Arrow Left Arrow Right Camera icon set icon set Ellipsis icon set Facebook Favorite Globe Hamburger List Mail Map Marker Map Microphone Minus PDF Play Print RSS Search Share Trash Twitter Video Camera Youtube
Sunni fighters opposing the Islamic State gather in formation along the front line near the Islamic State-controlled village of Haj Ali in the southern Mosul countryside near Makhmour, Iraq, on 19 November 2015. MAGNUM/Moises Saman

Counter-terrorism Pitfalls: What the U.S. Fight against ISIS and al-Qaeda Should Avoid

This report examines President Trump’s emerging counter-terrorism policies, the dilemmas his administration faces in battling ISIS and al-Qaeda across the Middle East and South Asia, and how to avoid deepening the disorder both groups exploit.

Executive Summary

In pledging to destroy the Islamic State (ISIS), U.S. President Donald J. Trump looks set to make counter-terrorism a centrepiece of his foreign policy. His administration’s determination against groups that plot to kill Americans is understandable, but it should be careful when fighting jihadists not to play into their hands. The risks include angering local populations whose support is critical, picking untimely or counter-productive fights and neglecting the vital role diplomacy and foreign aid must play in national security policy. Most importantly, aggressive counter-terrorism operations should not inadvertently fuel other conflicts and deepen the disorder that both ISIS and al-Qaeda exploit.

The new U.S. administration has inherited military campaigns that are eating deep into ISIS’s self-proclaimed caliphate. Much of Mosul, its last urban stronghold in Iraq, has been recaptured; Raqqa, its capital in Syria, is encircled. Its decisive defeat is still a remote prospect while the Syrian war rages and Sunnis’ place in Iraqi politics is uncertain. The threat it poses will evolve in its heartlands and elsewhere, as fighters disperse. But ISIS is in retreat, its brand diminished. For many adherents, its allure was its territorial expansion; with that gone, its leaders are struggling to redefine success. Al-Qaeda could prove harder to suppress. Its affiliates fight across numerous war zones in coalitions with other armed groups, its operatives are embedded in local militias, and it shows more pragmatic adaptability to local conditions.

Though the roots of ISIS’s rise and al-Qaeda’s resurgence are complex and varied, the primary catalyst has been the turmoil across parts of the Muslim world. Both movements grow when things fall apart, less because their ideology inspires wide appeal than by offering protection or firepower against enemies, rough law and order where no one else can or by occupying a vacuum and forcing communities to acquiesce. The U.S. can do only so much to reboot Arab politics, remake regional orders or repair cracked fault lines, but its counter-terrorism strategy cannot ignore the upheaval. So long as wars continue and chaos persists, jihadism will thrive, whatever ISIS’s immediate fate. In particular, the new administration should avoid:

  1. Angering communities. Campaigns against jihadists hinge on winning over the population in which they operate. Offensives against Mosul, Raqqa or elsewhere need to avoid destruction but also need plans to preserve gains, prevent reprisals, stabilise liberated cities and rebuild them; as yet, no such plan for Raqqa seems to exist. “Targeted” strikes that kill civilians and alienate communities, as appears to have been the case in the January Yemen raid and the 16 March strike in Syria’s Aleppo province, are counterproductive, regardless of immediate yield. Loosening rules and oversight designed to protect civilians, as has been suggested, would be a mistake.
     
  2. Aggravating other fronts. The new administration’s fight against ISIS and al-Qaeda intersects a tinderbox of wars and regional rivalries. No regional state’s interests dovetail precisely with those of the U.S.; few consider jihadists their top priority; most are more intent on strengthening their hand against traditional rivals. The U.S. should be careful that the Raqqa campaign does not stimulate fighting elsewhere, particularly among Turkish and Kurdish forces and their respective allies. Success in Mosul hinges on preventing the forces involved (the Iraqi army, Kurdish peshmerga units, Shiite militias and Sunni tribes; Turkey and Iran) battling for turf after ousting ISIS. Likewise, support for Gulf allies should not mean a blank check for the Saudi-led Yemen campaign, which – if wrongly prosecuted – would play further into al-Qaeda’s hands.
     
  3. Picking other fights. Confronting Iran, which the administration identifies as a priority alongside the fight against ISIS and al-Qaeda, requires careful consideration. Militarily battling Tehran in Iraq, Yemen or Syria, questioning the nuclear deal’s validity or imposing sanctions that flout its spirit could provoke asymmetric responses via non-state allies and put Iraq’s government in an untenable position. Iran’s behaviour across the region is often destabilising and, by aggravating sectarian tensions, provides fodder to jihadist groups; as with similar conduct by others, it calls for a calibrated U.S. response. But the answer ultimately lies in dampening the Iranian-Saudi rivalry, not stimulating it with the attendant risk of escalating proxy wars across the region and reinforcing sectarian currents that buoy jihadists. Similarly, sabre-rattling with China hinders diplomacy with Pakistan and thus efforts to stabilise Afghanistan; effective counter-terrorism in South Asia requires cooperation with Beijing.
     
