The Islamic State (IS), al-Qaeda-linked groups, Boko Haram and other extremist movements are protagonists in today’s deadliest crises, complicating efforts to end them. They have exploited wars, state collapse and geopolitical upheaval in the Middle East, gained new footholds in Africa and pose an evolving threat elsewhere. Reversing their gains requires avoiding the mistakes that enabled their rise. This means distinguishing between groups with different goals; using force more judiciously; ousting militants only with a viable plan for what comes next; and looking to open lines of communication, even with hardliners. Vital, too, is to de-escalate the crises they feed off and prevent others erupting, by nudging leaders toward dialogue, inclusion and reform and reacting sensibly to terrorist attacks. Most important is that action against “violent extremism” not distract from or deepen graver threats, notably escalating major- and regional-power rivalries.

The reach of “jihadists” (a term Crisis Group uses reluctantly but that groups this report covers self-identify with; a fuller explanation for its use is on page 2) has expanded dramatically over the past few years. Some movements are now powerful insurgent forces, controlling territory, supplanting the state and ruling with a calibrated mix of coercion and co-option. Little suggests they can be defeated by military means alone. Yet, they espouse, to varying degrees, goals incompatible with the nation-state system, rejected by most people in areas affected and hard to accommodate in negotiated settlements. Most appear resilient, able to adapt to shifting dynamics. The geography of crisis today means similar groups will blight many of tomorrow’s wars.

IS has reshaped the jihadist landscape: its strategy bloodier than that of al-Qaeda, from which it split in 2013; its declared caliphate across much of Iraq and Syria and grip on a Libyan coastal strip; thousands of foreigners and dozens of movements enlisted; its attacks in the Muslim world and the West. Fighting on multiple fronts – against Iran’s allies, Sunni Arab regimes and the West – it has woven together sectarian, revolutionary and anti-imperialist threads of jihadist thought. Its leadership is mostly Iraqi but the movement is protean: millenarian and local insurgent; to some a source of protection, to others of social mobility and yet others of purpose; with strands aiming to consolidate the caliphate, take Baghdad or even Mecca, or lure the West into an apocalyptic battle. Primarily, though, its rise reflects recent Iraqi and Syrian history: Sunni exclusion and anomie after the disastrous U.S invasion; harsh treatment under Prime Minister Nouri al-Maliki; and the brutality of President Bashar al-Assad’s regime and its allies. Any response must factor in IS’s many faces. But mostly it needs to address Sunni suffering in the Levant and the dangerous sense of victimisation that has helped spawn across the Sunni Arab world.

In part obscured by IS’s rise, al-Qaeda has evolved. Its affiliates in the Maghreb, Somalia, Syria and Yemen remain potent, some stronger than ever. Some have grafted themselves onto local insurrections, displaying a degree of pragmatism, caution about killing Muslims and sensitivity to local norms. Around the Lake Chad Basin, Boko Haram, the latest in a string of revivalist movements rooted in the marginalised political economy and structural violence of northern Nigeria, has morphed from isolated sect to regional menace, though formally joining IS has changed little about it. Movements of different stripes – the largely nationalist Afghan Taliban, resurgent as foreign troops draw down from Afghanistan, and Pakistani groups including sectarian movements, tribal militants fighting the central state and Kashmir- or Afghanistan-focused elements aligned to its military establishment – comprise an evolving South Asian jihadist scene.

The roots of this expansion defy generic description. Patterns of radicalisation vary from country to country, village to village, individual to individual. Autocrats, political exclusion, flawed Western interventions, failing governance, closing avenues for peaceful political expression, the distrust of the state in neglected peripheries, traditional elites’ declining authority and the lack of opportunity for growing youth populations have all played their part. So, too, has the dwindling appeal of other ideologies, particularly the Muslim Brotherhood’s peaceful political Islam – jihadists’ main ideological competitor – diminished by President Muhammed Morsi’s ouster and the subsequent crackdown in Egypt. Proselytising of intolerant strands of Islam has, in places, helped prepare the ground. The sectarian currents coursing through much of the Muslim world both are aggravated by IS and give it succour.

But if roots are complex, the catalyst is clear enough. The descent of most of the 2011 Arab revolutions into chaos has opened enormous opportunity for extremists. Movements have gathered force as crises have festered and evolved, as money, weapons and fighters flow in, as violence escalates. Mounting enmity between states means regional powers worry less about extremists than about traditional rivals, leverage the fight against IS against other enemies or quietly indulge jihadists as proxies. Especially in the Middle East, jihadists’ expansion is more a product of instability than its primary driver; is due more to radicalisation during crises than beforehand; and owes more to fighting between their enemies than to their own strengths. Rarely can such a movement gather force or seize territory outside a war zone or collapsed state.

Geopolitics hinders a coherent response. The starting point should be to dial back the Saudi-Iranian rivalry that drives Sunni and Shia extremism, deepens crises across the region and is among the gravest threats to international peace and security today. Easing other tensions – between Turkey and Kurdish militants, for example, Turkey and Russia, conservative Arab regimes and the Muslim Brotherhood, Pakistan and India, even Russia and the West – is also essential. In Libya, Syria and Yemen, tackling jihadists requires forging new orders attractive enough to deplete their ranks and unite other forces. Of course, none of this is easy. But redoubling efforts to narrow other fault lines would be wiser than papering them over in an illusion of consensus against “violent extremism”.

Vital, too, is to learn from mistakes since the 9/11 (2001) attacks. Each movement, notwithstanding the links between and transnational ties of some, is distinct and locally rooted; each requires a response tailored to context. They can, however, pose similar dilemmas and provoke similar blunders. Major and regional powers and governments in areas affected should:

• Disaggregate not conflate: Making enemies of non-violent Islamists, particularly the Muslim Brotherhood, prepared to accept political and religious pluralism and engage in politics is self-defeating. Also important is to distinguish movements seeking a place within the international order from those wanting to upend it. Even IS, its local branches and al-Qaeda affiliates, despite belonging to the latter category, are not monolithic. They have dedicated cores with transnational goals, but rank-and-files with diverse, mostly local motives whose loyalty can shift, and perhaps be shifted, with changing conditions. Governments should disaggregate even radical movements with an eye to ending violence, not lump others in with them looking for a fight.
   
• Contain if no better option exists: Foreign powers should always have a viable plan for what comes next if they undertake to oust militants; the same applies to governments in their hinterlands. Today’s strategy in Iraq – razing towns to defeat IS in the hope Sunni leaders in Baghdad can regain lost legitimacy through reconstruction – is unlikely either to meet Sunnis’ grievances or create conditions in which they can forge a new political identity. In Libya a heavy bombardment or deployment of Western troops against IS without a wider political settlement would be a mistake, likely to deepen the chaos. In both cases, slowing military operations also carries grave risks but, without a workable alternative, is the safer option – for those contemplating going in and those in areas affected alike.
   
• Use force more judiciously: Although force usually must be part of the response, governments have been too quick to go to war. Movements with roots in communities, tapping genuine grievances and sometimes with foreign backing are hard to extirpate, however unappealing their ideology. Wars in Somalia and Afghanistan show the shortfalls of defining enemies as terrorists or violent extremists and of combining efforts to build centralised state institutions with military action against them absent a wider political strategy that includes reconciliation. Nor can Russia’s scorched-earth approach in Chechnya – even leaving aside the human cost – be replicated in areas affected today, given porous borders, collapsed states and proxy warfare.
   
• Respect rules: Too often military action against extremists helps them recruit or leaves communities caught between their harsh rule and indiscriminate operations against them. Jihadists’ ability to offer protection against predation by regimes, other militias or foreign powers is among their greatest assets, usually more central to their success than ideology. While often guilty of atrocities, they fight in conflicts in which all sides violate international humanitarian law. Recovering the rulebook must be a priority.
   
• Curb targeted killings: Drone strikes can, in places, hinder groups’ operations and ability to hit Western interests and their leaders’ movements. But they feed resentment against local governments and the West. Movements weather the deaths of leaders, and the replacements that emerge are often harder-line. Foreseeing the impact of killings is hard in a reasonably stable order; doing so amid urban warfare and jihadist infighting – with al-Qaeda and others confronting IS – is impossible. Even leaving aside questions of secrecy, legality and accountability, targeted killings will not end the wars jihadists fight in or decisively weaken most movements.
   
• Open lines of communication: Notwithstanding the difficulties, governments should be more willing to talk, even with radicals. Opportunities to engage in ways that might have de-escalated violence – with some Taliban and al-Shabaab leaders, Boko Haram and Ansar al-Sharia in Libya, for example – have been lost. The decision whether a group is irreconcilable rests with its leaders not governments. Although policy-makers can entertain no illusions about the nature of the IS and al-Qaeda top commands, opportunities to open unofficial, discreet lines of communication, through community leaders, non-state mediators or others, are usually worth pursuing, particularly on issues of humanitarian concern, where there may be shared interest.
   
• Narrow the “countering violent extremism” (CVE) agenda: As a corrective to post-9/11 securitised policies, the CVE agenda, pioneered mostly by development actors, is valuable; so, too, are recognising the underlying conditions that can, in places, enable extremists’ recruitment and shifting funds from military spending to development aid. But re-hatting as CVE activities to address “root causes”, particularly those related to states’ basic obligations to citizens – like education, employment or services to marginalised communities – may prove short-sighted. Casting “violent extremism”, a term often ill-defined and open to misuse, as a main threat to stability risks downplaying other sources of fragility, delegitimising political grievances and stigmatising communities as potential extremists. Governments and donors must think carefully what to label CVE, further research paths of radicalisation and consult widely across the spectrum of those most affected.
   
• Invest in conflict prevention: IS’s and al-Qaeda’s recent expansion injects new urgency into prevention, both during crises, to halt their radicalisation, and upstream. Any further breakdown in the belt running from West Africa to South Asia is likely to attract an extremist element – whether these movements provoke crises themselves or, more likely, profit from their escalation. Although generic prescriptions are of limited value, nudging leaders toward more inclusive and representative politics, addressing communities’ grievances and measured responses to terrorist attacks usually make sense. Overall, in other words, preventing crises will do more to contain violent extremists than countering violent extremism will do to prevent crises.

The past quarter-century has seen waves of jihadist violence: a first in the early 1990s, when volunteers from the anti-Soviet jihad in Afghanistan joined insurrections elsewhere; a second pioneered by al-Qaeda culminating in the 9/11 attacks; and a third sparked by the U.S. invasion of Iraq. Today’s fourth wave is the most perilous yet. Partly this is thanks to IS’s territorial control and ideological innovation – its tapping of both local Sunni and wider anti-establishment discontent. Mostly, though, it is dangerous because of the currents propelling it, particularly the Middle East’s upheaval and fraying state-society relations there and elsewhere. World leaders’ concern is well-founded: IS’s attacks kill their citizens and threaten their societies’ cohesion. They face enormous pressure to act. But they must do so prudently. Missteps – whether careless military action abroad; crackdowns at home; subordinating aid to counter-radicalisation; casting the net too wide; or ignoring severer threats in a rush to fight “violent extremism” – risk aggravating those deeper currents and again playing into jihadists’ hands.

Brussels, 14 March 2016

In early 2011, revolutions in Egypt, Tunisia, Libya and Yemen heralded a new era of Arab politics. Protesters, often with women in the lead, took to the streets demanding greater dignity, opportunity and political pluralism. Among the main winners as authoritarians fell were Islamist parties prepared to participate peacefully in democratic politics.

Osama bin Laden’s ideology and tactics – a violent jihad targeting mainly Western powers – appeared increasingly out of step.^[fn] See "note on terminology" for an explanation of terminology, particularly the use of “jihad” and “jihadist”.Hide Footnote  Drone strikes in Pakistan’s tribal areas had by then decimated the al-Qaeda (AQ) core, and in May that year he was killed in Abbottabad. His most brutal franchise, best known as al-Qaeda in Iraq (AQI), looked beaten.^[fn]Al-Qaeda’s local branch in Iraq was Tanzim Qaidat al-Jihad fi Bilad al-Rafidayn (Organisation of Jihad’s Base in Mesopotamia), better known as AQI or al-Qaeda in Mesopotamia (see Section III.A).Hide Footnote  Except for al-Shabaab in Somalia, jihadists appeared peripheral to African crises.^[fn]

Note on terminology.^[fn]The root of the word “jihad” in Arabic refers to striving in the service of God. Many Muslims find its use in the context of political violence imprecise and offensive. It reduces a complex religious concept, which over centuries has taken many, often peaceful forms, to war-making. In the view of the vast majority of Muslims, today’s “jihadists” pervert Islam’s tenets. It is hard, however, to escape the term.
First, the groups this report addresses mostly self-identify as “jihadist”; Crisis Group normally lets actors speak for and label themselves. Secondly, while jihad has long been an element of virtually all schools of Islam, a nascent “jihadist” ideology has emerged that is more than a reflection of this history. Moving beyond the Islamist thought and practice that gave rise to modern jihad in recent decades, ideologues borrow from other traditions and at times show frustration with Salafi doctrinal rigidity that could constrain fighting tactics. Though big differences exist between “jihadist” groups, they share some ideological tenets: fighting to return society to a purer form of Islam; violence against rulers whose policies they deem in conflict with Islamic imperatives (as jihadists understand them); and belief in a duty to use violence when Muslim rulers abandon those imperatives. Our use of “jihadist” is not meant to add legitimacy to this interpretation or detract from efforts to promote alternative interpretations. 
We mostly avoid the term “violent extremist”, given that the groups covered in this paper represent only one form of “violent extremism” – namely Sunni extremism – and section IV.D explores some of the potentially dangerous policy implications of its use. Of course, lumping together movements with diverse goals and tactics under any single label, whether “jihadist” or “violent extremist”, is to a degree unhelpful. We disaggregate between and within even the hardest-line movements throughout this report and recommend policymakers do the same. We use “terrorist” only as an adjective to describe the attempt to use violence or intimidation, especially of civilians, to achieve political goals through the manipulation of fear. Though in principle both state and non-state actors can employ terrorist tactics, we use it here for actions of the latter.Hide Footnote

 This report uses the form al-Shabaab rather than Al-Shabaab (as the movement is commonly known in Africa and in Crisis Group publications) so as to maintain internal consistency of transliteration from Arabic.Hide Footnote

Today, the Middle East is at war, and the main winners so far are extremists. A wider belt, from West Africa to at least South Asia, appears vulnerable. The Islamic State (IS) claims a caliphate across large parts of Iraq and Syria, effacing the border between them and, in an amplification of the mostly Arab fighters who went to Afghanistan in the 1980s, has attracted tens of thousands of foreigners from the world over. Despite recent territorial losses, it has convinced dozens of movements elsewhere to sign up and coordinated or inspired attacks in the Muslim world and the West. An al-Qaeda affiliate, Jabhat al-Nusra, is among the most powerful Syrian opposition factions. Yemen’s escalating crisis has allowed another affiliate, al-Qaeda in the Arabian Peninsula (AQAP), to seize Mukalla, a strategic port on the Gulf of Aden, and surrounding areas.

An IS branch controls a 200-300km stretch of Libya’s Mediterranean coast and threatens the infrastructure for oil, the country’s main source of income. Other militants are ensconced elsewhere in its cities and towns. Jihadist groups, including al-Qaeda in the Islamic Maghreb (AQIM), though ousted from northern Malian towns in 2012, remain at large across the Sahel and claim responsibility for recent attacks in Bamako and Ouagadougou.^[fn]See also Section III.C.Hide Footnote  Boko Haram, a vicious insurgency indigenous to northern Nigeria, overran a swathe of the north east in 2013-2014 and still terrorises a large area around Lake Chad. Al-Shabaab poses an increasing threat beyond its Somali base, particularly to Kenya. In Afghanistan, the Taliban is resurgent, while other groups, including Pakistani, Central Asian and other foreign elements as well as Taliban splinters, join IS. Pakistan, despite efforts to rein in some extremists, still faces a multipronged threat from tribal militias, sectarian groups and its own proxies. Although Russia crushed a jihadist insurgency in the North Caucasus ahead of the Sochi Olympics, its operations displaced thousands of fighters to Iraq and Syria, while remnants in the Caucasus have joined IS.^[fn]Crisis Group Europe Report N°238, The North Caucasus Insurgency and Syria: Exported Jihad?, forthcoming 16 March 2016.Hide Footnote

Extremism in the Muslim world has ebbed and flowed over the past quarter century but has never looked as dangerous as today. IS and al-Qaeda-linked groups are among the most powerful protagonists in many of the world’s deadliest crises and may exploit divisions elsewhere, while their sophisticated recruitment, particularly that of IS, threatens countries hitherto unaffected.

Enormous differences exist between groups’ beliefs, strategies, tactics and targets, but all, according to their own statements, aim to return society to a purer form of Islam and believe that fighting a violent jihad to do so is a religious duty. Most to some degree define themselves as “jihadist”, however contested, varied and elusive the term’s meaning.^[fn]A notable exception here is now Ahrar al-Sham in Syria, see fn 87. See also note on terminology.Hide Footnote At some point, most have had ties, however loose, to al-Qaeda. Many policymakers erroneously lump them together.

This report examines today’s jihadist landscape. Why have these groups become so powerful? What do they want, and how are they pursuing it? How do they win support and control territory when their ideology has appeared, at least until recently, to have little natural constituency? How do they shape the conflicts they fight in and prospects for ending them? What threat do they pose elsewhere? How should the world respond? It draws from and extends findings from Crisis Group’s extensive body of work on the severest crises in which such movements are prominent, focusing in particular on the Middle East, given the pace of change there, but including West Africa, the Sahel, the Caucasus, North Africa, the Horn of Africa and Central and South Asia.^[fn]Crisis Group’s extensive work on violent extremism is available on our global issues page Jihad in Modern Conflict. This report mostly focuses on areas where IS- or al-Qaeda-linked groups have been able to seize territory or that appears a risk. It covers Europe – and many other places of origin of foreign fighters – only inasmuch as attacks there impact the calculations of its leaders. For similar reasons, it does not cover South East Asia: groups there are relatively small and, in the four areas of concern, southern Thailand, southern Philippines, Indonesia and the Rohingya in Myanmar, extremism per se has little attraction. Marginal groups have pledged allegiance to jihadists – Abu Sayyaf in the Philippines has released a video accepting the caliphate, as has Santoso, leader of the small Indonesian Mujahidin in central Sulawesi; Indonesia’s Jemaah Islamiyah has long-established al-Qaeda links – but none has gained traction. Despite outreach from IS and AQ, mainstream militant groups remain staunchly wedded to ethno-religious nationalism not global jihadism. Moreover, the states in which they operate are strong, with functioning institutions; repressive, but not on the scale that opens space for jihadists. Democratic and economic progress in the region over three decades allows for peaceful dissent, greater social mobility and a paradigm of growth that most people believe in. Jihadist groups exist and will continue to attack domestic and foreign targets, particularly in Indonesia, but their tactics and ideology are a hard sell in current regional conditions, and they are unlikely to reach the critical mass that would threaten society or the state. Even in the Southern Philippines, if peace talks collapse, most locals believe the danger is warlordism, not puritan Islam. For more on our reporting on South East Asia visit the regional page.Hide Footnote

The report does not examine the Muslim Brotherhood and its branches, including Hamas. Despite some shared roots, it has distanced itself over decades from the thinkers that inspire al-Qaeda and is perhaps jihadists’ main ideological competitor, though Cairo’s campaign against it has plunged it into disarray and left its future uncertain. IS and al-Qaeda attack many Brotherhood tenets and practices, including, on a political level, gradualism and participation in democratic politics. In terms of doctrine, the Brotherhood’s – and Hamas’s – relative flexibility and pragmatism sets them apart from the literalism of Salafis and the Taliban. Over the past few years, jihadists’ fortunes, particularly in the Arab world, have waxed as those of Muslim Brothers have waned.

