more heavily on remote and perhaps autonomous systems in the urban conflicts of the next two decades to mitigate the effects of the reductions in the number of human troops. There seems little prospect that even with introducing more remote and autonomous weapons, urban insurgencies will be suppressed easily. It seems rather more likely that emergent autonomous technology will only augment and enhance the capabilities of contracted state forces. Numbers are still likely to be crucial.

In the past two decades, states have also employed a second technique to offset declining troop numbers: proxies. Partnerships with local state, militia, or irregular forces have played an increasingly important role in urban conflict, providing additional mass. Weaker states fighting for their sovereignty, such as Iraq, Syria, and Ukraine, have been increasingly forced to mobilize irregular troops. However, even for major powers like the United States or Russia, the use of proxy forces has become the norm. In Iraq, for instance, the success of the surge was substantially the result not of increased troop numbers or tactics, but the Anbar uprising; Sunni militias, for example, the Sons of Iraq, changed sides and worked with the US and the Iraqi government. [lxxii] In the campaign against ISIS, the United States provided very few ground troops, relying rather on local Iraqi security forces and Iraqi and Kurdish militias. Similarly, in the Donbas, Russia has supported local militias in Donetsk and Luhansk. However, Iran has probably been the most successful and adept at using proxies. Indeed, since the 1980s, it has ‘perfected the use of human surrogates’. [lxxiii] The Iranian Republican Guard Corps and its Quds Force, originally conceived of as a multinational Islamicist force dedicated to the ‘liberation’ of Jerusalem, have played a key role in Iran’s proxy policy. For instance, the Quds Force established Hezbollah in 1982 and the Badr Corps in Iraq in 1983, and in 2005 consolidated Muqtada al-Sadr’s Jaish al-Mahi in Iraq. [lxxiv] Each of these militias have taken part in major urban battles over the past two decades. In Syria, too, the Assad regime has relied heavily on allies and proxy forces. In all the major battles to retake the cities of western Syria, proxy ground forces have played a vital role. In particular, Hezbollah and the Iranian Republican Guard Corps (especially the Quds Force) have made up a major part of the force in which Qasem Soleimani played a vital role as a commander. Irregular forces have become a critical element of urban battle today. However, although increasingly important, irregular proxy forces have mitigated rather than overcome the problem of mass. Even with the help of proxies, state forces have struggled to generate the densities that were typical of the twentieth century and ensured that urban areas were challenging environments for insurgents.

Like emergent technologies, proxy forces are likely to become even more important in future urban insurgencies. However, their appearance does not seem to alter the fundamental thesis of this article. The urban insurgency has become an increasingly common phenomenon in the past two decades and is likely to remain an important feature of future warfare, not only because of the demographics or the asymmetric advantages of cities but also because state forces themselves have declined so radically. Downsized state forces, probably augmented with disruptive technologies and supported by proxies, will contest urban areas with insurgents in the coming decades. Indeed, the contraction of military forces makes urban insurgency even more likely shortly.

---

[i] See e.g. Gregory Ashworth, War and the city (London: Routledge, 1991); Russell Glenn, Combat in hell (Santa Monica, CA: RAND Arroyo Centre, 1996), pp. 2, 5; Ralph Peters, ‘Our soldier, their cities’, Parameters, Spring 2000, 1996, p. 43; Michael Desch, ed., Soldiers in cities: military operations on urban terrain (Carlisle, PA: Strategic Studies Institute, 2001); Sean Edwards, Mars unmasked (New York: RAND, 2000); Louis DiMarco, Concrete hell: urban warfare from Stalingrad to Iraq (Oxford: Osprey, 2012); Stephen Graham, Cities under siege: the new military urbanism (London: Verso, 2010); Alice Hills, Future wars in cities: re-thinking a liberal dilemma (London: Frank Cass, 2004).

[ii] Michael Evans, City without joy: military operations in the 21st century, occasional series no. 2 (Canberra: Australian Defence College, 2007), p. 14; David Kilcullen, Out of the mountains: the coming of age of the urban guerrilla (London: Hurst, 2013), pp. 74, 76; Frank Hoffman, ‘Complex irregular warfare: the next revolution in military affairs’, Orbis: A Journal of World Affairs 50: 3, 2006, pp. 395–411.