  4. Defining the enemy too broadly. ISIS and al-Qaeda thrive on confusion generated by how the U.S. defines its foe: violent jihadists, political Islam or Muslims as a whole. Designating the Muslim Brotherhood a terrorist group would be a self-inflicted wound, alienating an ideological and political counterweight to jihadism. Similarly, many armed groups fight beside al-Qaeda in alliances that are tactical and do not signal support for jihadists’ goals of attacking the West or establishing a caliphate. Prising them away from al-Qaeda would be wiser than fighting them all.
     
  5. Neglecting peace processes. From Libya to Somalia, Syria, Iraq, Yemen and Afghanistan, no country where ISIS or al-Qaeda branches hold territory has a single force strong enough to secure the whole country. Without accommodation, factions will either ally with jihadists against rivals or use the fight against them for other ends. Backing forces for counter-terrorism while neglecting efforts to promote compromise will deepen instability.
     
  6. Fighting terrorism without diplomacy. Navigating allies’ rivalries, preventing a free-for-all in Mosul, managing the fallout from Raqqa, mediating between Afghan, Iraqi or Libyan factions – all are diplomats’ work. Multilateral engagement matters too, whether to back UN mediation, enlist its help for reconstruction and stabilisation or use UN and other multilateral frameworks for counter-terrorism cooperation. Staffing the State Department’s top levels and sustaining its expertise are priorities. The cuts proposed to U.S. diplomacy and foreign assistance, including to the UN’s budget, would damage U.S. security.

That the new administration wants to prioritise operations against groups that plot against the U.S. is understandable, but counter-terrorism does not exist in a vacuum. The U.S. administration’s executive order banning entry from certain Muslim countries; the troubling rhetoric of some of its officials; the calling into question of some of the restraints imposed on military operations; and the proposed slashing of the State Department and development budgets all undermine its goal of protecting Americans from terrorism. More broadly, it should be cautious not to overlook or aggravate other sources of instability even as it takes steps to defeat jihadists. The big winners from any new disorder in the Muslim world would be groups like ISIS and al-Qaeda – whatever guise they ultimately assume.

Washington/New York/Brussels, 22 March 2017

I. Introduction

U.S. Secretary of Defense James Mattis submitted an initial battle plan against ISIS in late February.[fn]This report adopts the acronym ISIS, as that is the version used by the new U.S. administration, to which it is primarily addressed.Hide Footnote Its precise contents are not public, but the administration appears ready to accelerate operations against ISIS and al-Qaeda-linked groups across the Muslim world.

The jihadist landscape has evolved fast in recent months.[fn]For a definition of “jihadist” and an explanation for its use, see Crisis Group Special Report N°1, Exploiting Disorder: al-Qaeda and the Islamic State, 14 March 2016.Hide Footnote ISIS has lost swathes of its Iraqi and Syrian heartlands. Its Libyan branch, with closest ties to the Iraqi leadership, has been ousted from the Mediterranean coastal strip it held. Boko Haram, whose leaders pledged allegiance to ISIS, menaces the four African states around Lake Chad but has split and lost much of the territory it held a year ago. Though smaller branches exist from Afghanistan to the Sinai and Yemen to Somalia, the movement has struggled to make major inroads or hold territory elsewhere. Fewer local groups are signing up. Fewer foreigners are travelling to join; a main danger now is their return to countries of origin or escape elsewhere.[fn]This report focuses on U.S. military operations against ISIS and al-Qaeda, not the danger posed by the potential dispersal or return of foreign fighters or the risk ISIS remnants may inspire attacks elsewhere.Hide Footnote

Al-Qaeda, too, has changed. Its affiliates, particularly its big branches in Somalia, Syria and Yemen, are more influential than the leadership in South Asia.[fn]Other affiliates include al-Qaeda in the Islamic Maghreb, now part of a new coalition of jihadist groups in the Sahel, and al-Qaeda in the Indian subcontinent.Hide Footnote Osama bin Laden’s successor, Ayman al-Zawahiri, still inspires loyalty and offers guidance but has less influence on daily operations. Many significant al-Qaeda operatives are in Syria or Yemen. The affiliates’ primary identity is more local than transnational. While cells within them aim to inspire attacks against the West, they fight local wars and have opened their doors to many fighters motivated by local concerns.

As grave a threat as ISIS or al-Qaeda is the disorder across parts of the Muslim world that has enabled their growth. Neither it nor the fraying social contracts and regional power rivalries beneath much of the chaos show signs of abating. The pool both movements draw from has deepened, as more young people have come into their orbit.