Nor does it examine Shia militancy, though the Iranian-sponsored radicalisation of Shia governments and militias across parts of the Middle East and the violence Iraqi and Syrian Sunnis have suffered at their hands have been major drivers of Sunni extremism. Many Crisis Group reports already cover this terrain, as well as Pentecostal fundamentalism and the Lord’s Resistance Army (LRA) in Africa, Buddhist and Christian extremism in Asia, parts of the Jewish far right in Israel and other forms of religiously-framed violence.^[fn]See among many, for example, Crisis Group Middle East Reports N°s 38, Iran in Iraq: How Much Influence?, 21 March 2005; 89, Israel’s Religious Right and the Question of Settlements, 20 July 2009; 104, Radical Islam in Gaza, 29 March 2011; 153, Lebanon’s Hizbollah Turns Eastward to Syria, 27 May 2014; and 154, The Huthis: From Saada to Sanaa, 10 June 2014; Middle East Briefing N°45, Yemen at War, 27 March 2015; Asia Report N°251, The Dark Side of Transition: Violence Against Muslims in Myanmar,  1 October 2013; Africa Report N°229, Cameroon: The Threat of Religious Radicalism, 3 September 2015; and Middle East Report N°147, Leap of Faith: Israel’s National Religious and the Israeli-Palestinian Conflict, 21 November 2013. See also, Asia Briefing N°114, Indonesia: “Christianisation” and Intolerance, 24 November 2010; Asia Report N°134, Sri Lanka’s Muslims: Caught in the Crossfire, 29 May 2007; and Africa Report N°182, The Lord’s Resistance Army: End Game?, 17 November 2011.Hide Footnote

The following sections examine the origins, trends and geopolitics beneath the recent jihadist expansion (II); give a snapshot of the evolving landscape (III); and explore policy options (IV). The main focus is less how individuals radicalise than how extremist movements have become prominent in so many of today’s deadliest crises; and less what groups and their leaders say than what they do. The report sets the stage for development of a wider body of Crisis Group work, identifying areas for further research on the nature of groups, their interaction with crises, the threat and policy dilemmas they pose and ideas on how to respond.

IS’s and al-Qaeda’s expansion over the past few years is the fourth in a series of waves of jihadist violence affecting mostly the Muslim world since the Soviet-backed government in Afghanistan fell in 1989.^[fn]Though jihadist ideology’s roots stretch back much further, the fourth wave’s modern origins can be traced to: first, the anti-Soviet jihad in Afghanistan, when thousands of foreign volunteers, often known as “Arab Afghans”, travelled to fight Soviet forces; the myth that these foreigners forced the invaders’ retreat, when their role was minimal compared with that of the Afghan mujahidin, became part of al-Qaeda’s founding narrative; secondly, the revolutionary violence inspired by Sayyad Qutb and his contemporaries in Egypt against President Gamal Abdul Nasser’s government; and thirdly, the Iranian Revolution and subsequent Iranian backing for Shia activism across parts of the Muslim world, which provoked in response Saudi and other Gulf funding for Sunni radicals. Different movements today draw from these several strands – anti-imperialist, revolutionary and sectarian – of jihadist thinking. To a degree, IS embodies them all (see Section III.A). Jason Burke, The 9/11 Wars (London, 2011) treats the past few decades’ ebbs and flows of extremist violence; as does Daniel Byman, Al-Qaeda, the Islamic State and the Global Jihadist Movement (Oxford, 2015).Hide Footnote  The first, in the early 1990s, saw many of the foreign volunteers fighting in Afghanistan return to Algeria, the Caucasus, Egypt, Libya, Sudan and elsewhere. In some places, small cells, clustered around charismatic leaders with Afghanistan experience, launched campaigns, mostly terrorist attacks with civilian casualties, against regimes they declared un-Islamic. Elsewhere, Afghanistan veterans joined irredentist struggles, revolutions or civil wars, sometimes, particularly in Algeria and Russia (Chechnya), contributing to their radicalisation. This wave subsided by the mid-1990s, as wars ended or movements were crushed or ejected from those countries. Many members retreated to Afghanistan, then under Taliban control.

From there, al-Qaeda launched a second wave targeting mostly what it called the “far enemy”. Its aim was to suck Western powers into wars in which they would be defeated, like the Soviets in Afghanistan, so withdraw support for regimes in the region, precipitating their downfall. As local-language satellite media outlets reached across the Islamic world, Osama bin Laden pioneered spectacular attacks, mostly against Western interests, to gain attention and cement his position at the vanguard of the global jihadist movement. This wave peaked with the 9/11 attacks in the U.S., which were opposed by Taliban leaders and many of the “Afghan Arabs” fighting for the Taliban against the Northern Alliance or in training camps dotted across the country. They rightly feared that the U.S. reaction bin Laden aimed to provoke would destroy the Taliban’s emirate and deny them their safe havens.^[fn]Mustafa Hamid and Leah Farrall, The Arabs at War in Afghanistan (London, 2015). See also Al-Gama’a al-Islamiya, “The Strategy and Bombings of al-Qaeda: Errors and Perils”, al-Sharq al-Awsat, 2004, for wider resistance among radical Islamist movements to the attacks. Mullah Omar’s refusal to hand over bin Laden, despite Taliban leaders’ apprehension about 9/11 and Pakistan’s urging, led to the U.S.-led intervention in Afghanistan.Hide Footnote  U.S.-backed forces ousted the Taliban quickly. Many of the foreign fighters were killed or captured; others sheltered in the Pakistani tribal areas or scattered.

The 2003 U.S. invasion of Iraq fuelled a third wave, reinvigorating the jihadist movement as thousands of Muslims, many from the Gulf and North Africa, travelled to fight the Americans in the heart of the Arab world.^[fn]Thomas Hegghammer, “The Rise of Muslim Foreign Fighters”, International Security, winter 2010/2011.Hide Footnote  The Awakening, a U.S.-backed tribal revolt against al-Qaeda’s franchise in Iraq that was partly motivated by the movement’s brutality, stemmed that wave.^[fn]For more, see Crisis Group Middle East Report N°s 74, Iraq after the Surge I: The New Sunni Landscape, 30 April 2008; 75, Iraq after the Surge II: The Need for a New Political Strategy, 30 April 2008; 144, Make or Break: Iraq’s Sunnis and the State, 14 August 2013; and 150, Iraq: Falluja’s Faustian Bargain, 28 April 2014.Hide Footnote  The Arab Spring protests that spread across towns and cities in 2011 then appeared to break it.

The collapse or suppression of most of those revolutions, however, has spurred a fourth wave. More powerful than its predecessors, it has seen IS- and al-Qaeda-linked groups seize territory, gain new footholds in Africa and pose a growing menace across much of the Muslim world and to the West. Generalising about the deeper currents driving this fourth wave is risky, particularly mid-flow. Dynamics vary between regions: from the Middle East’s war zones; to Africans’ struggles to cope with the instability that spills south; to the long legacy and Pakistan’s frequent support of jihad in South Asia. Each movement is unique and, despite the transnational ties of some, mostly rooted in local conditions. Patterns of radicalisation vary from place to place. Like any global trend, jihadists’ expansion results from different things happening in different places, some connected directly, some indirectly and some not at all.

Its immediate causes, however, are clear enough and explain why this fourth wave is potentially the most destructive and hardest to reverse. First and foremost, there is the upheaval across much of the Arab world. Jihadists’ gains have long been entwined with conflict, from Afghanistan to Algeria, from Iraq to Syria. The dramatic recent uptick in war and state collapse has opened up enormous opportunity for them. Enmity between states, meanwhile, in the Middle East at a level dwarfing that of previous waves, means regional powers worry less about extremists than about their rivals, or even quietly indulge such groups as proxies.

The sectarianism and deep sense of Sunni victimisation that the Iraq and Syria wars and the perception of an ascendant Iran have helped spawn play into extremists’ hands. So, too, do failed governance, authoritarian backlash and the elimination of legitimate and politically viable alternatives, all of which reinforce jihadists’ denunciation of corrupt local regimes and contribute to anti-establishment sentiment across the region. Weak states with limited writ across their hinterlands or borders have proven vulnerable, particularly in Africa. Aggressive proselytising over decades of intolerant strands of Islam and the dwindling appeal of ideologies that might be used to frame resistance have helped prepared the ground.

The grievances that took Syrians to the streets in 2011 were much like those motivating other Arab revolts. Most protesters did not initially call for President Bashar al-Assad to stand down but demanded that his increasingly sclerotic and repressive government reform, open politics and improve economic management. Over eighteen months, peaceful protests morphed into what has become, at least in parts of the north, a jihadist-dominated insurgency for very different reasons.^[fn]See Crisis Group Middle East Reports N°s 163, New Approach in Southern Syria, 2 September 2015; 155, Rigged Cars and Barrel Bombs: Aleppo and the State of the Syrian War, 9 September 2014; 146, Anything But Politics: The State of Syria’s Political Opposition, 17 October 2013; and 143, Syria’s Metastasising Conflicts, 27 June 2013. Also Charles Lister, The Syrian Jihad: al-Qaeda, the Islamic State and the Evolution of an Insurgency (Hurst, 2015); Nicolas Hénin, Jihad Academy: The Rise of the Islamic State (Bloomsbury, 2015).Hide Footnote  The most important was the regime’s response: deliberate radicalisation of the crisis through cruel, publicised violence; divisive sectarian discourse, pitting the ruling Alawite and other minorities against the Sunni majority; escalating collective punishment that destroyed cities and helped displace millions; and its release of jailed radicals and targeting of more pragmatic opposition factions.^[fn]Alawites, comprising roughly 12 per cent of the Syrian population, historically have lived principally in the mountain chains in the north west, along the Mediterranean coast; today, there are many in Damascus and Homs as well. Accounts of their religious origins vary; they are most likely an offshoot of the Twelver branch of Shia Islam. See Henri Laoust, Les Schismes dans L’Islam (Paris, 1977), p. 147. When Hafez al-Assad became president in 1971, he sought the help of Imam Musa al-Sadr, a leading Shia cleric in Lebanon, to certify that Alawites were Muslim and Shia. Sadr issued a fatwa (religious ruling) to that effect. Patrick Seale, Asad: The Struggle for the Middle East (Berkeley, 1988), p. 173. Hide Footnote

At the same time, friction between Qatar and Turkey on one side, Saudi Arabia and the United Arab Emirates (UAE) on the other, meant that their support to the opposition was incoherent and often flowed, like that channelled by Gulf-based clerics, to extreme proxies. Foreign fighters, who tended to be more radical, for a time entered freely through Turkey.^[fn]Sally Judson and Kadir Udson, “Turkey’s ISIS Challenge”, SETA, September 2014. Western officials admit that shutting down the border completely would be impossible and that Turkey, at least since March 2014, has worked to stem the flow. Crisis Group interview, Western diplomat, Ankara, February 2015.Hide Footnote  The gap between U.S. and other Western powers’ rhetoric – that Assad must go – and the reality that they would not commit troops, conduct airstrikes or arm his opponents enough to make that happen undermined less radical groups, whose strategy had hinged on drawing Washington in. As jihadists, many with Iraq combat experience, entered, some, notably Abu Mohammad al-Jolani, leader of the local al-Qaeda branch, Jabhat al-Nusra, proved effective commanders. Tactics like suicide bombing gave them an edge. The regime’s immense violence stoked desire for revenge among many Sunnis and numbed communities to jihadist atrocities.

The paths by which jihadists have become potent in today’s conflicts vary place to place, but this pattern – their exploitation of chances created by war and state collapse, their rise facilitated by the violence and mistakes of others – is common to many.

IS’s roots in Iraq (explored in more detail in Section III.A) lie in a similar mix. The U.S. invasion and occupation policies set the stage for the sectarian civil war (2005-2008) that Abu Musab al-Zarqawi, leader of AQI, the progenitor to IS, helped provoke. Equally important was failure of Baghdad and Washington to capitalise on the Awakening. Denial to the minority Sunnis of a sufficient stake in the state, then violence by mostly Shia security forces against largely peaceful protests in Sunni-majority cities in 2012-2013 undermined non-jihadist Sunni leadership and resistance. This cleared the way for IS, which had regrouped, to eradicate rivals and seize the Iraqi Sunni heartlands in 2014, with many Sunnis seeking its protection or seeing in it an opportunity to upset the status quo.^[fn]See, for example, Crisis Group Reports, Iraq: Falluja’s Faustian Bargain, and Make or Break, both op. cit.Hide Footnote

In Yemen, al-Qaeda’s local branch, AQAP, focused mostly on terrorist attacks until 2011. It was dangerous to the West because of its bomb-making expertise but largely peripheral to Yemeni politics and isolated in the remote east.^[fn]Gregory D. Johnsen, The Last Refuge: Yemen, al-Qaeda and America’s War in Arabia (New York, 2014).Hide Footnote  Only when the state collapsed – first as army factions faced off in the capital during the 2011 revolution, then in 2015 as Huthi insurgents advanced, and the Saudi-led coalition escalated in response – could it seize population centres.^[fn]Crisis Group Middle East Reports N°s 114, Breaking Point? Yemen’s Southern Question, 20 October 2011; and 167, Yemen: Is Peace Possible?, 9 February 2016.Hide Footnote

In Libya, too, IS and other extremist groups profited from the collapse of authority: first in the initial chaos after Muammar Qadhafi’s 2011 ouster, then, in 2014, from the escalating standoff between Tobruk- and Tripoli-aligned forces and their respective regional backers.^[fn]Crisis Group Middle East/North Africa Report N°157, “Libya: Getting Geneva Right”, 26 February 2015.Hide Footnote  In Mali, local al-Qaeda leaders, veterans of the Afghan and Algerian wars, had sheltered with tribes in the desert for years before they allied with, then usurped a Tuareg nationalist insurrection sparked largely by the return of mercenaries and weapons from Libya.^[fn]Crisis Group Africa Briefing N°90, Mali: The Need for Determined and Coordinated International Action, 24 September 2012; Africa Report N°92, Islamist Terrorism in the Sahel: Fact or Fiction?, 31 March 2005.Hide Footnote  The Taliban and al-Shabaab emerged only after decades of chaos in Afghanistan and Somalia, in both cases partly in reaction to the predation of warlords and the dwindling legitimacy of other armed groups.^[fn]Crisis Group Africa Report N°147, Somalia: To Move Beyond the Failed State, 23 December 2008; and Asia Report N°221, Talking About Talks: Toward a Political Settlement in Afghanistan, 26 March 2012.Hide Footnote

Boko Haram in northern Nigeria, is something of an outlier, in that it did not emerge in an existing war zone. Rooted in the north’s structural violence and marginalised political economy, it began as an isolated sect, then a protest movement demanding less corrupt Islamic governance. Its resistance to the state hardened after quarrelling with a local governor, who, according to its then leader, Mohammed Yusuf, had broken promises made to it for help mobilising votes. Even then, though, it was the 2009 crackdown in Maiduguri, in which some 800 supporters died; Yusuf’s extrajudicial killing in police custody; an inept government response to the mounting menace; and the arrival of weapons and expertise from Libya and the Sahel that drove the movement’s mutation into the insurgency under Abubakar Shekau that plagues the Lake Chad Basin today.^[fn]Boko Haram is the latest in a string of revivalist movements in northern Nigeria, long the hub for a two-way exchange of ideas running between there and other parts of the Muslim world. Crisis Group Africa Reports N°s 168, Northern Nigeria: Background to Conflict, 20 December 2010; and 216, Curbing Violence in Nigeria (II): The Boko Haram Insurgency, 3 April 2014. Much about Boko Haram remains murky, including the movement’s coherence and even if Shekau is still alive.Hide Footnote

Overall, therefore, jihadists’ growing prominence over the past few years is more the product of instability than its primary driver. Movements have gathered force as crises deepen and violence escalates. In some cases, particularly Boko Haram’s, extremists have helped provoke the conflicts they fight in – though there, as elsewhere, the state’s violence has been instrumental to its growth. More often, jihadists have exploited existing conflicts, as they did in Algeria and Chechnya two decades ago, infiltrating, profiting and making them harder to resolve. Their dramatic expansion in recent years owes more to the bloody genesis of crises, in other words, than to radicalisation beforehand. They have usually been able to graduate from terrorist tactics to insurgency only in conditions of war; IS’s strategy, as shown below, and to a degree al-Qaeda’s, rest on provoking precisely those conditions.

Escalating geopolitical rivalries have been another windfall for extremists. The modern jihadist movement was partly born of competition between states: Cold War rivalries in Afghanistan, which motivated the USSR’s invasion; the U.S. and Gulf monarchies funnelling and Pakistan radicalising Muslims to fight Soviet forces in response; and the explosion of Gulf funding for radical Sunni movements, partly to counter Iran’s sponsorship of Shia activism after its 1979 revolution. Mounting competition, particularly between Middle Eastern states, now drives and complicates efforts to end the crises jihadists feed off. It also means many leaders worry more about regional rivals than extremists. In Yemen, for example, the actions of Saudi Arabia and the UAE show they view the Huthis and the risk they perceive of Iranian influence on the Arabian Peninsula as graver threats than al-Qaeda. For months, AQAP-controlled areas were among the few Saudi-led coalition bombs avoided, strengthening the group relative to others.^[fn]Western diplomats’ claims that the UAE is more serious about AQAP are not borne out by its actions on the ground, particularly as Saudi Arabia sets military priorities. See Crisis Group Report, Yemen: Is Peace Possible?, op. cit.Hide Footnote

Regional politics present an even greater obstacle in Syria. First, as described, state policies helped facilitate the opposition’s initial radicalisation and Jabhat al-Nusra’s expansion, paving the way for IS’s advance. Even now, few of the diverse forces arrayed against IS treat it as the main enemy. The Assad regime, Iran, allied militias and Russia mostly attack other rebels, including those on the front lines against IS, believing them a graver threat to regime survival. Gulf powers and Turkey prioritise Assad’s removal, and the Turks fear the ascendance of the Kurdish People’s Protection Units (YPG), affiliated with their domestic insurgent enemy, the Kurdistan Workers’ Party (PKK).^[fn]Crisis Group observations, interviews, Ankara and Washington DC, January 2016. The YPG receives U.S. support, to the chagrin of Turkey, whose officials argue some of this weaponry ends up with the PKK.Hide Footnote  IS is first priority in Syria only for Western powers and Jordan.

Worse still, a common thread in the history of many movements is the support they have enjoyed from states hoping to use them as proxies against rivals. Pakistan’s jihadist milieu defies easy description, but the roots of some movements trace back to wars in Afghanistan and Kashmir, where they served as its foreign policy instruments. Even as some of these proxies cooperate with tribal militants that attack the Pakistani state or are actively engaged in radicalising a new generation of extremists, military and many civilian leaders still indulge Lashkar-e-Tayyaba, one of the world’s largest jihadist groups, and back the Afghan Taliban. The A^[fn]On the eve of 9/11, Pakistani militants could be categorised by their focus. Harkat al-Mujahidin (HuM), Jaish-e Mohammed (JeM), an HuM splinter, and Lashkar-e Tayyaba (LeT) fought in Kashmir. The Sipah-e Sahaba Pakistani (SSP) and its splinter Lashkar-e Jhangvi (LeJ) perpetrated sectarian attacks in Pakistan. Both were built decades earlier, largely with Saudi money to counter the increasing stridency of Shia militants backed by post-revolution Iran but also drawing from local resentment against wealthier Shia in Jhang. Numerous groups in the tribal areas had fought in Afghanistan. The last fifteen years have seen these distinctions gradually become less relevant, as many militants rubbed shoulders with each other and with al-Qaeda while fighting alongside the Afghan Taliban and training in the Pakistani tribal areas. The principle dividing line now is between those groups that fight the Pakistani state and those that do not – though even that is blurred. Groups that are military-sponsored and do not attack the state often provide training and infrastructure to those that do. A second dividing line is between those that attack Shia and other religious minorities and those that are less overtly sectarian. Crisis Group Asia Reports, N°s 164, Pakistan: The Militant Jihadi Challenge, 13 March 2009; 178, Pakistan: Countering Militancy in FATA, 21 October 2009; and 242, Pakistan: Countering Militancy in PATA, 15 January 2013.Hide Footnote ssad government funnelled jihadists into Iraq through the mid-2000s in an attempt to divert their attention and keep the U.S. bogged down; the latter motive drove Iran’s sporadic facilitation of al-Qaeda fighters’ transit to Iraq at the same time.^[fn]Hassan Abu Hanieh and Mohammed A. Rumman, The “Islamic State” Organization: The Sunni Crisis and the Struggle of Global Jihadism (Jordan, 2015).Hide Footnote

State support, direct or indirect, for jihadists appears to be on the rise, particularly as rivalry escalates between Iran and Gulf monarchies angered at what they see as Tehran’s growing geopolitical clout after the nuclear deal. Some of the weapons and ammunition flowing from the Gulf and Turkey to components of the Jaish al-Fatah rebel coalition in Syria almost certainly reach Jabhat al-Nusra, one of its most powerful members.^[fn]Crisis Group interviews, Turkey, April-December 2015.Hide Footnote Amid Yemen’s chaos, weapons delivered to local allies of the Saudi-led coalition seep into the arsenal of al-Qaeda, with which some of Riyadh’s partners align tactically against the Huthis.^[fn]Crisis Group Report, Yemen: Is Peace Possible?, op. cit. Also, Crisis Group interviews, Western journalist, November 2015; Adeni journalist, October 2015; Arab diplomat, tribal sheikh from Shebwa, August 2015.Hide Footnote

As Pakistan’s experience shows, jihadists make dangerous proxies. Iran’s non-state allies – the Iraqi Shia militias, Hizbollah and the thousands of Afghans and other Shia it has mobilised to fight beside Assad’s forces – are unlikely to turn on the Islamic Republic, given its reasonably coherent revolutionary narrative, their dependence on its support and its capable defence forces. By contrast, a centrepiece of many Sunni extremists’ strategy is to topple local regimes, including those on the same side of the sectarian line. The Gulf monarchies’ anxiety about Iran is understandable; Turkey has legitimate concerns about Kurdish separatism. But subordinating the threat from IS and al-Qaeda-linked groups and their ideas to these worries – or worse still, indulging such groups in the hope their sights remain on Iran’s allies – is likely to prove a miscalculation.