[iii] Hans Delbrück, History of the art of war within a political frame (London: Greenwood, 1975), p. 33; Stephen Biddle, Jeffrey A. Friedman and Jacob N. Schapiro, ‘Testing the surge: why did violence decline in Iraq in 2007?’, International Security 37: 1, 2012, pp. 7–40; John Hagan, Joshua Kaiser, Anna Hanson, Jon R. Lindsay, Austin G. Long, Stephen Biddle, Jeffrey A. Friedman and Jacob N. Shapiro, ‘Correspondence: assessing the synergy thesis in Iraq’, International Security 37: 4, 2013, pp. 173–98; Jacqueline Hazelton, ‘The “hearts and minds” fallacy: violence, coercion and success in counterinsurgency warfare’, International Security 42: 1, 2017, pp. 80–113; Enzo Nussio and Juan Ugarriza, ‘Why rebels stop fighting: organizational decline and desertion in Columbia’s insurgency’, International Security 45: 4, 2021, pp. 167–203.

[iv] For a longer discussion of the relevance of force size to urban warfare, see Anthony King, Urban warfare in the twenty-first century (Cambridge: Polity, 2021); Anthony King, ‘Will inter-state war take place in cities?’, Journal of Strategic Studies, publ. online Nov. 2021, https://www.tandfonline.com/doi/abs/10.1080/01402390.2021.1991797. (Unless otherwise noted at point of citation, all URLs cited in this article were accessible on 14 Jan. 2022.)

[v] See Christopher Phillips, The battle for Syria: international rivalry in the new Middle East (New Haven, CT: Yale University Press, 2018); Nicholas van Dam, Destroying a nation: the civil war in Syria (London: I. B. Tauris, 2017), pp. 106–8; Mouna Al-Sabouni, The battle for Homs: memoir of a Syrian architect (London: Thames & Hudson, 2016).

[vi] David Hodges, The philosophy of the urban guerrilla: the revolutionary writings of Abraham Guillén (New York: William Morrow, 1973), p. 234.

[vii] Antonio Sampaio, Illicit order: the military logic of organized crime and urban security in Rio de Janiero (London: International Institute for Strategic Studies, 2019), p. 9; Erica Robb Larkins, The spectacular favela (Oakland: University of California Press, 2015), pp. 32–3; Ioan Grillo, Gangster warlords: drug dollars, killing fields, and the new politics of Latin America (London: Bloomsbury, 2017), pp. 72–6; Desmond Enrique Arias, Drugs and democracy in Rio de Janiero (Chapel Hill: University of North Carolina Press, 2006); Graham Denyer Willis, The killing consensus: police, organised crime and the regulation of life and death in urban Brazil (Oakland: University of California Press, 2015).

[viii] Gian Gentile, David E. Johnson, Lisa Saum-Manning, Raphael S. Cohen, Shara Williams, Carrie Lee, Michael Shurkin, Brenna Allen, Sarah Solian and James L. Doty III, Reimagining the character of urban operations for the US Army (Santa Monica, CA: RAND Arroyo Centre, 2017); David Johnson, Wade M. Markel and Brian Shannon, The 2008 battle for Sadr City: reimagining urban combat (Santa Monica, CA: RAND, 2013).

[ix] David Johnson, Hard fighting: Israel in Lebanon and Gaza (Santa Monica, CA: RAND Arroyo Centre, 2011), p. 42

[x] Stephen Biddle, Nonstate warfare: the military methods of guerrillas, warlords, and militias (Princeton: Princeton University Press, 2021), p. 72.

[xi] Biddle, Nonstate warfare, p. 298.

[xii] Catherine Kelleher, ‘Mass armies in the 1970s: the debate in western Europe’, Armed Forces and Society 5: 1, 1978, pp. 3–30; Karl Haltiner, ‘The definite end of the mass army in western Europe?’, Armed Forces and Society 25: 1, 1998, pp. 7–36; Michel Martin, ‘Conscription and the decline of the mass army in France, 1960–75’, Armed Forces and Society 3: 3, 1977, pp. 355–406.

[xiii] Hazelton, ‘The “hearts and minds” fallacy’; David French, The British way in counter-insurgency 1945–1967 (Oxford: Oxford University Press, 2011); Douglas Porch, Counterinsurgency: exposing the myths of the new way of war (Cambridge: Cambridge University Press, 2011).

[xiv] Matthew Hughes, Britain’s pacification of Palestine: the British Army, the colonial state and the Arab Revolt, 1936–39 (Cambridge: Cambridge University Press, 2019).

[xv] Martin Thomas, Fight or flight: Britain, France and their roads from empire (Oxford: Oxford University Press, 2014), pp. 27–32.

[xvi] Benjamin Runkle, ‘Jaffa, 1948’, in John Antal and Bradley Gericke, eds, City fights: selected histories of urban combat from World War II to Vietnam (New York: Ballantine, 2003), pp. 289–313.

[xvii] Karl Hack, ‘The Malayan Emergency as counter-insurgency paradigm’, Journal of Strategic Studies 32: 3, 2009, pp. 383–414; Hazelton, ‘The “hearts and minds” fallacy’, pp. 99–100.

[xviii] Lieutenant-General Erskine, ‘The Mau Mau rebellion’, lecture to staff college, Camberley, 1955, p. 1; Huw Bennett, Fighting the Mau Mau: the British Army and counter-insurgency in the Kenya Emergency (Cambridge: Cambridge University Press, 2013), p. 112; Dan Branch, Defeating the Mau Mau (Cambridge: Cambridge University Press, 2009); Bruce Berman, Control and crisis in colonial Kenya (Oxford: James Currey, 1996).

[xix] David Anderson, Histories of the hanged: Britain’s dirty war in Kenya and the end of empire (London: Weidenfeld & Nicolson, 2005), p. 201; Anvil Area operation instruction no. 2, 8 April 1954, WO 276/188 (London: The National Archives (TNA)).

[xx] Anthony Burton, Urban terrorism: theory, practice and response (London: Leo Cooper, 1975), p. 169.

[xxi] Robert Taber, War of the flea: a study of guerrilla warfare theory and practice (Washington DC: Potomac, 2002), p. 127.

[xxii] Burton, Urban terrorism, pp. 175–6; Michael Dewar, War in the streets (London: BCA, 1992), p. 62.

[xxiii] Alastair Horne, A savage war of peace: Algeria 1954–1962 (New York: New York Review of Books, 2011)

[xxiv] Horne, A savage war of peace, p. 187; Jacques Massu, La vraie bataille d’Alger (Evreux: Librairie Plon, 1971), p. 15; Michel Bigeard, Ma guerre d’Algerie (Baume-les-Dames: Éditions du Rocher, 2010); Paul Aussaresses, The battle of the Casbah: counter-terrorism and torture 1955–57 (New York: Enigma, 2010); Roger Trinquier, Modern warfare: a French view of counter-insurgency (Fort Leavenworth, KA: US Army Command and General Staff College Press, 1985), pp. 16–17.

[xxv] Martin Thomas and J. F. V. Keiger, ‘France and the Algerian War: strategy, operations and diplomacy’, in Martin Thomas and J. F. V. Keiger, eds, France and the Algerian War 1954–62: strategy, operations and diplomacy (London: Frank Cass, 2002), pp. 6–7; Paul Alain Leger, ‘Personal account of chef de bataillon’, in Martin S. Alexander, Martin Evans and J. F. V. Keiger, eds, The Algerian War and the French Army, 1954–62 (Basingstoke: Macmillan, 2002), p. 241.

[xxvi] Alec Wahlman, Storming the city (Denton, TX: University of North Texas Press, 2015), p. 186.

[xxvii] Mark Bowden, Hué 1968 (London: Grove UK, 2017); James H. Willbanks, ‘The battle for Hue, 1968’, in William Robertson, ed., Block by block (Fort Leavenworth, KS: US Army Command and General Staff College Press, 2003), pp. 60–90.

[xxviii] Ernesto Che Guevara, Guerrilla warfare (Melbourne: Ocean Press, 2018, 2007; first publ. 1961), p. 15.

[xxix] Quoted in Régis Debray, The revolution in the revolution? (Harmondsworth: Penguin, 1972), p. 66

[xxx] Carlos Marighella, The minimanual of the urban guerrilla (Spade, 1969), p. 20.

[xxxi] Marighella, The minimanual, pp. 3–4.

[xxxii] Marighella, The minimanual, p. 17.

[xxxiii] Marighella, The minimanual, p. 40.