If wars, state collapse and geopolitics, particularly across the Arab world, are proximate causes of the fourth wave, other trends contribute. They are too complex to treat comprehensively, particularly as the dynamics are so varied, but a few stand out.

First, sectarianism has reached unprecedented levels across parts of the Middle East. Aggravated by Saddam Hussein’s overthrow, wars in Syria and Iraq and escalating Saudi-Iranian rivalry, it is more intense than any time since religion was conjoined with modern political identity. As states fail, many, not just Sunnis, are turning to other kinds of social organisation – tribe, clan, religion, sect – for protection and representation. The ramifications are still uncertain, but clearly sectarian hatred plays into the hands of IS, which both drives and feeds off it. It also moulds a new generation of jihadists who cut their teeth against Iran-backed forces on Syrian and Iraqi battlefields. It risks deepening Sunni-Shia tension in South Asia, as the Saudis cajole Pakistan, whose Shia population is the second largest in the world and has close ideological links to neighbouring Iran, to join its anti-Iran front in Yemen.^[fn]Shias compose around 20 per cent of Pakistan’s approximately 200 million population. Even where Sunnis have little contact with Shia world – like, for example, the Caucasus – sectarian solidarity helps drive local recruits to IS (Crisis Group interviews, North Caucasus fighters, Turkey, January-February 2016). An Egyptian taxi driver recently told Crisis Group that the main threat facing his country emanated from the Shia, though they are less than 1 per cent of Egypt’s population. Crisis Group interview, Cairo, September 2015. Beyond growing anti-Shia popular sentiment in countries with virtually no local Shia, officials’ concern about Iranian proselytising and intelligence operations are common even beyond the country’s usual area of influence. Crisis Group interviews, security officials and politicians, Tunis and Rabat, 2015-2016.Hide Footnote

It blends, too, with Sunni Arabs’ profound sense of victimisation, a sense deepened by the West’s focus on IS atrocities that largely overlooks – or, in the case of Iraq, appeared to facilitate – the slaughter of Sunnis by Iran-sponsored regimes and militias. As the 2011 unrest spread, the Arab Sunni world’s traditional power centres, such as Egypt, were destabilised, which left others scrambling to compensate. Saudi Arabia has tried to fill the vacuum, but in part by escalating sectarian sentiment: dangerous terrain on which to compete with IS.^[fn]The Saudis’ recent execution of the popular Shia sheikh, Nimr al-Nimr, both illustrated the depth of and, even if not deliberately, aggravated Sunni-Shia tensions. “Sheikh Nimr al-Nimr: Saudi Arabia executes top Shia cleric”, BBC News, 2 January 2016.Hide Footnote

Secondly, though a catalyst for the fourth wave was the toppling of dictators, its roots lie partly in persistent authoritarianism. Leaders and regimes, backed by major powers, have for decades clung to power through violence and repression. Their regimes provided relative stability, but their misrule did much to rot institutions, erode state-society relations and pave the way for the turmoil that followed their overthrow. In particular, the determination of Maliki (Iraq) and Assad (Syria) to consolidate or hold onto power largely provoked the wars that paved the way for IS; Assad deliberately radicalised the opposition as a regime-survival strategy.

Gloomy prospects for reform in countries, especially in the Arab world, that have not yet succumbed to violence contribute to anti-establishment sentiment, particularly among young people, and lend credence to jihadist criticism of corrupt local regimes. Western powers’ silence at their allies’, notably Egypt’s, backsliding and the dissipation, over the past few years, of their support for reform elsewhere confirms deep-seated perceptions of double standards, again strengthening jihadist narratives.

Thirdly, African leaders are for the most part more united against jihadists than their Middle Eastern counterparts, even if, in some cases, no less reluctant to let power go. Their challenge lies more in the weakness of states; their limited writ in neglected peripheries; and the inability of security forces, intelligence services and other institutions to respond with the required dexterity. The precedents of Boko Haram and jihadists in Mali, the former morphing from isolated sect to violent insurgency, the latter seizing towns after lurking for years in the desert, are especially troubling. Conditions that enabled both crises – underdevelopment, distrust of the state in its hinterlands, traditional elites’ declining authority, readily-available weapons and clumsy, heavy-handed and ineffective security forces – blight many other states, in Africa and elsewhere.

Lastly, ideological space has opened up. In the Arab world in particular, but also in parts of Africa, other ideologies once used to frame political activity and resistance against repression have lost appeal. Students across the Muslim world who once rebelled by joining socialist movements now have few moderate avenues to express discontent. Arab nationalism has diminished as much as socialism; neo-liberal reform and global governance failed to fulfil their potential and often worsened living conditions; the collapse of the 2011 revolutions has damaged liberal democracy and, particularly dangerously, peaceful political Islam.

Notwithstanding Muslim Brotherhood President Mohamed Morsi’s inept performance as Egypt’s president, the coup and repression under President Abdul Fatah al-Sisi have propelled the country in a still more perilous direction. Jihadist ideologues across the region portray the failure of the Muslim Brotherhood’s gradualism and political participation as vindication of their violent revolutionary strategy, arguments again strengthened by Western leaders’ silence as the Muslim Brotherhood was deposed and its former officials, whom they met officially only a few years ago, languish in Egyptian jails.^[fn]Overall, the Muslim Brotherhood has acted more as a firewall against jihadist movements than a conveyor belt toward them, certainly in the Middle East (in Pakistan, Jamiat-e-Ulema-e-Islam, JUI, has closer ties to Deobandi extremists and helps funnel fighters into Afghanistan but is mostly encouraged to do so by the military). For examples of jihadists attacking the Muslim Brotherhood, see, for example, Bill Roggio, “Zawahiri rebukes Muslim Brotherhood for trusting democracy”, Long War Journal, 3 August 2013; and William McCants, “Who exactly is Abu Bakr al-Baghdadi, the leader of ISIS?”, News­week, 6 September 2015. For the “firewall” versus “conveyor belt” discussion, see Marc Lynch, “Is the Muslim Brotherhood a terrorist organization or a firewall against violent extremism?” Washington Post, 7 March 2016.Hide Footnote

The spread of intolerant strands of Islam – often lumped together under a single label such as Wahhabism or Salafism – has clearly contributed.^[fn]The term Wahhabism refers to the religious revivalist movement initiated in Najd (central Arabia) in the early eighteenth century by Muhammad Ibn Abd al-Wahhab. Denouncing Islam’s perversion over the centuries and Muslim societies’ renewed descent into the state of ignorance (jahiliyya) that characterised the Arabian Peninsula before the advent of Islam, he preached a return to tawhid (exclusive worship of God) and the early practices of the “pious ancestors” – al-salaf al-salih, from which the English term Salafism derives – who comprise the first three generations of Muslims, including the Prophet Muhammad, his companions and their successors. The remedy to Islam’s plight, he argued, was to bypass Islam’s centuries-old legal and theological interpretive legacy and rely instead on the Quran, accounts from the Prophet’s life and the consensus of pious ancestors. Practically, this meant eradicating all forms of popular Islam, including Sufism, saint worship and Shiism, and imposing ritual austerity on believers. Crisis Group Middle East Report N°31, Saudi Arabia Backgrounder: Who are the Islamists?, 21 September 2004. See also Roel Meijer (ed.), Global Salafism: Islam’s New Religious Movement (New York, 2009). Crisis Group Asia Reports, N°s 73, Unfulfilled Promises: Pakistan’s Failure to Tackle Extremism, 16 January 2004; 49, Pakistan: The Mullahs and the Military, 20 March 2003; and 36, Pakistan: Madrasas, Extremism and the Military, 29 July 2002.Hide Footnote  Pakistan’s jihadist threat, for example, cannot be explained without reference to the deliberate Islamisation of laws and support for Islamist proxies by successive rulers, particularly Presidents Zia ul-Haq and Pervez Musharraf.^[fn]Crisis Group Asia Reports N°s 95, The State of Sectarianism in Pakistan, 18 April 2005; and 265, Women, Violence and Conflict in Pakistan, 8 April 2015.Hide Footnote  Across much of the Muslim world, decades of Gulf-sponsored proselytising – through imams, mosques and media, particularly Saudi-funded television – have created a pool of potential recruits who share a general theological disposition with jihadists.^[fn]See also, for example, “Extremism as Mainstream: Implications for Women, Development & Security in the MENA/Asia Region”, International Civil Society Network (ICAN), spring 2014.Hide Footnote  But although Salafis share some broad and conservative tenets, their religious practices and political proclivities are so varied, in no small part because the term tends to be one of self-ascription, connoting legitimacy, that it is hard to draw firm conclusions about a relationship to jihad. Many of today’s most ardent combatants do not come from a Salafi tradition. Nor do the vast majority of Salafis preach or practice violence. In many places they may prove useful allies against those who do.^[fn]See, for example, Rashid Abdi, “East Africa’s Sufi Path to Countering Violent Extremism”, Crisis Group, 15 September 2015.Hide Footnote

Mounting sectarianism, deepening authoritarianism, state fragility, even other ideologies’ dwindling appeal do not mean jihadists’ tenets will soon inspire mass appeal. Polls consistently show much of what they promote resonates broadly: opposition to corrupt local regimes, U.S. policy in the Muslim world, Israel and its treatment of Palestinians and Western influence, as well as a greater role for Islam in public life. But the strands distinguishing violent jihadists from political Islamists, inspire much less support. Their social vision tends to be too austere. Even for those to whom a caliphate might on some level be alluring, violent transnational revolt or drawing the West into an apocalyptic war to establish it is less so. Killing Muslim civilians is deeply unpopular without the kind of hatred only sustained conflict generates.^[fn]Cameron Glenn, Garrett Nada and Melissa Nozell, “Muslims Condemning Violent Extremism? Count the Ways”, U.S. Institute of Peace, 17 March 2015.Hide Footnote  The revulsion jihadists’ extreme bloodshed inspired in the past, notably in Algeria and Iraq, partly explains previous waves’ reversal – though the widening Sunni-Shia fault line and images of the Syrian carnage on local media across the Muslim world risk inuring many to violence.^[fn]In Algeria and Iraq, the enormous violence against civilians perpetrated respectively by the Sala­fist Group for Preaching and Combat, which later became AQIM, and AQI provoked widespread revulsion that partly enabled their defeats.Hide Footnote

That jihadist tactics and ideology look unlikely to resonate widely is partly moot. Revolutions throughout history have relied less on majorities than on a dedicated core able to exploit opportunities in chaos. The reach and resources these movements now command mean that any further breakdown in the Muslim world, from West Africa to South Asia, risks empowering an extremist element, whether jihadists provoke the crisis or, more likely, profit from its violent evolution. But it does suggest that countering their ideology should be but a small part of the response.^[fn]Clearly, though, in some countries it is more important than in others. In Pakistan, for example, unless radicalism through the brainwashing of youths in hundreds, if not thousands, of jihadist or sectarian madrasas ends, there will be no lack of foot soldiers for their causes.Hide Footnote  The more urgent priorities are to reinvigorate efforts to end wars, dial down rivalry between states and prevent other crises erupting, particularly by responding sensibly to terrorist attacks and by encouraging leaders toward inclusion and reform.

Although the pace at which the jihadist landscape is evolving means any description can offer only a snapshot, the main contours of the fourth wave are clear. Despite its loss of Ramadi, IS appears firmly in control of the Sunni heartlands in Iraq and parts of eastern Syria. It has not replicated elsewhere its dramatic success there, but it is expanding in Libya, the Sinai, Yemen and Afghanistan, winning recruits in other war zones and has coordinated or inspired attacks in the West.

In part hidden by IS’s rise, al-Qaeda has adapted. Some affiliates, particularly in Syria and Yemen, are increasingly powerful. Exploiting opportunities opened by local conflicts, they have shifted emphasis from attacking Western interests to capturing territory, targeting local regimes, often obscuring their links to al-Qaeda and, in places, acting with some pragmatism. Whether over time this will alter the identity of al-Qaeda or any local branch or help it recover ground lost to IS remains unclear.

The jihadist evolution has accelerated debate over tactics, strategy and doctrine: the killing of other Muslims, particularly Shia; how and when to impose Islamic rule; and whether the end goal is to overthrow the nation-state system or simply specific “un-Islamic” regimes. Since 2011, more movements have seized territory, supplanting the state while prompting, in some cases, a shift in relations with populations in areas they control.

In July 2014, IS routed the Iraqi army in Mosul, Iraq’s second largest city, capturing substantial U.S.-supplied weaponry. In a few weeks, it swept across the north and west of the country, linking up to strongholds in eastern Syria. Its previously almost-unknown leader, Abu Bakr al-Baghdadi (a nom de guerre), appeared in Mosul’s central mosque to declare a new caliphate and himself the “commander of the faithful” and demand fealty from Muslims worldwide. IS forces destroyed part of the Iraqi-Syrian border, the first time a jihadist group had claimed supranational territorial authority.

Tens of thousands of foreigners have joined, many lured by sophisticated online recruitment. Its choreographed violence, trumpeted over social media, are designed partly to sow fear and partly – like bin Laden’s attacks earlier – to generate headlines. Its enslavement of women generates headlines, too, and serves to recruit young men whose socially conservative background makes access to women difficult. It aims to expand by capturing territory and winning recruits in other collapsed states; dividing societies through terrorist attacks; and, it says, provoking a battle with Western powers that paves the way for a new Islamic order.

Above all, though, IS is a movement rooted in the recent history of Iraq and Syria and with a now predominantly Iraqi leadership. The ouster of Saddam Hussein, a largely secular dictator ruling a country with a limited history of Salafi-jihadism, and the policies adopted afterwards by the U.S. occupation were enormous gifts to extremists. De-Baath­i­f­ic­ation – the firing of many officials – and dismantling the army left hundreds of thousands of mostly Sunnis jobless. Power shifted from Sunni urban to Shiite and Kurdish provincial classes. The new political system, which expressly apportioned power by sect and to which Sunnis struggled to adapt, also served their interests poorly.^[fn]Sunni leaders struggled to redefine their political identity to fit an explicitly sectarian system; to a degree they still do. See Crisis Group Report, Iraq’s Sunnis and the State, op. cit.Hide Footnote  Violence and torture by U.S. forces and local allies was well known in Iraq even before the Abu Ghraib scandal and inspired wide outrage.^[fn]

To build the insurgent movement that became AQI and later IS, Zarqawi, a Jordanian militant who arrived in Iraq after fleeing Afghanistan as the Taliban were ousted, could thus tap a rich vein of Sunni discontent, as well as networks of Levantine militants he had forged in South Asia. Drawing on a new generation of jihadist ideologues, he found fertile ground for polarising the country along sectarian lines, an approach based on his deep hatred of Shia but also cold strategic logic, given the reversal of Sunni fortunes. In the early years, however, AQI was only one of many groups opposing the occupation and new government. While the leadership of his group included many foreigners, ex-regime elements dominated others.^[fn]Crisis Group Report, In Their Own Words, op. cit.; Hanieh and Rumman, The “Islamic State” Organization, op. cit., pp. 25-26. Also see Michael Weiss and Hassan Hassan, ISIS: Inside the Army of Terror (New York, 2015) and Burke, 9/11 Wars, op. cit.Hide Footnote

Though the U.S. invasion prepared the way for IS’s rise – without it there would be no IS – the aftermath of the Awakening, the tribal revolt against AQI, and the escalation of Syria’s war were as important. By the time the U.S. killed him in 2006, Zarqawi had helped provoke a sectarian civil war in Iraq. His brutal tactics, however, criticised locally and by al-Qaeda’s top leadership, cost his movement support. Particularly in parts of Anbar province, tribes chafed under foreign militants’ religious strictures, disregard for local power structures and attempts to monopolise smuggling revenue. These considerations, together with promises of U.S. support, push back against Iranian influence and substantial payments, led them to realign with the U.S. against al-Qaeda. More than 100,000 tribal fighters, their capacities reinforced by the U.S. surge, routed the militants.^[fn]See Crisis Group Reports, Iraq after the Surge I and Iraq after the Surge II, both op. cit.; also, Ala Ali, “Security, Religion and Gender in al-Anbar province; a Focus-Group based Conflict Analysis”, International Civil Society Action Network, 7 August 2014.Hide Footnote

The revolt against AQI was built on the understanding Sunnis would gain a greater stake in the state and its security forces. Instead, in the run-up to the 2011 U.S. troop withdrawal, Prime Minister Maliki increased sectarian rhetoric; stopped paying salaries of and otherwise cut off the tribal leaders who had risen up; and did not integrate their militias into the security forces as promised, instead arresting many. Syria’s crisis deepened the sense of escalating regional war, pushing him closer to Tehran, with which he shared fear that Assad’s overthrow could usher in a hostile, Sunni Islamist-led regime in Damascus.

The crushing by Iraqi security forces of protests that broke out in Sunni-majority towns (Falluja and Hawija) over the winter of 2012-2013 was the tipping point. It made it harder for Sunni leaders inclined to work across sects to do so and gave a green light to more extreme movements to stage armed retaliations, deepening both sides’ conviction that the clash was existential. As violence intensified, Maliki portrayed virtually all Sunni opposition as terrorist, while refusing to label as such no less brutal Shiite violence. U.S. and UN Security Council acquiescence – their support for Maliki belied token calls for political inclusion – fed the sense of Sunni victimisation that the Assad regime’s violence against Sunnis next door exacerbated.^[fn]Crisis Group Report, Iraq: Falluja’s Faustian Bargain, op. cit.Hide Footnote  Zarqawi’s successors by then had regrouped and, learning from his experience, prioritised their base in the Sunni community. The group had become predominantly Iraqi, partly because many foreigners had abandoned it for Syria, partly through tightening links with remnants of Saddam’s regime, many of whom had radicalised, with networks strengthened in U.S. and Iraqi jails.^[fn]Weiss and Hassan, ISIS, op. cit.Hide Footnote  It replenished its ranks and leadership via jailbreaks, then by paying disaffected tribesmen well. By mid-2014, it had infiltrated most Iraqi Sunni-majority cities. Though dynamics varied, local military councils and ex-insurgent factions often allied with jihadists, whose military superiority then translated into dominance. When the renamed IS captured Mosul and the Sunni heartlands in June 2014, the Iraqi army, hollowed out by corruption and incompetence and seen as a Shiite occupation force, mostly melted away. That many inhabitants of IS-captured areas celebrated “liberation”, despite the memories left by Zarqawi’s militants a few years earlier, showed the decay of their relations with the state.^[fn]Crisis Group Middle East Briefing N°38, Iraq’s Jihadi Jack-in-the-Box, 20 June 2014.Hide Footnote

Crisis Group Middle East Reports N°s 34, What Can the U.S. Do in Iraq?,  22 December 2004; 50, In Their Own Words: Reading the Iraqi Insurgency, 15 February 2006; and 52, The Next Iraqi War? Sectarianism and Civil Conflict, 27 February 2006.Hide Footnote

Zarqawi’s successors by then had regrouped and, learning from his experience, prioritised their base in the Sunni community. The group had become predominantly Iraqi, partly because many foreigners had abandoned it for Syria, partly through tightening links with remnants of Saddam’s regime, many of whom had radicalised, with networks strengthened in U.S. and Iraqi jails.^[fn]Weiss and Hassan, ISIS, op. cit.Hide Footnote  It replenished its ranks and leadership via jailbreaks, then by paying disaffected tribesmen well. By mid-2014, it had infiltrated most Iraqi Sunni-majority cities. Though dynamics varied, local military councils and ex-insurgent factions often allied with jihadists, whose military superiority then translated into dominance. When the renamed IS captured Mosul and the Sunni heartlands in June 2014, the Iraqi army, hollowed out by corruption and incompetence and seen as a Shiite occupation force, mostly melted away. That many inhabitants of IS-captured areas celebrated “liberation”, despite the memories left by Zarqawi’s militants a few years earlier, showed the decay of their relations with the state.^[fn]Crisis Group Middle East Briefing N°38, Iraq’s Jihadi Jack-in-the-Box, 20 June 2014.Hide Footnote

Developments within the Sunni community as much as its distrust of Baghdad aided IS’s advance. The broken promises to the Awakening destroyed or discredited much of the non-jihadist Sunni opposition that had gambled on working with the U.S. and the Iraqi state and distanced Sunnis from their elites. With the help of ex-officials of Saddam Hussein’s regime skilled in the repressive tactics of an authoritarian state, IS faced little resistance as it fragmented social and political structures that it feared could some day resist its rule. The most notorious way it did this was ruthlessness with potential rivals, particularly those involved in the Awakening who refused to join. No less crucially, however, it provided an avenue for social mobility to Sunnis who lacked a champion within their community.

IS has thus weaved a web of marginalised groups and classes whose interests, if not beliefs, align with its own. Its “Tribal Bureau” exploits tribal divisions, peeling off support, empowering younger leaders or weaker clans and turning clans against each other. Many youths, especially but not only within tribes, backed it to protest their elders’ enrichment by Maliki’s patronage. Some business people, former bureaucrats and others in the middle classes in places like Mosul, whose livelihoods were upended after Saddam Hussein’s overthrow, could recover their standing and profit under IS. Rural classes found in it a way to strike back at what they saw as exploitative urban elites. Paradoxically for a group that promotes an uncompromisingly austere vision of Islam, IS leaders initially showed, at least in Iraq, some flexibility in enforcement of religious codes, depending on what they believed the local market would bear.^[fn]Crisis Group telephone interviews, residents of IS-controlled areas, October 2015-January 2016.Hide Footnote

Of course, even those who benefit live under painful strictures: movement restrictions, imposed in early 2015, which create a sense of isolation; a war economy’s deprivations; and an escalating bombing campaign. But some have profited, and for many IS still inspires less resentment than Baghdad. Plus, many Iraqis are inured to repressive rule stretching back decades.

The story is different in Syria, into which what was becoming IS expanded in 2011. Baghdadi deployed Jolani, a top lieutenant, who quietly built Jabhat al-Nusra into a large insurgency, thanks partly to IS financing but mostly by working with others, keeping al-Qaeda ties quiet, winning support through his movement’s relative discipline and profiting from the war’s radicalisation. In April 2013, Baghdadi announced IS would subsume al-Nusra. Jolani rejected the merger and pledged allegiance to al-Qaeda leader Ayman al-Zawahiri. After a failed attempt to mediate, Zawahiri ruled that the Iraqi and Syrian branches would be separate al-Qaeda affiliates, in effect siding with Jolani. Baghdadi rejected this.^[fn]Crisis Group Report, Rigged Cars and Barrel Bombs, op. cit. Though the clash between Baghdadi and Jolani was the spark, the split between al-Qaeda and IS had long been brewing. As far back as 1990s Afghanistan, relations between Zarqawi and al-Qaeda leaders had been strained. His tactics in Iraq drew regular criticism from Zawahiri and leading al-Qaeda ideologues, who questioned his brutality against other Muslims and focus on killing Shia and capturing territory rather than targeting the U.S. No IS leader since Zarqawi appears to have pledged allegiance to either bin Laden or Zawahiri. See Hanieh and Rumman, The “Islamic State” Organisation, op. cit.Hide Footnote  The schism has since played out in public recriminations and aggressive IS efforts to win over al-Qaeda loyalists elsewhere. In Syria, many Iraqi and other foreign jihadists defected to IS, radicalising it further. Though some al-Qaeda veterans stayed with it, al-Nusra became increasingly Syrian, and most of its rank-and-file, if not leaders, focus on Syrian, not transnational concerns.

IS initially targeted not the regime but rebel-held areas, trying to conquer the Sunni opposition in Syria as it had in Iraq. The regime left it mostly undisturbed and escalated against rebels, viewing them as a graver threat and IS’s expansion as an opportunity to portray all opposition as terrorist. Fractious rebel groups at first veered between subordinating to IS and confronting it, but by early 2014, IS’s actions, including killing popular rebel leaders, led to more coordinated opposition. Initially al-Nusra stayed out of the fray, but was drawn in against IS. Beaten back from the north west around Aleppo, IS was forced to retreat to eastern Syria, but this also freed up resources for its dramatic capture of Mosul and expansion in Iraq.^[fn]Crisis Group Report, Rigged Cars and Barrel Bombs, op. cit.Hide Footnote

Thus, although its de-facto capital is Raqqa, connected since ancient times to Iraq by Euphrates River trade, IS’s Syrian roots are shallower. Within Iraq’s Sunni minority, it has eradicated opposition, empowered marginal groups, invested in governance and shown flexibility. In Syria, where Sunnis are a majority and powerful alternatives exist, it controls only some Sunni-majority areas and relies more on force, despite forming some alliances and often operating by persuasion or bribery. These differences notwithstanding, its defeat in either country appears remote. Though unlikely to advance into Iraq’s Shia or Kurdish heartlands or mount a serious assault on Damascus or Syria’s Alawite regions, it appears resilient in core areas – partly thanks to its military prowess and ties to elements of the Sunni community, partly, as described, because its foes are divided.

The degree to which, over time, it can maintain support or acquiescence, particularly in Iraq, is uncertain. Dwindling revenues might tip its balance of coercion/co-option toward the former, which could fray its roots in communities. However, it is as embedded in the local economy as in society. It generates part of its revenue through oil production, looted banks, gold mines, wheat farming and sale of antiquities, but most now comes from taxes of various sorts, confiscation and extortion, all hard for international sanctions to squeeze without inflicting wide suffering. Even as it has faced greater military pressure and lost territory over the past year, it appears durable.

IS aims to expand beyond its regional base by establishing provinces (wilayaat) through aggressive recruitment and luring in other groups. It appears less discerning in allowing groups to join than al-Qaeda is about accepting new affiliates.^[fn]See, for example, Barak Mendelsohn, The al-Qaeda Franchise: The Expansion of al-Qaeda and Its Consequences (Oxford, 2016).Hide Footnote  It has had some success elsewhere but nothing like in Iraq – perhaps unsurprising given its strong Iraqi identity and roots in conditions there.^[fn]A recent UN report argues that by mid-December 2015, 34 groups had declared their affiliation to IS. “Report of the Secretary-General on the threat posed by ISIL (Da’esh) to international peace and security and the range of United Nations efforts in support of Member States in countering the threat”, United Nations Security Council, 29 January 2016. This report treats only the largest.Hide Footnote

In Libya, around the coastal town of Sirte, a former stronghold of the Qadhafi regime, and nearby towns, IS recruited from the local Ansar al-Sharia branch, taking advantage of a security vacuum. Although consisting of only a few hundred men, it made inroads by brokering deals with local leaders who had nowhere else to turn for protection; the area has no significant militias of its own, as most residents are former regime loyalists “defeated” in the 2011 war. Over 2015, IS won control of a 200-300km coastal stretch between Sirte and Ben Jawwad. Its emissaries appeared in greater numbers after June 2015, both Libyan returnees from Syria and foreigners, including notable Iraqi IS commanders.^[fn]Until early 2015, most local IS leaders were Libyan, but over time the flow of foreigners increased. Crisis Group interviews, Sirte and Harawa residents, Harawa, March 2015; refugees from Sirte, al-Bayda, November 2015.Hide Footnote

Initially, IS did not impose strict rules on residents, provided women were veiled, and local groups did not attempt to take up arms against it. Killing primarily targeted foreigners, especially Christian refugees. But over time, especially after a group of Sirte residents (led by a Salafi imam) tried to rise against it in summer 2015, repression became more violent. Militants began to publicly execute security officials and residents accused of spying or engaging in un-Islamic practices; demand young girls be handed over for forced marriage and de-facto rape; and, at checkpoints along Libya’s main coastal road, arrest individuals identified as state employees or oil sector workers.^[fn]Crisis Group telephone interviews Ben Jawwad and Brega residents, January 2016; interviews, Sirte residents, al-Bayda, November 2015.Hide Footnote  IS funding sources in Libya are murky but appear to include local taxation (including on smuggling), extortion, looting of banks, kidnapping and wealthy sponsors. The group ransacked oil fields and attacked ports and refineries, but there is no evidence that it smuggles oil.^[fn]Crisis Group interviews, residents and security guards, Harawa, March 2015; telephone interviews, residents and security guards, Sidra, January 2016.Hide Footnote

The Libya branch appears to have the closest operational ties of all IS-linked groups to the leadership in the Levant. The longer it can hold on, and the more Iraq and Syria veterans and foreigners flow in, the more dangerous it will become. In early 2016, it expanded east, tightening its grip on Ben Jawwad (the last town before major oil facilities on the coast) and attacked oil and gas infrastructure around Sidra. Its expansion westward is checked by the Misrata-aligned revolutionary brigades, which are distrusted by Sirte locals but could perhaps oust IS were their leaders not reluctant to lose men or risk being outflanked in their hometowns.

Elsewhere in Libya, IS has not made significant progress. It has a limited, static presence in Benghazi (where it is believed to have coordinated with the Shura Council of Benghazi Revolutionaries, a mostly non-jihadist coalition fighting against forces under the command of General Khalifa Haftar).^[fn]General Khalifa Haftar commands Operation Dignity, an offensive led by army units and other armed groups aligned to the Tobruk-based government against Islamist armed groups in Benghazi. Haftar and his supporters purport to be fighting terrorist groups; critics accuse them of also attacking non-radical groups.Hide Footnote  It has been pushed out of Derna, another city with a history of jihadist activity, where Ansar al-Sharia and some al-Qaeda-linked groups dominate. Libya is not torn along the sectarian fault lines of Iraq or Syria, and its chaotic and fluid militia scene is more difficult for IS to exploit, although some Iraq dynamics, notably the rifts between the state and communities associated with the former regime, are evident.

In Egypt’s Sinai region, Ansar Bayet al-Maqdis (ABM), a mostly Bedouin group rooted in the area’s radicalisation in the early 2000s (partly the result of the second Palestinian intifada) and a wave of repression in 2005-2007 that followed terrorist attacks on tourist resorts in Taba, Dahab and Sharm al-Sheikh, declared allegiance to IS in November 2014.^[fn]The neglect of the Sinai’s populated north east, security and intelligence services’ heavy-handed tactics, the Gaza blockade and smuggling economy it encouraged that distorted the local economy, the weakening of traditional tribal authority and the crackdown on the Muslim Brotherhood all helped create the social conditions in which armed groups thrive. Crisis Group interviews, Sinai residents, al-Arish and Sheikh Zuwayed, May and July 2015.Hide Footnote  IS-Sinai recruits mostly locally, as it did while still ABM, but can draw on militants from the Nile Valley, as well as carry out major attacks there, including in Cairo. In north-eastern Sinai, it has mounted a significant challenge to the Egyptian military through truck bombings against security installations, the wide­spread use of improvised explosive devices (IEDs) and at times large-scale battles in towns. Some of its expertise may have come from veterans of Syria or Iraq. It has advanced weaponry – having used MANPADS (man-portable air defence systems) at least once in 2014 and Russian-made anti-tank Kornet missiles in 2015 – and claimed responsibility for the downing of a Russian civilian airliner in October 2015.^[fn]Ibid. Also Crisis Group interviews, Egyptian and foreign military officials, Cairo, 2014-2015; “Notice Regarding Egypt Sinai Peninsula”, U.S. Federal Aviation Authority, 5 November 2014; and Jeremy Binnie, “Sinai militants attack Egyptian patrol boat”, IHS Jane’s Navy International, 19 July 2015.Hide Footnote

In Yemen, IS, which announced itself in November 2014, has to contend with a well-established and strong al-Qaeda movement that has demonstrated its staying power. Still, various old al-Qaeda and other militants have pledged loyalty to Baghdadi, most prominently Jalal Mohsen Saeed Baleedi, a former AQAP member from Abyan, who was killed in a suspected U.S. drone strike in February 2016. IS appears strongest in Hadramout, Aden and Lahj, with a growing presence in Abyan. It is more brutal and less concerned about heeding local norms and forging local alliances than al-Qaeda but recruits from the south’s disillusioned and impoverished youth. Attacks on holy sites of Zaydis, the Shiite Islam sect to which Huthis belong, appear aimed at stoking sectarian divisions so IS can present itself as the protector of Sunnis, tactics that serve it well in Iraq. Al­though for now fighting is not only along sectarian lines, and traditionally primary identities in Yemen have been tribe, clan, region or political affiliation rather than sect, deepening sectarian polarisation may play into IS’s hands.

Some former Pakistani Taliban commanders, traditionally more sectarian than their Afghan counterparts, established IS in Afghanistan’s easternmost provinces. Throughout 2015, Taliban splinter groups also sporadically re-hatted for diverse reasons.^[fn]These include discontent with the Pakistan-based Taliban leadership, IS recruiters offering higher salaries, competition over drug or extortion routes and ties to Salafism, among others. See, for example, Antonio Giustozzi, “A Gathering Storm? The Islamic State campaign in Eastern Afghanistan”, Jane’s Terrorism and Insurgency Monitor, 13 November 2015; “Why Taliban special forces are fighting Islamic State”, BBC World Service, 18 December 2015; and Hekmatullah Azamy and James Weir, “Islamic State and Jihadi realignments in Khorasan”, The Diplomat, 8 May 2015.Hide Footnote  Some districts have seen fierce fighting between Taliban and IS militias. The Taliban conglomerate, however, remains the preeminent armed opposition, with deep roots in parts of Pashtun society and growing reach in the north.^[fn]The Taliban insurgency combines various Afghan factions, joined by Pakistani militants like Lashkar-e-Tayyaba. The Islamic Movement of Uzbekistan in August 2015 announced it had declared allegiance to IS, but appears to have fought alongside the Taliban in its Kunduz offensive in early 2015. The Taliban now control more territory countrywide than at any point since the U.S.-led intervention, according to the U.S. Special Inspector General for Afghanistan Reconstruction (SIGAR)’s 30th Report to Congress (6 January 2016).Hide Footnote  In the southern heartlands, IS’s Salafi-jihadist ideology is alien to the Deobandi and rural Pashtun traditions the insurgency draws from.

Taliban leaders nonetheless appear to take the IS threat seriously. The caliphate declaration, with Baghdadi as “commander of the faithful”, directly challenged the legitimacy of the Taliban’s emirate and Mullah Omar, who was thought to be still alive and to whom al-Qaeda leaders and the Pakistani Taliban had pledged bayat (allegiance, fealty). Though Zawahiri has since pledged bayat to Omar’s successor, Mullah Mansour, the latter enjoys nothing like his predecessor’s prestige or legitimacy.^[fn]Al-Qaeda leaders’ bayat to Taliban leaders stems from bin Laden’s pledge of allegiance to Mullah Omar as commander of the faithful in 1990s Afghanistan. Zawahiri also pledged allegiance to Omar and now pledges allegiance to his successor, Mullah Mansour.Hide Footnote  Recent Taliban battlefield successes – in the north east, where it briefly captured a provincial capital, Kun­duz, for the first time since 2001, and then in the southern heartlands – have solidified support for Mansour, but this would weaken if he were to opt, under Pakistani pressure, for a negotiated settlement.

By mid-2015 most of Russia’s North Caucasus insurgency, the Caucasus Emirate, which had loose ties with but was never an affiliate of al-Qaeda, had sworn allegiance to Baghdadi. Shortly thereafter, IS announced creation of its “Wilayaat Kavkaz”. The Caucasus branch, however, has been decimated since Russian security services cracked down in 2013. Together with the allure of fighting in Syria, that appears to have driven many Russian jihadists to the Levant. Militants in the North Caucasus reportedly have also not received the financial support they expected from Raqqa. Thus far, the Caucasus appears less a priority for IS than Libya or South Asia, though IS fighters with roots in the region often call for Muslims there to attack the Russian state in its name.^[fn]According to Russian officials, some 5,000 Russian citizens now fight in Syria and Iraq. See, for example, Crisis Group Report, The North Caucasus Insurgency and Syria, op. cit. For background on the Caucasus insurgency see Crisis Group Europe Reports N°s 220 and 221, The North Caucasus: The Challenges of Integration (I), Ethnicity and Conflict, and The North Caucasus: The Challenges of Integration (II), Islam, the Insurgency and Counter-Insurgency, both 19 October 2012.Hide Footnote

Boko Haram’s joining IS in March 2015 appears to have been motivated partly by Shekau’s desire, after suffering territorial losses, for publicity and the legitimacy harnessing the movement to the global jihad might garner. Thus far, little has changed about the organisation’s capability, tactics or identity beyond more polished online promotion. It is not clear that operational ties to Raqqa exist. Although there are fighters from outside the Lake Chad Basin region among its ranks, foreigners are less numerous than in other African jihadist movements.^[fn]A Senegalese preacher, arrested by Nigerien authorities in Niamey, admitted the presence of Mauritanians, Senegalese and Sudanese as well as Chadians in Boko Haram’s ranks. Crisis Group’s viewing of police interrogation records, Dakar, October 2015.Hide Footnote  Boko Haram is likely to continue causing tremendous suffering in the hinterlands it plagues and elsewhere, but linking it too directly to the global jihadist movement risks misdiagnosing the threat it poses.

IS’s inability thus far to repeat its Iraq success does not diminish its significance. Understanding its Iraqi roots and armed capability is critical but only partly captures its protean nature: both Iraqi Sunni resistance and transnational millenarian force; a source for some of protection, for others of adventure or identity; a state structure, but also a revolutionary idea. Its resources and military capability and the remote prospects for eradicating it in the near term make it a more difficult challenge than any prior jihadist movement. It nimbly exploits cleavages, particularly along the Sunni-Shia fault line, but also others, like that between Ankara and the Kurds, where its attacks risk contributing to the instability of a country threatened on multiple fronts.^[fn]See, for example, Jean-Marie Guéhenno, “Don’t sacrifice Turkey to save Syria”, The Guardian, 29 February 2016.Hide Footnote

The lack of avenues for peaceful dissent and opportunities for young people makes many societies vulnerable to its recruitment, even if it lures only tiny minorities. IS has devised a paradigm of mobilisation both local and opposed to a global establishment. By recruiting online as much as through religious networks that earlier move­ments relied on, and by filling the void left by many states’ failure to provide an alternative, it taps new markets for jihadist recruitment.

As IS has emerged, al-Qaeda has evolved. Drone strikes and military offensives have weakened its core in the Pakistani tribal areas, and Zawahiri’s grainy video sermons appear drab beside IS’s flashy online promotion. But despite IS efforts to win over al-Qaeda affiliates in the Maghreb, Somalia, Syria and Yemen, no top commanders, most of whom rubbed shoulders with bin Laden and Zawahiri in South Asia, have defected. Some affiliates have become more powerful than ever, seizing territory, grafting themselves onto local insurrections and fighting beside rather than seeking to crush or absorb other Sunni movements.

In Syria, as described, Jabhat al-Nusra initially lost out from IS’s rupture with al-Qaeda. Many of its foreign fighters joined IS, but it has regrouped and with a stronger Syrian identity is second in strength among rebels in the north only to Ahrar al-Sham.^[fn]Ahrar al-Sham is the most powerful member of the rebel Jaish al-Fatah coalition, with strongholds particularly around Aleppo. See Section III.D. Also Crisis Group Reports, Rigged Cars and Barrel Bombs and New Approach, both op. cit.Hide Footnote  Even before the split, it was more restrained in attacks on civilians, tempered emphasis on ideology in its governance while attempting to serve the local population, and worked with other rebels, with whom it maintains close operational ties. Its fighters and suicide bombers are the insurgency’s elite attack force, pivotal to offensives around Aleppo and Idlib in summer 2015.^[fn]Crisis Group interviews, rebel factions’ officials, Turkey and Jordan, 2013-2015.Hide Footnote

U.S. officials say there are still individuals in the movement with close ties to al-Qaeda’s leadership and who plot against the West.^[fn]Crisis Group interviews, U.S. officials, Washington DC 2015; also Crisis Group Report, Rigged Cars and Barrel Bombs, op. cit.Hide Footnote  A peace process that offered some prospect of Assad’s departure might split the fighting majority, whose priority is a new order in Syria, from those with transnational goals – a cleavage that for now Jolani’s rhetoric tends to straddle.^[fn]See, for example, “Nusra leader: Our mission is to defeat Syrian regime”, Al Jazeera America, 28 May 2015; and “For the First Time on Orient News, Comments of the Leader of Jubhat al-Nusra, Abu Mohammad Jolani” [English trans.], video, YouTube, 12 December 2015.Hide Footnote  Efforts by rebels to convince al-Nusra’s leadership to end the group’s al-Qaeda affiliation thus far have been unsuccessful. A growing tendency to assert unilateral authority at other rebels’ expense also damages its reputation within the rebellion, as do public criticisms of rebels (including Ahrar al-Sham) for ties to state backers and engagement in UN-sponsored talks.^[fn]Crisis Group interviews and communications with rebel officials, Turkey, 2015-2016.Hide Footnote  Yet so long as the war continues, al-Nusra is likely to remain potent and mostly Syria-focused, and other rebels will not confront it for fear of losing its vital contribution against the regime.