[xxxiv] Quoted in Hodges, The philosophy of the urban guerrilla, p. 234.

[xxxv] Quoted in Hodges, The philosophy of the urban guerrilla, p. 250.

[xxxvi] Quoted in Hodges, The philosophy of the urban guerrilla, p. 250.

[xxxvii] Frank Jellinek, The Paris Commune of 1871 (London: Gollancz, 1937); Robert Tombs, The Paris Commune of 1871 (Harlow: Pearson, 1999); Alastair Horne, The fall of Paris (London: Macmillan, 1965).

[xxxviii] International Institute for Strategic Studies, The military balance 1970 (London, 1970), p. 22.

[xxxix] See e.g. Tim Pat Coogan, The IRA (London: Fontana/Collins, 1987); Patrick Bishop and Eammon Mallie, The Provisional IRA (London: Corgi, 1988); Richard English, Armed struggle: the history of the IRA (London: Macmillan, 2003); Frederick Boal, ed., Ethnicity housing: accommodating the differences (Abingdon: Routledge, 2018); Matt Dillon, The dirty war (London: Arrow, 1991); Michael Dewar, The British Army in Northern Ireland (London: Arms and Armour, 1997). For a British Army perspective, see Chief of the General Staff, Operation Banner: an analysis of military operations in Northern Ireland (London: Ministry of Defence, 2006).

[xl] Sunday Times Insight Team, Ulster (Harmondsworth: Penguin, 1972), pp. 126–43; English, Armed struggle, pp. 101–2.

[xli] Ciaran de Baroid, Ballymurphy and the Irish War (London: Pluto, 2000), e.g. p. 42; Rod Thornton, ‘Getting it wrong: the crucial mistakes made in the early stages of the British Army’s deployment to Northern Ireland’, Journal of Strategic Studies 30: 1, 2007, pp. 73–107; Anthony Burton, Politics of legitimacy (London: Routledge and Kegan Paul, 1978.), p. 113.

[xlii] Dewar, The British Army in Northern Ireland, p. 96.

[xliii] Tony Geraghty, The Irish war: a military history of a domestic conflict (London: HarperCollins, 2000), pp. 33–40; Nick van der Bijl, Operation Banner: the British Army in Northern Ireland 1969–2007 (Barnsley: Pen & Sword, 2017), p. 35; David Barzilay, The British Army in Ulster (Belfast: Century, 1973), pp. 11–16.

[xliv] Many of the archives for this period have not been released. Others have been reclassified in the light of retrospective legal cases. However, for examples of British Army actions, see WO 305/4740, July 1972; WO 305/5703 30 June 1972 (TNA). For the Operation Motorman/Carcan Operations order, see DEFE 70/1477, 29 July 1972, TNA.

[xlv] Andrew Sanders, ‘Operation Motorman (1972) and the search for a coherent British counter-insurgency strategy in Northern Ireland’, Small Wars and Insurgencies 24: 3, 2013, pp. 465–92.

[xlvi] Van der Bijl, Operation Banner, pp. 66–7; Huw Bennett, ‘From direct rule to Motorman’, Studies in Conflict and Terrorism 33: 6, 2010, p. 522; Barzilay, The British Army in Ulster, p. 47; Desmond Hamill, Pig in the middle: the army in Northern Ireland 1969–1985 (London: Methuen, 1986).

[xlvii] See Martin Melaugh, Brendan Lynn and F. McKenna, Background information on Northern Ireland society: security and defence, CAIN Web Service, n.d., https://cain.ulster.ac.uk/ni/security.htm#01.

[xlviii] Frederick Boal, ‘Integration and division: sharing and segregating Belfast’, Planning Practice and Research 11: 2, 1996, pp. 151–8; Frederick Boal, ‘Territoriality on the Shankill–Falls divide, Belfast’, Irish Geography 41: 3, 2008, pp. 349–66; Frederick Boal, P. Doherty and D. Pringle, Social problems in the Belfast urban area: an exploratory analysis, Occasional paper no. 12 (Belfast: Queen Mary College, 1978).

[xlix] Robert Thompson, Defeating communist insurgency: experiences from Malaya and Vietnam (London: Chatto & Windus, 1974), pp. 47–8.

[l] David Galula, Counterinsurgency warfare (Westport, CT: Praeger, 2006), p. 21.