In Yemen, AQAP is a main beneficiary of the Saudi-led bombardment. Unlike IS, which is new to the country, it has a long history and an extensive social and family network there. It is ensconced in Hadramout and, following the Huthis’ expulsion, parts of Aden. The group also is now active in Taiz and al-Bayda. After the 2011 revolution, it created a network of affiliates known collectively as Ansar al-Sharia, that are associated with al-Qaeda but have less rigorous membership standards, allowing them to recruit more widely and avoid an explicit al-Qaeda association. It has weathered the death of its leader, Nasir al-Wuhayshi, killed by a drone strike in June 2015. His longtime deputy, Nasir al-Raimi, a trainer in an al-Qaeda camp in the 1990s, appears to have quickly cemented his authority. His personal ties, the movement’s prominence as the affiliate closest to the al-Qaeda leadership – as well as the significance of breaking a pledge of allegiance – mean it is unlikely to abandon al-Qaeda for IS.

Precise relations between AQAP and other anti-Huthi militias in the south, notably the strong, non-Islamist, secessionist Hiraak, are difficult to define. In some places – Aden after its liberation, for example – they already fight each other. In others, such as Taiz, where for now they align against Huthis, these alliances may prove temporary. Clearly, though, the war is a massive boon for al-Qaeda. Even if UN mediation yields a peace deal between the Huthis and their foes – which still appears some way off – ousting it militarily will be tough, especially with the southern question unresolved.

Though expelled by French and Chadian forces from towns in northern Mali they controlled for half of 2013, AQIM militants have gained footholds in Libya, which has become a hub for jihadist networks stretching south into the Sahel, west to Tunisia and Algeria and east to the Levant battlefields. Libya’s security vacuum enabled the attack on the Amenas hydrocarbon complex in eastern Algeria in January 2013, carried out under the leadership of former AQIM commander Mokhtar Belmokhtar.^[fn]Crisis Group interviews, security officials Derna residents, al-Bayda, Tripoli, 2015.Hide Footnote  In the Sahel’s fragmented militant scene, groups regularly strike alliances and splinter, but for now, Belmokhtar, who has formed a new group (al-Mourabitoun) and AQIM leader Abdelmalik Droukdel, both with Afghan-generation ties to al-Qaeda, look unlikely to switch allegiance to IS. The former has claimed a hand in the recent Bamako and Ouagadougou attacks.^[fn]Jason Burke, “Mokhtar Belmokhtar: the ‘uncatchable’ chief of Africa’s Islamic extremists”, The Guardian, 21 January 2016.Hide Footnote

Lastly, al-Shabaab in Somalia has withstood in the past few years offensives by an African Union (AU) mission, the loss of major population centres, ideological attacks from other Islamists, including earlier jihadist leaders, and, in 2013, an internal power struggle. Part of its resilience lies in the weakness of its rivals: the transitional authorities’ inability to develop credible alternative local governance across rural south-central Somalia and AU forces’ often clumsy operations. But it lies also in the movement’s strengths, particularly its roots in parts of that region and its tactical flexibility.^[fn]See Crisis Group Africa Report N°99, Somalia: Al-Shabaab – It Will Be a Long War, 26 June 2014; and Africa Briefings N°s 85, Kenyan Somali Islamist Radicalisation, 25 January 2012; and 74, Somalia’s Divided Islamists, 18 May 2010. Since 2008, al-Shabaab blends insurgent tactics with terrorist attacks: besieging towns, breaking supply lines, conducting night raids while striking in urban areas beyond its direct control. It pays fighters well thanks to diverse income sources: donations, extortion, even in parts of Mogadishu, looting, kidnapping and taxing piracy and smuggling. Its outreach is effective; online content targets the Somali diaspora and appeals, like IS’s, to young men’s desire to belong to a group as much as religious credentials. Outreach in villages stresses need to defend Somalia and Islam from invaders. Foreign influence has shaped its ideological and tactical development but not swamped its Somali core. It still aspires to create an East African regional emirate, and much outreach is now in Kiswahili not Somali.Hide Footnote  Over the past six months, it has been launching set piece attacks against AU bases and retaking as many locations as it loses. At least by night, it again controls much of Mogadishu.^[fn]Crisis Group observations, interviews and telephone interviews, Mogadishu, January 2016.Hide Footnote  Once erroneously accused of being foreign, it is now the longest-lived force – politically, socially and militarily – in Somalia.

Abdiqadir Mumin, an al-Shabaab ideologue linked to the diaspora and based in northern Somalia, recently defected to IS with a handful of men. However, al-Shabaab’s new leader, Abu Ubeidah, and his top circle look unlikely to break al-Qaeda ties.^[fn]Zawahiri accepted al-Shabaab’s allegiance only in 2012 – bin Laden’s Abbottabad letters suggest he viewed it as unruly and a liability – but al-Qaeda links go back to 1990s Afghanistan. Don Rassler, Gabriel Koehler-Derrick, Liam Collins, Muhammad al-Obaidi, and Nelly Lahoud, “Letters from Abbottabad: Bin Ladin Sidelined?”, CTC West Point, 3 May 2012.Hide Footnote  The resurgence in Yemen of AQAP, with which al-Shabaab enjoys close links, also makes a split less likely. The movement’s threat to Kenya is especially worrying. Missteps by the government or security forces, like indiscriminate arrests or violence or scapegoating Somalis, could alienate Muslims, drive them into al-Shabaab’s arms and make parts of that country ungovernable. For now, the appointment of Kenyan Somali national security officers in the north has gone some way to bridge the gap between the state and affected communities, although authorities should work more with elders, resolve local disputes al-Shabaab exploits and improve living conditions. Actions have been clumsier in Coast, another region with many Muslims and at risk of al-Shabaab infiltration.^[fn]See Crisis Group Africa Briefing N°114, Kenya’s Somali North East: Devolution and Security, 17 November 2015. Also Crisis Group interviews, Coast province, Kenya, November 2015. A resident of Lamu, for example, said the “year long curfew in Lamu feels like collective punishment”.Hide Footnote

Al-Qaeda’s evolving strategy, documented in letters between affiliate leaders and borne out on the ground, is partly a pragmatic response to new opportunities and the imperative to adapt after the 2011 Arab protests appeared to render it obsolete.^[fn]For example, in 2012, then AQAP leader Nasir al-Wuhayshi urged the AQIM emir to “take a gradual approach” regarding the implementation of Sharia. “First letter from Abu Basir to Emir of al-Qaida in the Islamic Maghreb”, Al-Qaida Papers, Associated Press, undated. Droukdel also wrote to his lieutenants urging them not to alienate locals and even lamented their splitting from Tuareg rebels. See Pascale Combelles Siegel, “AQIM’s Playbook in Mali”, CTC Sentinel, 27 March 2013; also James Cockayne, Hidden Power: The Strategic Logic of Organized Crime (Hurst, forthcoming July 2016). Hide Footnote  It may also reflect the leadership change to Zawahiri and that the split with IS has allowed him to distance the movement from more extreme tactics. If Zarqawi’s experience and the Awakening taught IS to show even less mercy to potential rivals, some al-Qaeda local branches appear to have drawn different conclusions, all of which make strategic sense: more pragmatism with other militants and communities; more caution about killing Muslims; more sensitivity to local norms and popular opinion.

Whether the new strategy heralds a change in the longer-term aspirations of any affiliate is unclear. Some affiliates still attack civilian and predominantly Western targets: AQIM’s recent West Africa hotel attacks, partly aimed at asserting al-Qaeda’s prominence over IS in the region, are an example.^[fn]In Bangladesh, a new branch, al-Qaeda in the Indian Subcontinent, set up by Zawahiri in September 2014, has announced its presence with attacks mostly on bloggers or activists espousing what it considers atheist ideas.Hide Footnote  Even those showing more pragmatism contain contingents, if small, whose goals stretch beyond existing borders. Local commanders have, however, allowed international humanitarian organisations to provide aid in areas they control.^[fn]Crisis Group telephone interviews, humanitarian officials, November 2015.Hide Footnote  Some Qatari officials quietly promote “moderating” al-Nusra – a stretch, but given its strength perhaps worth exploring.^[fn]Crisis Group interview, Qatari official, Autumn 2015; Syrian rebels and Islamist political figures, Turkey and Syria, 2015.Hide Footnote  Other engagement is probably happening, too, given the tactical links between local al-Qaeda branches and Sunni forces backed by regional states, though this may be aimed more at co-opting movements as proxies against Iran and its allies than at taming them.

IS’s emergence and new Arab wars have transformed jihadists’ tactics, strategies and doctrines. A global typology is beyond this report’s scope and would be hard, given the speed at which the scene mutates, groups’ amorphous nature and fluid memberships and the tendency for movements with different ideologies, goals and targets to cooperate. Some debates, nevertheless, have important policy implications. Often framed theologically, they rarely stray far from the strategic: arguments over what Islam permits track closely what works on the ground.

IS and al-Qaeda differences, at least at leadership level, tend to revolve more around tactics and strategy than goals. Both disavow local regimes as un-Islamic and want to expel the West and Russia from Muslim lands and destroy Israel. For both, the aspiration remains a caliphate that upends the international order. Their paths and timeline for getting there, however, diverge sharply, reflecting the contrasting experiences of their leaders and the contexts in which they emerged.

Theologically, the cornerstone of both groups’ armed campaigns is the doctrine of takfir – deeming persons or groups appearing to be Muslim in fact not Muslim, thereby permitting them to be killed with impunity and circumventing the Quranic general prohibition on a Muslim killing another. Takfir can be invoked in three circumstances: against Muslim tyrants; against Muslims serving tyrants or operating in foreign interests; and against Muslims improperly practicing their religion, a provision particularly targeting Shia, who are referred to by so-called takfiris as rawafid (rejectionists of the Sunni-endorsed lines of succession from the Prophet Muhammad). With notable exceptions that jihadists take as inspiration, takfir was used infrequently in Islamic history, was limited to individual cases and had a high juridical bar.^[fn]Mohammed M. Hafez, “Takfir and violence against Muslims”, in Moghadam and Fishman (eds.), Fault Lines in Global Jihad, op. cit.Hide Footnote  Recent jihadist ideologues have reversed all three constraints.

While al-Qaeda and IS, in theory at least, share this expansive conception of takfir, their behaviour differs considerably. Al-Qaeda has usually tried to avoid gratuitous Muslim casualties. Zarqawi’s targeting of Shia in Iraq was a departure that in part reflected his personal hatred of the Shia, but also Iraq’s emerging battle lines and the perception of Iran’s ascendance. Takfir legitimised, for those who believed in it, an all-out Sunni assault on Iran’s perceived proxies in Baghdad.

Zarqawi’s approach was shaped further by new jihadist ideologues, who also borrow from non-Islamic traditions.^[fn]In Architect of Global Jihad: The Life of Al-Qaida Strategist Abu Mus’ab al-Suri (London, 2008), Brynar Lia shows that Suri was an eclectic thinker, sometimes sceptical of Salafism’s doctrinal rigidity because of the constraints it imposes on jihadist strategy.Hide Footnote  Abu Bakr al-Naji, the pseudonym of an unknown author, explained in his Management of Savagery how to create and exploit pervasive violence to unseat a tyrant and consolidate power.^[fn]The Management of Savagery: The Most Critical Stage Through Which the Umma Will Pass (tr., Will McCants, 2006), John M. Olin Institute for Strategic Studies at Harvard University.Hide Footnote  Abu Abdullah al-Muhajir, building on the work of others, including some with al-Qaeda links, such as Abu Yahya al-Libi, fleshed this out, arguing the propriety and utility of specific tactics, including suicide bombings, collateral damage, kidnapping, assassinations and beheadings.^[fn]For more on Abu Abdullah al-Muhajir’s book, Issues in the Fiqh of Jihad, see, “Takfiri literature makes headway in Lebanon”, al-Akhbar English, 11 September 2012. Abdullah Warius and Jarret Brach­man, “Abu Yahya al-Libi’s Human Shields in Modern Jihad”, CTC Sentinel, 15 May 2008. For a helpful discussion of al-Naji’s and al-Muhajir’s contributions to IS, see Hanieh and Rumman, The “Islamic State” Organization, op. cit.Hide Footnote  These writers advocated violence not only to protect a marginalised Sunni community, but also to remake society and give direction to a generation crushed under decades of oppressive governance and an unfriendly global order.

At least in its propaganda, IS aims to extinguish the “grey zone”, what it calls any space for neutrality between the caliphate and heretical regimes and Western powers. Muslims must fight for the former or be seen as non-believers, part of the latter. Local IS commanders have shown occasional pragmatism in Iraq and Syria and are likely to do so elsewhere, given that eradicating all other forms of Sunni opposition would be impossible. Still, IS fights a simultaneous war on all fronts: against primary enemies, Iranian proxies and the Shia; other Sunni rebels; Sunni powers it sees as Western stooges; Russians as infidel supporters of Assad and Iran; Western powers and so forth. It has woven together sectarian, revolutionary and anti-imperialist strands of jihadist thinking.

Al-Qaeda and its affiliates have responded differently to the popular upheaval. AQAP and al-Nusra may fight in sectarian wars and target Huthis and Alawites; and al-Qaeda is hardly shy about killing civilians or cooperating, in Pakistan for example, with deeply sectarian allies.^[fn]For al-Qaeda’s cooperation with Pakistani sectarian groups, see Crisis Group Report, Pakistan: The Militant Jihadi Challenge, op. cit.; Moeed Yusuf (ed.), Pakistan’s Counterterrorism Challenge (Washington DC, 2014); and Khaled Ahmed, Sectarian War: Pakistan’s Links to Sunni-Shia Violence and its Links to the Middle East (Karachi, 2012).Hide Footnote  But Zawahiri, like bin Laden before him, tends to maintain that making enemies of Shia as a whole and alienating Muslim public opinion through indiscriminate killing work against the main goals of attacking the West, driving it out of the Muslim world and overthrowing tyrannical local regimes.^[fn]“English Translation of Ayman al-Zawahiri’s letter to Abu Musab al-Zarqawi”, The Weekly Standard, 11 October 2005; and Brian Dodwell, “The Abbottabad Documents: The Quiet Ascent of Adam Gadahn”, CTC Sentinel, 22 May 2012; and The Osama Bin Laden Files: Letters and Documents Discovered by SEAL Team Six during their Raid on Bin Laden’s Compound, The Combating Terrorism Center (West Point, 2012).Hide Footnote  Characteristic was the celebration by some al-Qaeda supporters at the “discretion” shown by not spilling Muslim blood during AQIM’s November 2015 Bamako attack, as compared with IS’s indiscriminate attacks in Paris the previous week (in fact some non-Western casualties in Bamako and Ouagadougou were Muslim).^[fn]Liam Stack, “al Qaeda supporters celebrate Mali attack on Twitter”, The New York Times, 20 November 2015; also “Pro-AQ Jihadists Celebrate Bamako Attack, Contrast to IS Tactics”, SITE Intelligence Group Jihadist Threat, 20 November 2015.Hide Footnote  In Syria and Yemen, al-Qaeda also worked with militias backed by powers it claims it wants to overthrow and, as described, has received on occasion help from states.^[fn]Crisis Group Reports, Rigged Cars and Barrel Bombs; and Yemen: Is Peace Possible?, op. cit.Hide Footnote

The debate between “near” and “far” enemies has also shifted. Though al-Qaeda pioneered the switch from “heretical” states in the region to the West in the late 1990s, most affiliates are now fighting locally. Conversely, IS initially, as AQI and in its current guise, focused mostly on Iraq, Syria and other parts of the Muslim world. It appears to have moved over the past year, however, from simply encouraging “lone wolf” attacks elsewhere to actively dedicating resources for blows against the West – like, for example, its coordination of the Paris killing spree. This is partly to sow unrest within Western society and provoke a backlash against the Muslim diaspora that would generate additional supporters; and partly to cement its position as leader of the jihadist movement and bin Laden’s true successor.^[fn]See, for example, IS material summary compiled by Aaron Zelin.Hide Footnote  Its focus on Muslim populations in Europe and exploitation of the internet, with multilingual propaganda, have also, to a degree, collapsed the distance between near and far.

Attitudes toward the nation-state system are, in some conflicts, perhaps a variable in determining who can be engaged diplomatically. At their top level, IS and al-Qaeda have transnational goals. Despite its primary identity as an Iraqi insurgency, IS – at least according to its own statements – wants to provoke a war across the Muslim world as a step to expanding its caliphate; Zawahiri and al-Qaeda affiliate leaders view their local struggles as fronts in a wider transnational jihad.

Other movements, including some self-identified as jihadist, espouse national goals: ousting an illegitimate government, fighting foreign “occupiers” or establishing their conception of Sharia (Islamic law). The Taliban has many elements, but its core is nationalist, if mostly Pashtun, dedicated to recovering its emirate in Afghanistan and expelling Western forces. Ahrar Al-Sham repeatedly says it wants to change Syria’s political order, not remake the Muslim world, despite a senior al-Qaeda operative, Abu Khalid al-Suri, being among its founders and its tight battlefield coordination with Jabhat al-Nusra.^[fn]Ahrar al-Sham’s Abu Yazan: “It’s our country and our revolution”, Abu al-Jamajem. Since 2014, Ahrar al-Sham has also rejected the Salafi-jihadist label, ibid.
Hide Footnote  It openly takes Turkish support, and its leaders declare willingness to work with the West to oust Assad. Ansar Dine, which aligned with al-Qaeda in Mali in 2012-2013, and some Ansar al-Sharia factions in Libya similarly appear to aspire to Islamic rule within existing borders.^[fn]Crisis Group interviews, Libyan politicians and members of Ansar Sharia, Tripoli and Benghazi, 2014.Hide Footnote  

Even among movements with nationalist goals, few accept political or religious pluralism. The Taliban leadership aspires to a government under the authority of a divinely-appointed emirate. While it appears open to compromise – and in the past some of its leaders have been willing to accept other forms of government – it still insists that any new political order must be based on its version of Sharia; it would have to perform ideological gymnastics to justify power sharing and a government based on the people’s will.^[fn]See, for example, Michael Semple, “Rhetoric, Ideology and Organisational Structure of the Taliban Movement”, U.S. Institute for Peace, 2014.Hide Footnote  Ahrar al-Sham leaders, on the other hand, concur since at least mid-2014 with other rebels that Syrians should determine the country’s system of governance and select its leaders. They remain Salafi, define vaguely if at all how popular will would be determined and avoid the word “democracy”, but they agree on the principle, and al-Qaeda ideologues attack them for it.^[fn]Crisis Group interviews and communications, current and former senior Ahrar al-Sham officials, Istanbul, March 2014-December 2015; also Crisis Group Report, Rigged Cars and Barrel Bombs, op. cit. For attacks on Ahrar al-Sham from al-Qaeda-linked ideologues, see articles by senior Jab­hat al-Nusra figure Abu Firas al-Shami at and AboFiras.Hide Footnote

Identifying groups’ goals can, of course, be difficult. Official messaging may not reflect positions of the rank-and-file or even the leadership: some are clearly committed to radical ideals; others express them to curry favour with Gulf-based donors or may feign pragmatism to win state backing. To a degree, identities are defined as much by strategy, tactics and sources of funding and support as by longer-term goals, given the often remote nature of those goals. But jihadists’ increasing prominence in war zones and the speed with which some mutate make it vital to monitor ideology, between and, to the extent possible, within movements. What they want, particularly related to the state system, their openness to sharing power and tolerance toward other sects or religious groups, bears on policy. Any sign of evolution or possibility of influencing or splitting them along these lines may open new ways to diminish their threat.

Controlling territory, among the thorniest challenges for any insurgency, has proven especially hard for jihadists. Their harsh, literal implementation of Sharia has rarely inspired much support. More importantly, most have proven inept rulers. But given the conditions of extreme violence or state collapse that enable them to seize territory, communities may find them better than the alternatives or have little choice but to acquiesce. Also, some movements show signs of learning to govern in ways that avoid fully alienating those under their control.

In recent history, few radical Islamist movements had held territory before 2011. The Taliban, first as it advanced north and then as the government of most of Afghanistan in the mid-1990s, initially brought some basic law and order, but its puritanical mores, economic mismanagement, sporadic attempts to curb poppy cultivation, forced conscription and war-time atrocities soon alienated many, particularly in cities and towns.^[fn]Crisis Group Report, Talking About Talks, op. cit.; and Crisis Group Asia Report N°256, Afghanistan’s Insurgency after the Transition, 12 May 2014.Hide Footnote  Its leaders’ poor performance left them isolated after their rout in 2001 by Northern Alliance and other U.S.-backed forces.

It was, in turn, mostly the failures of the new government and the U.S., its primary sponsor, particularly allowing local powerbrokers to manipulate the U.S. war on terror to abuse or eliminate rivals, that enabled the Taliban, excluded from the new political order and whose leaders had sheltered across the border in Pakistan, to re-emerge as an insurgency, rekindling ties and offering protection. Its courts, often mobile, dispense fast, predictable and enforced, if harsh, justice that by most accounts is reasonably popular, at least outside cities. Its “shadow” state administration operates across much of rural Afghanistan, though is dedicated more to the military campaign than service delivery. In some places, insurgents allow Kabul’s education and health ministries into areas under their control to run schools and clinics, sometimes even shaping the former’s curricula.^[fn]For example, Nick Walsh, “Taliban tightens grip on Afghan schools”, CNN, 22 May 2012.Hide Footnote

Similarly, al-Shabaab’s rule at the height of its territorial control (2007-2011) was austere but brought some initial order. Some villagers at first welcomed schools for Quranic education, basic medical services, reasonably predictable tolls on roads, regular, safe market days and local dispute resolution. As an insurgency, al-Shabaab now combines unpopular violence with pragmatism and political acuity. It deals ruthlessly with potential rivals, while mediating between clans or backing weaker ones against rivals and avoiding too close an association with any.^[fn]Over recent months al-Shabaab has lost locations mostly where clans have closed ranks, including recently in Middle Shabelle, apparently in reaction to its tax demands for drought victims elsewhere in Somalia. It does best amid outright rivalry between clans or where clans feel frozen out of power. Abdul Khalif and Cedric Barnes, “Why is Al-Shabaab Still a Potent Threat?”, Crisis Group, 11 February 2016.Hide Footnote  Both the Taliban and al-Shabaab have permitted, if uneasily and by no means universally, aid groups to work in areas they control, which involves engagement to agree on terms.^[fn]Ashley Jackson, “Negotiating Perceptions: Al-Shabaab and Taliban View of Aid Agencies”, Overseas Development Institute, August 2014; also, Crisis Group interviews, aid workers after the Somali famine, January-March 2011.Hide Footnote

Neither movement is popular. Many villages are caught between their harsh rule and violence and the predation of local government-aligned strongmen; for many, survival hinges on working with whomever holds sway locally. Both, however, deliver some basic public goods and exploit local grievances, conflicts and tribal or clan relations to win support, while playing on intra-tribal or clan tensions between traditional authorities and those marginalised, particularly younger men. They exert their authority in captured territory through an often carefully calibrated mix of coercion and co-option.

Since 2011, more jihadists have seized territory. IS’s rule is difficult to assess given the dearth of information and that it varies considerably across Iraq and Syria, but it is far more sophisticated than that of Zarqawi’s AQI a decade earlier. Its violence raises the cost of dissent, while its leaders have forged closer ties to parts of society. More importantly, in contrast to any past jihadist movement, it appears able to run a state, its recent setbacks notwithstanding. Unlike the Taliban and al-Shabaab, it inherited a largely functioning infrastructure and civil service and has co-opted parts of the local bureaucracy. In most cities and towns, sanitation, rubbish collection, schools and clinics still work. Its law enforcement may be draconian but reportedly is not yet corrupt; its internal revenue generation is often extortive but at least so far appears sustainable. It has, like other movements, emphasised the quick and enforced resolution of often longstanding disputes.^[fn]Crisis Group telephone interviews, residents of IS-controlled areas, October 2015. Khales Joumah, “Mean but Clean: Extremists Fix Roads, Make Mosul a Nicer Place”, Niqash, 14 May 2015.Hide Footnote

The evolution in AQAP’s governance in Yemen is as striking. During the 2011 revolution, it overran part of Abyan governorate, including its capital Zinjibar. Army reinforcements took time to deploy – the army split during the revolution, some factions siding with protesters – but then ousted militants swiftly, with local support. This led Nasir al-Wuhayshi, AQAP’s leader at the time and al-Qaeda’s general manager (in effect its number two after Zawahiri), to write to al-Qaeda affiliate leaders elsewhere, notably in Mali, to share his experiences and urge increasing sensitivity to local opinion.^[fn]See “First letter”, op. cit. Al-Wuhayshi urged pragmatism to avoiding alienating inhabitants of captured areas, advice that AQIM’s leader, Abdelmalik Droukdel, appears to have heeded, though his commanders followed haphazardly. Their rule again varied across different parts of the country – overall less brutal than Boko Haram, IS or even AQIM’s own splinter, the Movement for Monotheism and Jihad in West Africa (MUJAO), but still harsh, with unpopular music and smoking bans, corporal punishment and destruction of Sufi shrines and cultural artefacts. Even so, numerous reports suggest some villagers welcomed AQIM’s and to a degree MUJAO’s ability to enforce their dispute and conflict resolution. Crisis Group interviews, former MUJAO member, Bamako, June 2015; also, residents, Gao, August 2013.Hide Footnote

When AQAP seized parts of Hadramout governorate, including Mukalla, as Yemen’s war escalated in 2015, it appointed a local council headed by prominent elders, including Salafis but not al-Qaeda members. New religious courts are viewed by many locals as fair and swift in contrast to the corrupt and slow official system, which in any case has collapsed. Civil servants are paid, and the city has not suffered the chaos of elsewhere, partly because it is among the few areas not hit by Saudi-coalition bombs. AQAP looted local banks, but the council generates revenue mostly through taxes on goods, particularly fuel. Shipping companies continue to trade with the al-Qaeda controlled town; though wary of docking in its port, they stop in international waters and smaller boats ferry in goods, including gas.^[fn]Crisis Group observations and interviews, Mukalla, October 2015.Hide Footnote

AQAP’s fighters make locals uneasy but have reduced petty crime. Its leaders meet representatives of Western aid organisations to coordinate relief, as jihadist leaders did in northern Mali in 2012.^[fn]Crisis Group telephone interview, humanitarian representatives, November 2015.Hide Footnote  It has destroyed several Sufi shrines and mausoleums in Hadramout but interfered less with dress norms and has not forced people to pray or pay religious taxes. Selling qat is forbidden, but music and TV are not. It has also responded differently to dissent. In a town just east of Mukalla, after demonstrations against AQAP’s assassinations of religious scholars and its fighters’ behaviour at checkpoints, the local commander met with town leaders and agreed to withdraw most of his men to a nearby military camp.^[fn]Crisis Group observations and interviews, Mukalla, October 2015.Hide Footnote

AQAP’s and IS’s evolving governance has certainly not been replicated by all extremists. Boko Haram claims to want to bring Islamic rule to the Lake Chad Basin but pillages captured areas of northern Nigeria, bringing not even the blend of coercion and co-option deployed by some others, let alone any pretence of Sharia.^[fn]Crisis Group Report, Boko Haram, op. cit. Also see Alex Thurston, “‘The Disease is Unbelief’: Boko Haram’s Religious and Political Worldview”, The Brookings Project on U.S. Relations with the Islamic World Analysis Paper, no. 22, January 2016.Hide Footnote Although many in northern Nigeria distrust the state, identify with Boko Haram’s criticisms of its abuses and aspire to a greater governance role for Islam, the movement’s brutality in towns it seized, the havoc it wreaked and the kidnapping of schoolgirls have stripped it of popular support. Its tactics resemble more those of the LRA or other militias plaguing the African Great Lakes than IS or al-Qaeda affiliates.^[fn]EJ Hogendoorn, “Boko Haram’s Evolution: How it got this far and how to stop it”, Open Canada, 21 October 2015.Hide Footnote  The disparate tribal militias loosely aligned under the Pakistani Taliban banner perform little better in areas they sporadically seize; any initial support evaporates quickly in the face of their cruelty.^[fn]Crisis Group Report, Pakistan: Countering Militancy in PATA, op. cit. Also Abubakr Siddique, The Pashtun Question: The Unresolved Key to the Future of Afghanistan and Pakistan (London, 2014).Hide Footnote

Nor would even the more adept groups be credible alternatives in reasonably functioning states. Conditions must be awful before communities accept them or are forced to do so to survive – illustrating again how war and state collapse create settings in which jihadists thrive. But where their governance is evolving, there are clearly policy implications. It has been common for extremists to win some initial support by bringing basic law and order – especially predictable and enforced dispute resolution – but for that to dissipate fast as their violence becomes arbitrary and their punishments draconian, as they ban music and empower criminals, as services collapse and rubbish piles up. Will that model hold? Can groups be contained geographically in expectation that over time inhabitants will revolt or support their ouster? Or will they hold territory and deliver services in a way that deepens their ties to communities, furthers their agenda and safeguards a haven from which to launch attacks?

It is too early to say, but more such movements hold land now than ever before, many of the crises that permit them to do so show little sign of abating, and some are learning to calibrate their approach toward those they rule.

The extending reach of IS and al-Qaeda-linked groups poses thorny policy dilemmas, especially where they hold territory, but also in places facing an increased risk of terrorist attacks. World leaders ramping up their rhetoric against IS must learn from mistakes, while redoubling efforts to understand evolving dynamics.

Many Western politicians overstate the threat. This is, to a degree, understandable: jihadist attacks target their citizens. But even IS poses no major, let alone existential, peril to their countries. Beyond the human misery it already causes, the gravest risk is that its violence provokes reactions – xenophobia, curtailing of civil liberties, selective policing at home or military adventurism abroad – that aggravate the conditions that enabled its rise, open new opportunities for it in the Muslim world and facilitate recruitment in the West.

Over the past few years, however, jihadist movements have become more powerful than ever before. Standard counter-terrorism toolkits – designations, financial sanctions, travel bans, targeted killings and special forces’ operations, for example – are insufficient against movements that control cities, towns and supply lines, provide public goods, generate revenue locally and have tens of thousands of fighters. Some of their leaders’ ideology and aspirations complicate engaging them politically, but there is scant modern precedent for defeating an entrenched insurgent movement through military means alone. Sri Lanka’s approach to the Tamils, for example, even leaving aside its law-of-war abuses and horrific human cost, would not work in much of the belt from West Africa to South Asia, given porous borders, wars’ often proxy nature and states that have collapsed or have limited writ in hinterlands. Similarly, replicating in Syria Russia’s scorched-earth tactics in Chechnya would more likely bolster the ranks of IS than defeat it; in any case, Russian airstrikes have primarily targeted other rebels, not IS. Elsewhere, military gains have often merely relocated the problem.^[fn]Operations against Boko Haram, for example, have dispersed its fighters across the border. Algeria’s campaign against the remnants of the GSPC, which became AQIM, pushed militants into the Sahel; French operations in turn appear to have shifted many to Libya. The U.S. ousting of the Taliban largely displaced its leadership and much of the al-Qaeda top leadership to Pakistan. Russian operations in the North Caucasus have partly caused many jihadists to go to the Levant.Hide Footnote

What makes the fourth wave so perilous, however, is less the groups’ strength than the geopolitical upheaval that they profit from.^[fn]See also Darryl Li, “A Jihadist Anti-Primer”, Middle East Research and Information Project 276, Fall 2015.Hide Footnote  First, decisively reversing jihadist gains often requires ending the wars they fight in. In Yemen, without a peace deal between the Huthis and loyalists of former President Saleh on the one hand and forces aligned to the Saudi-led coalition, prospects of ousting al-Qaeda from the territories it controls are bleak. The longer it brings a semblance of order amid chaos, the stronger it will grow. Even with a peace deal, it may have deepened local ties to such a degree and Yemeni security forces may have become so debilitated that they will struggle to oust violent jihadists as they did in 2012. A deal would further fracture the anti-Huthi alliance of which AQAP is part, though what the net effect of that on the movement would be is unclear: it might simply reshuffle alliances and mark the start of the war’s next phase.

Similarly, reversing jihadist gains in Libya will depend on resolving rivalries between other local forces and persuading them to collaborate against IS. It will depend, too, on giving areas associated with the Qadhafi regime, which are most vulnerable to IS recruitment, a stronger position in the national fabric and probably also self-defence opportunities.^[fn]For how to do this, see Issandr El Amrani, “How much of Libya does the Islamic State control?”, Foreign Policy (online), 18 February 2016.Hide Footnote  A bombing campaign could hamper IS operations, especially near oil facilities, and degrade its materiel; in Libya such targeted strikes may make sense. But so long as rivalries between its enemies persist, it will continue to hold the area around Sirte and may extend further east. If the U.S. or others decide – mistakenly – to press ahead with heavier bombing, better they do so without demanding that the fledging, contested unity government invite or endorse foreign military action, notwithstanding the legal obstacles that would create, lest that further diminish its credibility. More can also be done to engage with diverse Libyan security actors – and promote contact between them – to both build support for the political process and find potential partners against IS.

Secondly, much as smaller groups profit from the Libya and Yemen wars, so IS profits in Iraq and Syria from its enemies’ regional confrontation. The best starting point against it would be a grand bargain to dial back the Iran-Saudi rivalry that drives both Sunni and Shia radicalism, is a principal obstacle to ending crises across the region and poses a graver threat to global stability than jihadists. Prospects appear bleak, but urging an entente should be as vital a priority as fighting IS. Without it there is risk of mounting confrontation, with Syria its epicentre and both sides describing their violence as counter-terrorism, that pits an Iran-Baghdad-Damascus-Hizbollah axis, with Russia joining opportunistically, against the mostly Sunni powers in the new Saudi alliance, backed uneasily in the West. Efforts to narrow other fault lines that open space for jihadists, – between, for example, conservative Arab regimes and the Muslim Brotherhood, Turkey and Kurdish armed groups, now Turkey and Russia and India and Pakistan, should also be redoubled – even if rapprochement seems remote.

Thirdly, there is the nature of many affected states. The largest movements have filled vacuums left by state collapse in Iraq, Libya, Somalia, Syria and Yemen and, to a degree, Afghanistan. Jihadists thrive, too, in parts of more capable states like Egypt, Mali, Pakistan, Russia and those around the Lake Chad Basin where the government’s writ has traditionally been limited. In many vulnerable states and those at war, government behaviour is a main source of grievances driving support for jihadist movements or provoking crises they profit from. Capable, resilient states should be the foundation of efforts against extremism. However, the outlook for recovery, reform and regeneration, particularly in the Arab world, is gloomy. Little suggests that governments largely responsible for the fourth wave are ready to adapt in ways needed to counter it.

Fourthly, leaders in many of the countries most affected simply view the threat differently than their Western counterparts. Some, as described, are more focused on regional rivalries or may fear that action against jihadists would anger religious establishments. Others see opposition movements as graver threats to their rule or jihadists as useful leverage with the West and a pretext for repressing other rivals.^[fn]Former Yemeni President Saleh, for example, co-opted mujahidin returning from Afghanistan as he battled in the south for power in newly-unified Yemen. He then gave government posts to some, while sidelining others, even if often retaining ties to them through intelligence services. Throughout the last two decades of his rule, he used the jihadist threat to win Western support, receiving training and weapons to fight al-Qaeda. Despite sporadic crackdowns, usually under U.S. pressure – particularly after the attack on the naval ship USS Cole in the Aden harbour (2000) and 9/11 the next year – his dealings with Sunni radicals, al-Qaeda in particular, tended to be guided by his divide-and-rule approach to politics in general.Hide Footnote  The variance in jihadists’ targets – Western powers, local regimes and Shia – means that governments in affected areas confront dilemmas different from those facing Western powers, which are threatened, for the most part, from afar: cracking down can stir a hornets’ nest, shift jihadists’ anger at foreign powers homewards and increase local terrorism. Some states, notably Pakistan, have badly miscalculated this balancing act, a mistake Turkey may have replicated in Syria. But contrasting incentives mean anti-jihadist alliances tend to be flimsy, and the U.S. and European focus on the threat to the West, while understandable, can have a distortive local impact.

There is, of course, no single solution. The diversity of groups and the wars they fight in mean that any approach must be developed case-by-case, with accurate diagnoses of the relevant movement’s strength, goals and relationship to communities, of those communities’ grievances, the motives of governments, militaries and outside powers and of whether a credible force exists that can act without making matters worse and is not distracted by rivals.

Options against groups like those that captured northern Mali, for example, – that initially enjoyed shallow support, fled when confronted by a serious force and some of which appear to have had transnational goals – differ from those against the Afghan Taliban, which is firmly entrenched in the Pashtun heartlands, largely nationalist, enjoys at least intelligence support and safe havens in Pakistan and has weathered U.S. troop numbers in the six figures. Tackling unpopular Boko Haram, which can hide in the vast desert and bush around Lake Chad but against which regional governments are now reasonably united, requires a very different strategy than in Libya against militants in Benghazi and Derna that other revolutionary brigades view as allies and many residents more as wayward youth than hardened extremists. Understanding local dynamics is critical. Each movement should be tackled individually, not as a global phenomenon.

That said, many pose similar dilemmas. First is on the use of force. Where jiha­dists have seized territory, does military action to oust them make sense; if so how and by whom; and, most importantly, what local administration follows? Secondly, does the targeted killing of leaders help reduce the threat, either locally or to the West? Thirdly, what engagement is feasible, what ends should it serve and what risks does it entail? And lastly, as jihadists’ ability to profit from war and state collapse brings new urgency to efforts to prevent crises that may open opportunities for them, what role can the emerging agenda of countering violent extremism (CVE) play in shoring up states’ resilience?

Part of IS’s allure rests on its momentum, its ability, in its own words, to “remain and expand”, to portray itself as having assumed the mantle of Sunni leadership across the region. The longer it holds a swathe of Iraq and Syria, the stronger its aura of invincibility and the greater its appeal will be. Ousting it or at least putting it on the back foot should thus be a priority.

But IS also thrives in chaos. Woven within its narrative are both its inexorable advance and a strand of apocalyptic thinking that envisages an eventual final battle with Western forces. Most importantly, it is a product of Sunnis’ suffering and, in Iraq, their struggle, after Saddam Hussein’s ouster, to forge a new political identity. Reclaiming territory is vital, but doing so at the cost of further alienating Sunnis – having already lost them in the aftermath of the 2003 invasion and then by a betrayal of the Awakening – would be counterproductive. The lynchpin of any approach and that must shape any use of force has to be a political strategy to win over the communities in which IS is embedded.

Bombs alone will not do the job. Disrupting IS’s service delivery risks harming communities as much as jihadists; history shows that affected communities tend to rally behind local oppressors against external attackers. Pounding Raqqa after the Paris attacks had no strategic value; further flattening and driving more residents from homes risks playing into the hands of extremists as much as weakening them. Airstrikes, even if intensified, only work if they reinforce allies on the ground, which raises the question of which forces can lead offensives.

Even when the U.S. was deployed in Iraq in large numbers – a height of some 160,000 troops during the Surge – it was effective against IS only in partnership with local forces. During the Awakening, the U.S. backstopped the tribal “Sons of Iraq” by giving structure to their formations, providing rudimentary training, reinforcing them when necessary, paying salaries and running interference with the Iraqi state.^[fn]Peter R. Mansour, Surge: My Journey with General David Petraeus and the Remaking of the Iraq War (New Haven and London, 2013), p. 140.Hide Footnote

Replicating that today would be hard, for many reasons. Even hawks in the U.S. have little appetite for a massive redeployment and with good reason. Committing larger numbers of Western (or Russian) ground forces would reinforce IS’s narrative of infidel crusaders, accelerate its intake of fighters, foreign and local, and play to its apocalyptic narrative. Even a more limited Western deployment, as some recommend – in numbers ranging up to 25,000, including military advisers, Special Forces and Quick Reaction Forces – to back local and regional elements would pose enormous hazards for an uncertain return.^[fn]U.S. Senators John McCain and Lindsey Graham, among others, call for U.S. ground troops in Syria. Stephen Kalin, “U.S. senators call for 20,000 troops in Syria and Iraq”, Reuters, 29 November 2015. Kimberly Kagan, Fredrick Kagan and Jessica D. Lewis estimated the needs of a first phase alone at 25,000. “A Strategy to Defeat the Islamic State”, Institute for the Study of War, September 2014.Hide Footnote  Russian involvement in Syria means a risk of global escalation; even were that danger somehow eliminated, the U.S. has no state ally to work with in Syria and would risk getting sucked into fighting simultaneously IS, al-Qaeda, its rebel allies and regime loyalists.

In Iraq, the U.S. has to worry less about Russian involvement, but Iran and Shia politics in Baghdad could prove no less daunting. Even during its eight-year occupation, the U.S. failed to convert the eventual military success of the Awakening into a political one by brokering a deal between its Sunni allies and the Iraqi state – and that was when it had more troops on the ground than anyone contemplates today, Iran’s influence was weaker, and Shia militias were less active.