[li] Trinquier, Modern warfare, p. 6.

[lii] Force ratios are even higher in Belfast when the RUC is factored in; it is excluded because at the height of the battle of Belfast it assumed a supporting role.

[liii] See e.g. Haltiner, ‘The definite end of the mass army in western Europe?’.

[liv] Headquarters, Department of the Army, Counter-insurgency FM 3-24 (Washington DC, 2006), p. 1.2.

[lv] Lieutenant-General Sir Richard Shirreff, evidence to Chilcot Inquiry, 11 Jan. 2010, https://webarchive.nationalarchives.gov.uk/20110119123313/  http://www.iraqinquiry.org.uk/transcripts/oralevidence-bydate/100111.aspx, p. 4.

[lvi] Chief of the General Staff, Operations in Iraq, 29 Nov. 2010, p. 50.

[lvii] Warren Chin, ‘Why did it all go wrong? Reassessing British counterinsurgency in Iraq’, Strategic Studies Quarterly, vol. 128, 2008, p. 128.

[lviii] Peter Mansoor, ‘The British Army and the lessons of the Iraq War’, British Army Review, no. 147, 2009, p. 14; Daniel Marston, ‘Operation TELIC VIII to XI: difficulties of twenty-first-century command’, Journal of Strategic Studies 44: 1, 2021, pp. 63–90.

[lix] David Johnson, Agnes Gereben Schaefer, Brenna Allen, Raphael S. Cohen, Gian Gentile, James Hoobler, Michael Schwille, Jerry M. Sollinger and Sean M. Zeigler, The US Army and the battle for Baghdad: lessons learned and still to be learned (Santa Monica, CA: RAND, 2019), p. xviii.

[lx] Joel Rayburn and Frank Sobchak, eds, The US Army in the Iraq War, vol. 1: Invasion, insurgency, civil war (Carlisle, PA: US Army War College Press, 2019), pp. 539–41.

[lxi] Rayburn and Sobchak, eds, The US Army in the Iraq War, p. 570.

[lxii] Johnson et al., The US Army and the battle for Baghdad, p. 160.

[lxiii] Rayburn and Sobchak, eds, The US Army in the Iraq War, p. 48.

[lxiv] John Spencer, Stealing the enemy’s urban advantage: the battle of Sadr City, Modern War Institute, 31 Jan. 2019, https://mwi.usma.edu/stealing-enemys-urban-advantage-battle-sadr-city/.

[lxv] Phillips, The battle for Syria, p. 120.

[lxvi] Nicholas van Dam, Destroying a nation: the civil war in Syria (London: I. B. Tauris, 2017), pp. 106–8; Phillips, The battle for Syria, p. 161.

[lxvii] Graham, Cities under siege, p. 162.

[lxviii] Paul Sharre, Army of none: autonomous weapons and the future of war (New York: Norton, 2019), p. 4.

[lxix] Sharre, Army of none, p. 4.

[lxx] Gentile et al., Reimagining the character of urban operations for the US Army; Johnson et al., The 2008 battle for Sadr City; Mosul Study Group, What the battle of Mosul teaches the force, Sept. 2017, https://www.armyupress.army.mil/Portals/7/Primer-on-Urban-Operation/Documents/Mosul-Public-Release1.pdf.

[lxxi] Tim Ripley, Operation Aleppo: Russia’s war in Syria (Lancaster: Telic-Herrick, 2018), pp. 50, 52, 110, 190.

[lxxii] Biddle et al., ‘Testing the surge’; Stephen Biddle, Stabilizing Iraq from the bottom up, statement before the Committee on Foreign Relations, United States Senate, Second Session, 110th Congress, 2 April 2008, https://backend-live.cfr.org/sites/default/files/pdf/2008/04/Biddle%204-2-08%20Testimony.pdf.

[lxxiii] Andreas Krieg and Jean-Marc Rickli, Surrogate warfare (Washington DC: Georgetown University Press, 2019), p. 164; Eli Berman and David Lake, eds, Proxy warfare (Ithaca, NY: Cornell University Press, 2019); Tyrone Groh, Proxy war: the least bad option (Stanford, CA: Stanford University Press, 2019).

[lxxiv] Krieg and Rickli, Surrogate warfare, p. 176.

.