Marshalling local and regional forces for the U.S. to back would also be challenging. Other rebels and their al-Qaeda allies have done the most in Syria against IS, repelling it from the north west, but they cannot fight it successfully in the east while hemmed in by the regime and pounded by Russian airstrikes. So long as the war between regime and rebels rages, training the latter to fight only jihadists has no chance, as shown by the dismal results of U.S. attempts to do so at a time when rebel prospects were much less bleak than today.^[fn]Many trained at significant cost were quickly captured or killed by al-Nusra and their weapons confiscated. Noah Bonsey, “Turkey and the U.S. in Syria: Time for Some Hard Choices”, Crisis Group blog, 10 August 2015.Hide Footnote  In Iraq, Kurdish and Shia militias are among the most capable, but neither appears keen to fight for the Sunni heartlands or has local support there; to the contrary, their involvement would aggravate tensions with local communities, potentially driving them further into IS’s arms. The same is true of the Kurdish YPG in Syria. Arming militias also further degrades the Iraqi state.

Most important, while Baghdad and the U.S. have in places raised Sunni allies against IS, another uprising like the Awakening looks remote. Tribes joined against AQI only after being convinced that the U.S. would be a reliable ally. Their bitter experience in the aftermath means that any foreign force would face an uphill battle to win their trust. Their suffering at the hands of Baghdad and IS’s infiltration of local social structures and crushing of its opponents have heightened the local population’s belief that jihadists will be around long after foreign forces eventually leave. Unless Western states make an open-ended commitment of troops at far higher levels than seem possible, it will be hard to win back former allies.

With a U.S. re-invasion off the table, the campaign against IS has been conducted on a more limited scale. Recent offensives have involved warnings to civilians to leave towns and massive airstrikes to oust militants, followed by the Iraqi government, in cooperation with para-state forces, advancing a patchwork of small units – including counter-terrorism forces, retrained Sunni local and federal police and Kurdish forces – to retake territory. Former Sunni political leaders, displaced by IS, are waiting out the fighting in Baghdad and elsewhere, hoping to recover their legitimacy and reestablish their authority by rebuilding the infrastructure the offensive against IS destroys. The Iraqi government, with the support of the U.S., Saudi Arabia and Turkey, is pushing for decentralisation, with a Sunni-majority enclave centred in Anbar province and extending to Mosul, whence Gulf countries and Turkey would support the local Sunni leadership and assist with reconstruction funds.^[fn]“The U.S. will fully support the plan endorsed by the [Iraqi] Council of Ministers on May 19 for the liberation of Anbar, as well as the Iraqi government’s priority of de-centralization to empower local communities in line with the Iraqi Constitution. This ‘functional federalism’ effort being pursued by the Iraqi government is integral to ensuring that ISIL – once defeated – can never again return to Iraqi soil”. Statement by the White House press secretary on “Additional U.S. Steps in the Counter ISIL-Effort”, 10 June 2015.Hide Footnote

This strategy is unlikely to succeed. Iran and, to a degree, Russia oppose any devolution that could empower Sunnis. Decentralisation would need to be defined along administrative rather than sectarian lines and relate to all provinces and districts, not just Sunni areas, to undercut Iranian resistance, provide flexibility to other provinces that resist Baghdad’s tight control and, hopefully, start to reverse the communal logic of Iraqi governance.^[fn]Divisions within the Shia and Kurdish parties and loss of support among their constituents could lead to more fragmentation but might also help break community-based politics and force the formation of cross-communitarian alliances.Hide Footnote

More importantly, the anti-IS strategy, which largely concerns the mechanics of governance, does not in itself address the Sunni community’s anomie, which, alongside its marginalisation, IS feeds off. Renovating the structure of governance will not necessarily imbue it with substance. The key to broad Sunni re-engagement is narrowing the gap between the Sunni leadership and its constituents, particularly young people. This is especially so if non-ideological supporters of IS are to be prised from its ideologically motivated core, which would not disappear even if ousted from towns. Massive destruction and backing largely discredited leaders who abandoned Sunni areas after the Awakening would be a weak base on which to build a new Sunni political project.

Attempting to replicate in Mosul the 2015 capture of Anbar province’s Tikrit and Ramadi, which all but destroyed the cities in the name of saving them, will be far riskier. The Sunni character of Anbar is undisputed, but the longstanding regional competition over the multi-ethnic and strategically located Mosul will complicate stabilising the city in the wake of any campaign, which itself will be more complex than any previous ones against IS. Turkey, the Iraqi government, Iran and Shia militias, and the Kurds (including both the Kurdish Democratic Party and PKK, themselves at odds with one another) are all determined to secure their own interests and, perhaps more important, deny their rivals the same.

What, then, is the alternative? If territory cannot be usefully won because of the difficulty of also winning over its inhabitants and creating conditions in which Sunnis can build a viable political agenda, a better bet is patience and containment, preventing IS’s advance, avoiding action that plays into its hands, redoubling efforts to cut its funding – albeit difficult now that much comes from taxation and extortion – and other measures to degrade its relations with those under its control.

This does not mean leaving those under IS’s rule to their fate, but slowing the battle tempo to give political strategy a chance to catch up and allow for more outreach before offensives. An essential first step would be to secure local communities’ trust, as the Awakening did slowly and methodically, not leaving it for later. This starts by limiting the bombing campaign to vital targets and imminent threats, and preventing IS expansion, while squeezing it in every other way so as to erode the aura of invincibility that has convinced communities to cooperate with it and attracted new recruits from around the world. Circumstances are different, of course, from a decade ago, when the Sons of Iraq switched sides: IS is more potent than AQI; the Iraqi government is less amenable to Sunni aspirations; the U.S. cannot provide the same military backup nor bridge gaps with that government; and other actors, such as Kurdish and Shia militias, have a greater presence and ability to defend their interests. The principle, however, should be the same: that trust of residents is a more important asset than territory.

Containment, or slowing the pace of the campaign, would be, of course, a significant gamble, given IS’s ability to disrupt and attack elsewhere, in the West, but first and foremost in Muslim countries. It would involve risks that either Iran assumes the lead in combatting it and does so in a counterproductive manner, or that IS endures and its rule normalises; and political costs, including domestically, that the U.S. and other countries would pay for being seen by some as irresolute or impotent, even if their restraint is sensible. But the track record in Iraq and around the region provides compelling evidence that without a strategy that includes a convincing, locally accepted political alternative for the day after IS’s defeat, military escalation is not the answer.

Options against IS are especially poor, but other groups pose similar dilemmas. Any calculation rests partly on a group’s potency and local ties, as discussed above, but partly, too, on what forces can take it on. Even reasonably capable states’ armies are often not built for internal threats.

Early Pakistani operations against militants hosting al-Qaeda in the tribal areas, for example, launched mostly at U.S. urging in 2002, were disastrous. The army stirred up resistance, was repeatedly forced to retreat and struck deals ceding militants more local authority.^[fn]Crisis Group Reports, Pakistan: Countering Militancy in FATA, op. cit.Hide Footnote  After waves of offensives and with military elites more resolute, at least against some militants, the army can now clear and hold some areas, though operations still exact tremendous civilian tolls.^[fn]Crisis Group Asia Report N°255, Policing Urban Violence in Pakistan, 23 January 2014.Hide Footnote  Some Pakistani Taliban leaders, however, have crossed to Afghanistan, while militants dispersed across Pakistan have escalated attacks ranging from Peshawar’s Badaber base to Charsadda’s Bacha Khan University, to military targets in Quetta. Unless the tribal areas are brought under regular constitutional rule, which would require reforms that the security establishment appears reluctant to accept because these areas traditionally host the training infrastructure for their militant proxies, the army’s occupation is at best a stopgap.

Nigeria’s initial response to Boko Haram was similarly clumsy, lurching from denial to brutal crackdowns, to military operations, including air assaults that killed many civilians.^[fn]See “Nigeria: At Least 1,000 Civilians Dead Since January”, Human Rights Watch, 26 March 2015; and CRF.Hide Footnote  Many youths were executed or imprisoned without trial.^[fn]See, for example, “Stars on Their Shoulders. Blood on Their Hands. War Crimes Committed by the Nigerian Military”, Amnesty International, 3 June 2015.Hide Footnote  Troops from outside the north and without knowledge of local customs or languages were distrusted. Corruption, insufficient logistics and poor leadership meant desertions were rampant, mutinies common.^[fn]Crisis Group Report, The Boko Haram Insurgency, op. cit.; also see Hilary Matfess, “Don’t repeat mistakes against Boko Haram in Cameroon”, Al Jazeera America, 2 October 2015.Hide Footnote  Even now, more competent Nigerian and Chadian operations that have reversed Boko Haram’s gains tend to be heavy-handed and indiscriminate. They may not drive communities to support Boko Haram, but they make them less likely to offer government cooperation, as militants hide in more remote areas. As in Pakistan’s tribal areas, Nigeria and its neighbours must engage more benevolently in their peripheries, particularly around Lake Chad. Egypt’s Sinai operations against IS, again a location where the state’s writ is weak, risk similar problems, as collateral damage mounts, and the population finds itself living under increasingly arduous conditions with little government relief.

Working through auxiliaries is potentially more problematic still. Nigeria’s and Pakistan’s arming of militias against Boko Haram and tribal extremists has, perhaps, yielded occasional short-term gains but causes problems over time.^[fn]On Nigeria, see, Haruna Umar, “Nigeria’s army accuses 2 soldiers of arming Boko Haram”, Associated Press, 10 February 2016; and Michelle Faul, “Report: 10 Generals guilty of arming Boko Haram”, The Huffington Post, 3 June 2014. On Pakistan, see, Crisis Group Report, Countering Militancy in FATA, op. cit.Hide Footnote  Arming anti-Taliban militias in Afghanistan has often entrenched predatory local forces and exclusionary patronage networks that drive support for the insurgency and fuel local disputes. Such dynamics almost certainly facilitated the Taliban’s encroachment around Kunduz in 2015.^[fn]See, for example, Deedee Derksen, “The Politics of Disarmament and Rearmament in Afghanistan”, U.S. Institute of Peace, 20 May 2015.Hide Footnote  In Iraq, Libya, Syria and Yemen, militias may be the best-equipped forces, but backing them contributes to the disintegration of state structures, escalating arms races and radicalisation on all sides. Non-state forces may still be required; the Iraqi army needs Sunni allies against IS as it did against AQI. But policymakers need to be more cognisant of the risks and factor in militias’ relations with communities as much as their keenness to fight.

Foreign boots on the ground involve other challenges. There have been some successes: the French Serval operation in Mali quickly ousted al-Qaeda-linked groups from northern towns, creating space for an eventual deal between Tuareg factions and the government. Even there, though, foreign operations failed to eradicate movements that melted into the Sahel, and the peace deal’s slow implementation has opened space for them again; Ansar Dine’s support is growing, particularly in Kidal.^[fn]Crisis Group telephone interviews, mediation specialist, December 2015 and January 2016.Hide Footnote  Elsewhere, the record of direct foreign military intervention is dismal. The 2003 Iraq invasion, though at first only tangentially linked to counter-terrorism, breathed new life into a global jihadist movement disoriented after the loss of Afghan sanctuaries. Even the 2006 U.S. surge, often heralded as a turning point, had a mixed record: the Awakening it supported was an initial military success against AQI, but its aftermath a political disaster, as Maliki further alienated Sunnis and undermined non-jihadist opposition.

In Afghanistan, U.S.-backed forces initially ousted the Taliban and weakened al-Qaeda, but now the insurgency is stronger than ever and the anti-Taliban alliance in Kabul shakier.^[fn]Crisis Group Report, Afghanistan’s Insurgency after the Transition, op. cit.; and Asia Report N°268, The Future of the Afghan Local Police, 4 June 2015.Hide Footnote  In 2006, when NATO deployed across the south, insurgents shifted to asymmetric tactics. A further influx of mostly U.S. troops in 2009 temporarily reversed some Taliban gains but at the cost of a massive upsurge in violence. As in the Iraqi surge, political failures outweighed military success: a tarnished presidential vote and potential openings for talks with Taliban leaders squandered by U.S. commanders determined to fight and U.S. announcement of a withdrawal date.^[fn]Crisis Group Asia Report N°207, The Insurgency in Afghanistan’s Heartland, 27 June 2011; and Crisis Group Report, Afghanistan’s Insurgency after the Transition, op. cit. For the military leadership’s reluctance to talk, Crisis Group interviews, U.S. officials, November 2015; also Rajiv Chandrasekaran, Little America: the War within the War for Afghanistan (New York, 2012).Hide Footnote  Reasons for the difficulties are many and complicated, including insurgent safe havens in Pakistan, but it is hard to conclude Western forces have made the region more stable or safer from Islamist radicalism. Instead, their presence has contributed to radicalisation across the region; in some Central Asian states, already threatened by the Afghan upheaval, reliance on closed regimes to keep open supply lines deepened destabilising patterns of rule.^[fn]For Pakistan, see, Moeed Yusuf, Pakistan’s Counterterrorism Challenges, op. cit. For Central Asia, see, for example, Crisis Group Europe and Central Asia Briefing N°78, Tajikistan Early Warning: Internal Pressures, External Threats, 11 January 2016; and Crisis Group Asia Report N°183, Central Asia: Migrants and the Economic Crisis, 5 January 2010.Hide Footnote

In Somalia, too, foreign forces gave impetus to radicals. Al-Shabaab won backing from both Islamists and nationalists opposing the Ethiopian invasion in 2007-2008. Many Somalis view troops from neighbouring countries now in the AU mission as occupiers with suspect motives, sentiments al-Shabaab, much like the Taliban, exploits.^[fn]See, for example, Crisis Group Africa Briefing N°74, Somalia’s Divided Islamists, 18 May 2010; and Report, Somalia: Al-Shabaab, op. cit.Hide Footnote  Western priorities, like counter-terrorism or national elections, are also out of step with those of communities that need to be won over and are more interested in local reconciliation or dispute resolution. Even if viewed as a containment strategy to keep regionally ambitious jihadists from power in Mogadishu, al-Shabaab’s attacks in Kenya suggest the military policy at best only a partial success.

More broadly, the Afghan and Somali experiences highlight the flaws in an approach that combines building centralised state institutions with counter-insurgency but without a wider political strategy that includes reconciliation.^[fn]The Afghan intervention’s early years focused solely on counter-terrorism, not state building. Even as Western powers gradually began to dedicate resources to institutions, they neglected those that interfaced with citizens, like local authorities and rule of law institutions, and the assumptions underpinning the intervention’s early stages regarding the Taliban persisted.Hide Footnote  Given the fragile regimes Western and African forces defend, neither the Taliban nor al-Shabaab look likely to be defeated nor their support sapped by improved governance soon. The military campaigns in fact work at cross-purposes, relying on local allies whose behaviour is part of the problem and, in some cases, have an interest in perpetrating insecurity. Military aid, meanwhile, has often fed corruption.^[fn]For example, see “Corruption: Lessons from the International Mission in Afghanistan”, Transparency International UK, February 2015;  “Operationalizing Counter/Anti-Corruption Study”, Joint and Coalition Operational Analysis (JCOA), 28 February 2014; or Dana Hedgpeth, “$13 Billion in Iraq Aid Wasted Or Stolen, Ex-Investigator Says”, The Washington Post, 23 September 2008.Hide Footnote  And if the record of foreign deployments is unhappy, more sobering still is that withdrawal can make things worse, or at least throw into stark relief their troubled legacies. In Iraq, the U.S. departure precipitated IS’s rise. In Afghanistan, the reduction in foreign forces has left some provincial capitals vulnerable to insurgents, with the U.S. now forced to recommit troops to prevent a Taliban takeover.^[fn]Barack Obama, “Statement by the President on Afghanistan”, Washington DC, 15 October 2015.Hide Footnote  Were AU forces to leave, al-Shabaab would retake Mogadishu.In Mali, perhaps, and certainly against Boko Haram, military action has been necessary. Elsewhere, too, it must usually be part of the response – even just to prevent jihadists’ expansion or avert atrocities. But recent history suggests governments and foreign partners have been too quick to go to war. Framing wars as struggles between governments and extremists is far too simplistic a dichotomy and overlooks complex, multi-layered and often old drivers of violence, a misdiagnosis that inevitably leads to mistakes. Many groups prove more resilient than anticipated. Insurgents with strong bonds to communities and who tap genuine grievances that are hard to resolve quickly and military action often aggravates are difficult to uproot. In the sense of their local roots, IS, al-Nusra and perhaps now even AQAP resemble more closely the Taliban and al-Shabaab than they do Boko Haram or al-Qaeda in Mali circa 2013.^[fn]Ansar Dine, al-Morabitoune/MUJAO and even al-Qaeda now appear to be putting down roots in Mali among, respectively, Iforas factions, Fulani nomads along the Niger border and Arab tribes around Timbuktu. Crisis Group observations, interviews and telephone interviews, Mali, January-February 2016.Hide Footnote  Without a workable strategy for a durable political order, military action against militants is unlikely, over time, to diminish the threat they pose.

When force is required, too often insufficient regard is paid to its wider impact. The past decade is littered with examples of violence either deepening support for extremists or leaving communities caught between their harsh rule and brutal campaigns against them. Jihadists’ ability to protect against predation by governments, other militias or foreign powers is far more central to their success than ideology. They perpetrate horrific acts of violence; the suicide bomber, reviled a few years ago as alien across much of the Muslim world, is now ubiquitous. Many fight, however, in conflicts in which all sides violate international law. Recovering the rulebook – starting with jihadists’ opponents showing greater respect for the legality of their actions – must be a priority.

Targeted killings are a tactic only as effective as the strategy that guides their use. They can disrupt extremist networks and potential attacks on the West across great distance and, in the case of drones, without immediate risk to U.S. military personnel. Certainly they have disrupted al-Qaeda in the Pakistani tribal areas and appear to have impacted IS’s ability to operate in Afghanistan.^[fn]Crisis Group telephone interview, international expert in contact with Taliban representatives, March 2016.Hide Footnote  They can hinder leaders’ movements and have a strong psychological impact on groups. But their greatest strength is also a weakness: by taking asymmetrical warfare to the extreme – with all risk of harm born by the target population, including non-combatants, and none by the attackers – drone strikes can destabilise local political conditions and fuel anger. Unless they are integrated into a broader strategy to calm a conflict, their tactical gains come at a cost.

Outside Pakistan, targeted killings have had less impact on militants’ strength. Drone strikes in Yemen, for years a central component of U.S. policy toward AQAP, have killed leaders, including al-Wuhayshi and, earlier, Ansar al-Awlaki, a top al-Qaeda ideologue. The movement has weathered this, while collateral civilian deaths have fuelled anger, particularly among tribes whose support against al-Qaeda is essential, and driven anti-Western sentiment, even if not direct backing for jihadists.^[fn]Christopher Swift, “The Boundaries of War: Assessing the Impact of Drone Strikes in Yemen”, in Peter Bergen and Daniel Rothenberg (eds.), Drone Wars (Cambridge, 2015) p. 79.Hide Footnote  If the impact of assassinating AQAP leaders in reasonably stable pre-2011 Yemen was uncertain, it is completely unpredictable in today’s chaos, as al-Qaeda competes with IS and is enmeshed in local alliances and conflicts. Al-Nusra’s alliance with rebels in Syria means that there, too, killing its operatives may have unintended consequences, particularly deepening anger against the West among potential allies and strengthening IS.^[fn]This is particularly so when there is collateral damage, which contributes to the outrage of many rebels that Western powers can deliver strikes but do nothing against regime airstrikes.Hide Footnote  This assumes, of course, that drone operators can reliably distinguish among insurgents, targeting some but not others – which is difficult, particularly in urban areas.^[fn]For more statistics, see: “Drone Wars Yemen: Analysis”, New America Foundation.Hide Footnote

In Somalia, the U.S. has killed commanders, including al-Shabaab’s military chief, Aden Hashi Farah Ayro (with a cruise missile in 2008) and its leader, Ahmed Abdi Godane (with a drone strike in 2015). Successors emerged fast, however, and the Ayro to Godane transition may have contributed to the movement’s increasing radicalisation, as efforts were accelerated to affiliate to al-Qaeda.

Elsewhere, too, harder-line commanders have replaced assassinated leaders: the deeply sectarian Hakimullah Mehsud of the Pakistani Taliban replacing Baitullah Mehsud; and Abubakar Shekau replacing Boko Haram’s Mohammed Yusuf (killed in police custody).^[fn]Crisis Group Report, Pakistan: Countering Militancy in FATA, op. cit.Hide Footnote  During both the Afghan and Iraqi surges, killings of mid-level commanders appear to have brought in a more radical, brutal generation.^[fn]For more, see Crisis Group Reports, Afghanistan’s Insurgency after the Transition, p. 26; and Iraq after the Surge I, both op. cit.Hide Footnote  While this may, in some cases, have frayed relations between insurgents and communities, killing leaders in the hope of radicalising groups, in the hope they will then alienate communities, in the hope those communities can subsequently be won over seems a flimsy strategy given the track records of state and foreign forces in both places.

In sum, assassinations can help disrupt leaders’ and groups’ ability to operate, but predictability tends to be low and the risk high. Against large insurgent movements in war zones, particularly those like IS whose inner workings and command structures are opaque, the impact is particularly uncertain. Though it may fragment some groups, in the case of a well-organised group like IS a replacement, perhaps more radical, is likely to emerge quickly.^[fn]A wide-ranging study of insurgencies found that leadership decapitation works best when movements are “weak organisationally and focused around a cult of personality”, neither of which applies to IS. Max Boot, Invisible Armies: An Epic History of Guerrilla Warfare from Ancient Times to the Present (New York, 2013). Studies examining the impact of killing leaders of criminal organisations that provide basic public goods suggest decapitation is unlikely to cause the organisation’s demise in the absence of a wider strategy, usually involving the state moving in to provide those services. See, for example, Cockayne, Hidden Power, op. cit. Hide Footnote  An era of jihadist infighting – al-Qaeda and others confronting IS in Afghanistan, Libya, the Sahel, Syria and Yemen – makes the impact still less sure. Little suggests targeted killings will help either end the conflicts jihadists fight in or decisively weaken their movements.

Talking to IS- and al-Qaeda-linked groups, whether to negotiate over hostages, humanitarian access or an end to violence, poses practical and substantive challenges. There is physical danger to mediators. Movements’ hierarchy and structures are often obscure. Leaders may hold views different from those on the front lines. Mediators often face resistance from states that have suffered attacks. Obstacles can also be legal. Some states prohibit material support of groups designated terrorist in ways that would penalise dialogue; others ban facilitating transport of their representatives to a safe meeting place.^[fn]The U.S., for example, prohibits material support including “expert advice or assistance” to foreign terrorist organisations, though the secretary of state can approve exceptions. “Providing material support or resources to designated foreign terrorist organizations”, Title 18, U.S. Code 2339B. UN Security Council Resolution 2178 (24 September 2014) resolves to limit the ability of terrorist groups to travel or acquire material/equipment to conduct their activities.Hide Footnote

IS and al-Qaeda leaders’ transnational ideology also closes space, at least for political engagement. Top IS leaders make no demands; even negotiating relief delivery with local commanders has been hard.^[fn]Crisis Group telephone interviews, humanitarian workers and organisations, November 2015.Hide Footnote  IS may tap genuine grievances, but neither its leaders nor many within al-Qaeda indicate their struggle would end were those addressed; little suggests attempts to negotiate would end violence. Some of their objectives – the restoration of a caliphate from southern Spain to Indonesia, the destruction of Israel, Westerners’ complete withdrawal from Muslim world – are unattainable by negotiation. Though their austere social vision, including literal interpretation of the Quran, is not unique to them, ending the wars they fight in will require some degree of political and religious pluralism.

At times, too, negotiations have emboldened movements with scant popular support. In the Pakistani tribal areas, the military’s deals with Pakistani Taliban factions have backfired. Similarly, the federal and Khyber Pakhtunkhwa regional government endorsed repeated peace deals with Mullah Fazlullah’s Taliban faction after it captured Swat, each ceding it more authority, until it seized Buner, a few hundred kilometres north of Islamabad, sparking domestic and international outrage and a more serious military offensive against it.^[fn]Crisis Group Report, Pakistan: Countering Militancy in PATA, op. cit.Hide Footnote

On balance, though, governments have unhelpfully shied from dialogue, a tendency that the reframing of movements of many stripes as “violent extremists” risks deepening. With hindsight, the U.S. rejection in 2001 of some Taliban leaders’ offers to accept the new order in return for government posts or their safety looks unwise.^[fn]See Crisis Group Report, Insurgency in Afghanistan’s Heartlands, op. cit. According to experts with contacts in the insurgency, the Taliban was sending envoys up to 2005. Crisis Group telephone interview, March 2016.Hide Footnote  Bringing them in would not have prevented some form of insurgency without an accompanying shift from the counter-terrorism focus, a more inclusive settlement in Kabul, better administration there and in the provinces and greater efforts to bring along Pakistan. But it would have changed that insurgency’s form. Now, Kabul and its foreign allies will have to surrender much more to persuade the Taliban to stop fighting, if indeed the movement intends to or can without fragmenting.

Reluctance to engage at the height of the war on terror has meant opportunities with al-Shabaab have been missed, too.^[fn]Crisis Group Reports, Somalia: To Move Beyond the Failed State; and Somalia’s Divided Islamists, both op. cit.Hide Footnote  In Mali, involving Ansar Dine leader Iyad ag-Ghali in the peace process would have been challenging, but many believe that without him, peace around Kidal will remain elusive. Nor did the Mali deal explore the role of religion in politics; doing so might have undercut radical groups’ support by taking up one of their main demands.^[fn]Some Tuareg rebels tried to introduce discussion on religion during the Algiers peace talks, but most international mediators warned that was a red line. Crisis Group interviews, mediation team members, September 2014.Hide Footnote  Efforts to persuade Ansar al-Sharia leaders in Libya to accept democracy after the revolution appeared to bear some fruit, before being scuppered by escalating violence.^[fn]From late 2012 to early 2014, Libyan politicians, especially those linked to the Libyan Muslim Brotherhood, engaged with Ansar al-Sharia militants in Benghazi and Derna seeking to persuade them that democracy, if founded on Sharia-compliant laws, was not un-Islamic. They had some success – a commander visited the elected legislature for example – but ceased in mid-2014, when Ansar al-Sharia and the individuals they were talking to were targeted by Haftar’s forces. Crisis Group interviews, observations, Tobruk, al-Bayda, Benghazi, 2014-2015.Hide Footnote

Similarly, after the 2009 Maiduguri crackdown, Boko Haram called for the restoration of its mosque (destroyed in the fighting), and for those responsible for its leader’s killing to be held accountable.^[fn]Various attempts have been made to engage Boko Haram. Former President Olusegun Obesanjo met with Yusuf’s brother-in-law, Babakuru Fuggu, whose father also died in the Maiduguri battle; Babakuru was shot by a suspected Bok0 Haram member shortly afterwards. On occasion individuals claiming to represent Boko Haram have been dismissed by Shekau. See, for example, Crisis Group Report, Boko Haram, op. cit., and Virginia Comolli, Boko Haram, Nigeria’s Islamist Insurgency (UK, 2015).Hide Footnote  Engagement would have been hard, but those demands might have offered a starting point. Instead, both sides escalated, and Boko Haram metastasised into a regional menace. The Nigerian government should continue to offer to talk to any member ready to engage – partly to counteract the movement’s narrative of a cruel, oppressive state and partly because there may be more pragmatic factions that can be brought in. It should also bring Yusuf’s killers to justice and release the wives of Boko Haram’s leaders it has imprisoned. But ending violence through a mediated settlement with the radical and increasingly nihilist core looks remote.

Refusing in principle to engage jihadists seems an anachronism, given their prominence, the ties some enjoy to communities and the spotty records of military action against them while trying to sap their support through better governance. Efforts are underway already with some movements previously cast as “irreconcilable”, including the Afghan Taliban; discreet efforts are ongoing with parts of al-Shabaab; Ahrar al-Sham is now rightly seen, at least by Western and some Gulf powers, as a viable interlocutor for Syrian peace talks, though al-Qaeda operatives were among its founding members.^[fn]Abu Khalid al-Suri, among Ahrar al-Sham’s original founders, was as an al-Qaeda operative later appointed by Zawahiri to mediate between al-Nusra and IS but killed in a suspected IS suicide bombing. See Hanieh and Rumman, The “Islamic State” Organization, op. cit., or Charles R. Lister, Syrian Jihad, op. cit.Hide Footnote  As noted, al-Qaeda affiliates’ seizure of territory, coordination with aid organisations and ties to state-supported armed groups may open opportunities even with these groups.

Contact with many groups should be approached without much expectation their core will easily move off global jihad, let alone toward peaceful political participation or Salafi quietism. Prospects are probably brighter with groups with national goals and even more so with those prepared to accept pluralism. Nor should governments themselves necessarily attempt to engage. But policymakers, certainly in Western capitals, could take advantage of often longstanding contacts between those in radical movements and others and of the engagement that already takes place, including by religious or other community leaders, non-state mediators and humanitarian groups. All these can help shed light on dynamics within groups, facilitate humanitarian access and, in places, alleviate suffering. Although many jihadist movements have perpetrated horrific violence against civilians, the wars they fight in have featured atrocities by many other actors as well. Crimes should be dealt with through transnational justice, if feasible, not shape decisions on whether to talk.

Mediators always face questions. What is the purpose of engagement? What are the risks? Will it empower unpopular hardliners at the expense of those more inclined to compromise? Will it incur costs with others? Who is best placed to do it? Can it delegitimise the use of violence by those that do not participate? Although the answers may differ, these questions are the same for the most extreme group as for any armed movement. Particularly important now with all groups – those with transnational as well as national goals – is to monitor them as prominent forces in conflicts, not just as threats to the West; keep the door to engagement ajar; and identify and assess prospects as they arise. Opportunities to open discreet lines of communication to at least try to define whether groups have demands that could be used as the basis for talks and can be moved away from those that are irreconcilable, are usually worth pursuing.

The recent expansion of IS and al-Qaeda-linked groups injects new urgency into conflict prevention, particularly in the belt running from West Africa to South Asia. Since such movements are likely to profit from any new crisis, and prospects for reversing their gains or ending the crisis diminish once they do, it is important to shore up states that are still standing but vulnerable. Beneath a veneer of stability, some – in the Lake Chad Basin, Sahel, North Africa, Middle East, even the Gulf and certainly Central Asia – are brittle.

How the emerging Countering Violent Extremism (CVE) agenda contributes to this is still unclear. The agenda was conceived as a soft counterweight to the militarised response to 9/11 and initially pioneered by development actors who recognised the flaws in an approach rooted only in force.^[fn]Georgia Holmer, “Countering Violent Extremism: A Peacebuilding Perspective”, U.S. Institute of Peace, Special Report 336, September 2013.Hide Footnote  Its action points tend to include civic engagement with communities; push-back – or a “counter-narrative” – against intolerant strands of religion; a focus on stemming the flow of foreign fighters; and addressing “root causes” of radicalisation, often relating to the lack of opportunity for young people and, in some cases, poor or abusive governance. Different states and the UN emphasise different aspects: some ideology; others the “pull” factors or specific recruitment paths that entice individuals to join up; yet others “root causes” or “push” factors. The UN Secretary-General’s recent Plan of Action on Preventing Violent Extremism calls on states to develop their own plans of action that include measures that address diverse sources of fragility.^[fn]UN Plan of Action. The plan refers to preventing, rather than countering, violent extremism, but the thinking is much the same.Hide Footnote

Much within the CVE agenda makes sense. The emphasis in the UN plan is vital, for example, on the grievances that underpin extremists’ ability to recruit; state responsibilities; and the links between radicalisation and human rights abuses, repressive and abusive governance, crushed aspirations and marginalisation. So, too, is its call for member states not to violate human rights as they respond. Though the plan stops short of explicitly linking jihadists’ recent gains to major and regional powers’ policies in the Middle East, it recognises that violent extremism does not arise in a vacuum and calls for redoubled efforts to end protracted conflicts.

Given that the fourth wave owes much to the failures of securitised policies since 9/11, criticising the CVE agenda, devised precisely to correct those failures, might seem churlish. But there may be dangers in countries using CVE as the main prism through which to see threats to their stability.

First, while recognising the diverse factors that can drive extremism and shifting resources toward efforts to tackle them is valuable, re-hatting efforts explicitly as CVE may be less so. Many are worthwhile without vesting them with de-radicalisation expectations they may be unable to meet or that could undermine them. Creating jobs for youths is sensible, for example, but prevents them joining extremist groups only in some conditions. Helping marginalised communities is vital, but doing so to win support against “extremists”, or, worse, conditioning development accordingly, can work against aid and those delivering it. Education is a child’s basic right; reframing it or any government obligations to its citizens as CVE may distort delivery of basic public goods. Similarly women activists should be engaged to help develop policy, not inform on their children, as has happened in places.^[fn]Crisis Group telephone interview, Sanam Anderlini, co-founder and executive director, ICAN, 14 February 2016.Hide Footnote  Encouraging governments toward inclusion and gradual reform is usually the most valuable contribution allies can make to preventing the crises that open opportunities for extremists. But branding such diplomacy as CVE adds no value.

Secondly, governments and the UN may not be best positioned to develop counter-narratives on religion themselves, while co-option can weaken “friendly” imams. Governments should allow and protect space for diverse Muslim voices, Salafi and otherwise. Perhaps more important, as shown, ideology’s role in driving extremists’ rise is not straightforward. Although Salafi proselytising and often state-sponsored Islamisation of parts of society have helped set the stage, the fourth wave owes more to jihadists’ exploitation of war and state collapse, or armed groups adopting more extreme tactics as crises deepen, than to earlier radicalisation. During crises, support extremists may enjoy from communities is, in most cases, based less on shared values and more on what else they provide when things fall apart: protection against a hated regime, quick dispute resolution, social advancement or opportunity for profit.

Chad is an example worth study. After initially staying out of Nigeria’s fight against Boko Haram, President Idriss Déby sent troops in early 2015, as violence began to cut off cattle trading routes and affect Chad’s economy. His forces spearheaded offensives that routed militants from the villages they had captured across north-eastern Nigeria. In response, Boko Haram began to threaten Chad and Déby in online state­ments. By then, the crisis had spilled over the border, with militants penetrating Lake Chad’s surroundings and launching suicide attacks in N’Djamena. Déby cracked down on lake communities, accusing them of ties to Boko Haram and, like other riparian governments, limited their fishing, thus restricting livelihoods and alienating potential allies against the militants.

The gradual, mostly Gulf-funded encroachment of Salafism preceded Boko Haram. As elsewhere in Africa, Sufi leaders in Chad lament ground lost, particularly with youth, to more radical Salafi imams. Déby promotes what he calls “African Islam”, locally flavoured Sufism, and tries to limit the activities of Salafi mosques and preachers.^[fn]The government, for example, banned women’s veils after the suicide attacks in the summer of 2015. “Chad’s ban on burqa divides Muslims”, The Express Tribune, 21 June 2015.Hide Footnote  Nothing suggests that Chad’s Salafis have ties to or even sympathy for Boko Haram, but harsh action against non-violent Salafis risks furthering what it is meant to prevent.

Boko Haram is likely to remain disruptive, particularly if Chad and its neighbours cannot offer hope to people in affected areas. To a lesser degree, some Salafis may strain the country’s social cohesion. But the gravest mid-term threat to stability almost certainly emanates from Déby’s personalised rule and accumulation of power – a trend that his tightening alliance with Western powers and the training they give his forces to fight jihadists elsewhere aggravate. Without reform, he is likely to either provoke internal instability before he departs office or leave chaos behind. Little suggests that radical Islam would be used to frame either resistance to his rule or the succession in-fighting, though Salafism’s spread perhaps makes that somewhat more likely. More probably, jihadists, whether Boko Haram or more sophisticated North African and Sahel movements, will infiltrate and profit from any crisis, much as they have done elsewhere, even in places with little history of radicalisation.

So while African and other leaders are justifiably angry at the unregulated flow of Gulf money to intolerant preachers, focusing on that to the detriment of other sources of fragility risks missing the forest for the trees. The likeliest way IS or al-Qaeda-linked groups can capture part of the Chadian state is if it collapses in a struggle over power and resources. The same applies in other Lake Chad Basin states, particularly Cameroon and Niger, in parts of Central Asia and many other places. Vital is that measures against jihadists do not inadvertently make violent breakdown more likely by propping up exclusive, destabilising patterns of rule.

Perhaps most worrying across the CVE agenda is that the term “violent extremist” is loosely defined, if at all. Does it refer to doctrine, tactics, outreach or aspirations? Some Western governments mostly use the label as a euphemism for the jihadists this report covers; others so classify different kinds of Islamic militants like Hamas; yet others include violent right-wing movements in Europe.^[fn]Crisis Group interviews, New York and Washington DC, September 2015-February 2016.Hide Footnote

The label thus obscures more than illuminates, potentially casting diverse forms of protest, rebellion and radicalism together as “violent extremist”. If confusing the Taliban and al-Qaeda was a mistake fifteen years ago, creating a category that might include IS, Hamas, the FARC insurgents in Colombia and right-wing extremists in the West is analytically flawed and risks setting policy on a course that allows leaders to portray their enemies as irreconcilable and lock their countries into endless wars against them. Even the movements this report discusses – among the most extreme contemporary non-state armed groups in terms of their beliefs and goals – comprise a dedicated core and then many others fighting for a diverse array of often local, non-ideological motives. Policymakers should disaggregate even the most radical movements and look for opportunities to end violence, not lump others in with them.

The label “violent extremist”, much like that of “terrorist”, also risks delegitimising groups’ political grievances and agendas – however remote some of their goals – and pushing policy away from politics. The UN plan, for example, despite stressing the importance of dialogue between conflict parties, still appears underpinned by the assumption that “violent extremists” are beyond the pale. This leaves an empty political middle ground between the mostly development- and de-radicalisation-oriented policies usually considered part of CVE and counter-terrorism or counter-insurgency policies. By buying into the “violent extremism” language, the Secretary-General risks reinforcing the mindset that justifies the hard security measures he warns against.

The CVE agenda has value, of course – and not only as a corrective to previous mistakes. It might help in tackling IS recruitment, which in many places hinges less on imams and religion than on social media and appeals to fraternity, belonging and purpose. It might, for example, advance de-radicalisation in prisons, a main recruitment venue, and measures to assist particularly vulnerable youth groups, a main recruitment pool.

But governments as they develop approaches to counter the influence of extremist movements would be wiser to narrow CVE to a handful of context-specific activities against “pull” factors and to funding research on radicalisation, patterns of which are still little understood. Efforts to address root causes of instability and conflict should, naturally, be redoubled; donors can usefully shift resources from military and security spending toward addressing those underlying factors. However they and governments they support should think carefully about the benefits in each case of labelling these efforts CVE. Most of all they need to involve a wide range of people, including women, from communities affected in developing whatever policies are adopted and how they should be framed.

IS’s rise in Iraq and Syria, its reach elsewhere and the growing potency of al-Qaeda-linked groups over the past few years pose a major threat. Their violence, particularly IS’s theatrical displays, their intolerance and much else in their thinking are affronts to the vast majority of Muslims. Their prominence on many of today’s battlefields complicates efforts to end wars and deepens humanitarian suffering. World

leaders must do whatever possible to diminish the threat they pose, stop them recruiting, curtail the spread of their ideology and prevent similar groups emerging.

Reversing the fourth wave, however, requires focusing on not only an enemy easy to hate but also the conditions that have enabled its rise: the enormous violence Sunnis have suffered in Iraq and Syria; upheaval and escalating Middle East power rivalries; the dangerous sense of victimisation among the Arab world’s Sunni majority; increasing identity politics and sectarian hatred; the Libyan and Sahel instability after Qadhafi’s ouster; the ideological space that has opened up with the Muslim Brotherhood’s demise; dim prospects for reform in countries that have not yet succumbed; and many states’ struggles to meet needs of citizens, particularly those in peripheries, Muslim minorities and growing youth populations. IS’s emergence throws into stark relief Sunnis’ desperation in Iraq and Syria. Its ability elsewhere to recruit, even tiny minorities, shows states’ failures to deliver as much as the power of what the movement sells. IS provokes justifiable outrage, but blame for its rise is widely shared and should provoke introspection beside condemnation; compassion as much as revulsion.

Exactly how further expansion would play out is unclear. The interaction between the threat jihadists pose and other sources of fragility varies from place to place. Despite their contrasting strategies, both IS and al-Qaeda have shown they can exploit cleavages along multiple lines – particularly sectarian in the case of IS, but also generational, between communities and within them, between those with power and those without. Their terrorist attacks, like those of many groups before them, aim to deepen divides, aggravate conditions that enable them to expand and provoke reactions that do the same.

What the past few years show clearly, however – especially but not only in the Middle East – is that war and state collapse are massive boons for both movements. Dialling back the conflicts they fight in and preventing breakdowns elsewhere are ambitious agendas, requiring shifts in some major and regional powers’ strategic calculations and that leaders thus far displaying little inclination to reform do so. But trying to counter IS’s and al-Qaeda’s influence while wars rage and bloodshed plays out on local media across the Muslim world is likely to prove futile. And while either movement could itself provoke a major crisis in a new theatre, the more probable path along which either captures territory or establishes a serious presence elsewhere is by profiting from a collapse in which it initially plays no central role. Their increasing potency notwithstanding, the gravest danger these groups pose, at a particularly perilous moment of world history, is that they provoke reactions that deepen the conditions they feed off and, like mistakes after the 9/11 attacks, create new instability that again plays into their hands.

Brussels, 14 March 2016