A main dilemma facing the Trump administration is to find the right balance between military action against jihadists and policies aimed at tackling the conditions they exploit. This report, drawing from Crisis Group’s decades of research on war and jihadism, explores potential pitfalls in getting the balance wrong. It poses four questions the new administration’s ultimate plan should answer: (i) how to fight ISIS in Iraq and Syria; (ii) how to tackle jihadists elsewhere without aggravating the chaos on which they feed; (iii) what direction for Afghanistan and Pakistan policy?; and (iv) how to define the enemy. Though jihadists pose a threat elsewhere, with Africa of particular concern, it focuses mostly on the Arab world and South Asia – roughly corresponding to U.S. Central Command’s area of responsibility – as the main arena for U.S. counter-terrorism operations and the regions hosting the largest numbers of U.S. forces.[fn]For analysis of the Sahel, see Jean-Hervé Jezequel and Vincent Foucher, “Forced out of Towns in the Sahel, Africa’s Jihadists go Rural”, Crisis Group commentary, 11 January 2017. For Crisis Group’s extensive Boko Haram work, see https://www.crisisgroup.org/boko-haram-insurgency.Hide Footnote

II. How to Fight ISIS in Iraq and Syria?

In Mosul, ISIS is hemmed in. The agreement the Obama administration forged before the offensive has largely held: U.S.-trained elite Iraqi counter-terrorism forces, Iraqi army divisions and local Sunni auxiliaries are fighting ISIS in the city, with support from Western advisers and special forces; Iran-backed Shiite militias and the Turkey-backed peshmerga of the Kurdistan Democratic Party, both forces distrusted by Mosul’s inhabitants, have mostly remained in the outskirts. Parts of the city east of the Tigris have been recaptured. The warren of alleys in its old quarter and adjacent neighbourhoods to the north west, where ISIS fighters are entrenched, are likely to see even fiercer battles.

Greater challenges will follow Mosul’s capture. The first will be to secure the city and, to the extent possible, prevent reprisals. Divisions within the local Sunni Arab community mean that intra-Sunni bloodshed is as much a risk as Shiite, Kurdish or Yazidi violence against Sunni Arabs. Preventing clashes among forces involved in the campaign and among their foreign backers over spheres of influence in the city and its surroundings is another challenge. Whatever remains of ISIS may escape into the desert but is likely also to operate cells in Mosul and other cities – perpetrating attacks, sowing division, extorting reconstruction funds, offering a path for those angry at whatever arrangements follow its rule or simply lying low to await more opportune times.

The best way to inoculate Iraq against the return of ISIS or a jihadist successor is to help Prime Minister Haider al-Abadi’s government re-enfranchise Sunnis and bring them back into the political fold.

Sunnis’ role in Iraq’s politics and security, or even what Sunni political identity will emerge, is unclear. Sunnis are traumatised and atomised, fragmented between tribes, within tribes and between generations. Shiite and Kurdish forces entrenched in Mosul’s surroundings will not easily relinquish areas under their control, which also hinders any potential devolution of political authority and security responsibility. The best way to inoculate Iraq against the return of ISIS or a jihadist successor is to help Prime Minister Haider al-Abadi’s government re-enfranchise Sunnis and bring them back into the political fold. Strengthening the state is vital, and will help as long as it is also inclusive.

Iran is vastly influential in Iraq, but its outsized role is resented and contested by a large array of Iraqi politicians. These include Shiite leader Muqtada al-Sadr, who has adopted a nationalist platform ahead of provincial and national elections, as well as the Shiites’ foremost religious authority, Grand Ayatollah Ali al-Sistani. Many residents of Baghdad, regardless of religious affiliation, complain that Iran-backed militias are morphing into a multi-functional and corrupt para-state, running businesses as part of an extending patronage network. Efforts by Iran-backed militias in the north to forge a land corridor to Syria risk setting the stage for a next phase of conflict.  

Paradoxically, however, more aggressive U.S. efforts to turn Iraq into a battle-field to reverse Iran’s influence would likely have the opposite effect. They would not only be destabilising, given Tehran’s sway over the Iraqi government, powerful Shiite militias and parts of the army, federal police and body politic, but would also vastly complicate counter-ISIS operations and – by placing Prime Minister Abadi squarely at the centre of U.S.-Iranian competition – undercut Baghdad’s efforts to forge a path more independent from Tehran.

A better way to sustain momentum against ISIS and promote Iraq’s stability would be for the U.S. to play a balancing role: preserving Baghdad’s independence by supporting its military as well as economic and humanitarian efforts, while keeping Iran, Turkey and their respective allies at bay. Flashpoints – such as Sinjar, where Turkey’s archenemy the Kurdish insurgent group Kurdistan Workers’ Party (PKK) holds territory; Bashiqa, where Turkey maintains a military presence despite Iraqi objections; and the Turkman town of Tel Afar, where Shiite militias’ advances sow fear among Sunnis – will necessitate deft diplomacy and management. In particular, the U.S. should invest in deepening security cooperation among the Abadi government, the Kurdistan Regional Government and Sunni Arab fighters, while acknowledging that Shiite militias also will have to play a role lest this delicate balance unravel.

Critical for the new U.S. administration is to learn from mistakes made after the Sunni Awakening and U.S. surge defeated ISIS’s precursor, al-Qaeda in Iraq (AQI), a decade ago. The Awakening movement was subsequently handed over to and promptly betrayed by the government of Abadi’s predecessor, Nouri al-Maliki. Military gains were not translated into a sustainable political order. Doing so now will be harder still, given considerably diminished U.S. military presence and leverage, Iran’s entrenchment, the fracturing of Sunni politics and a legacy of distrust deriving from that betrayal. It requires rebuilding liberated areas but also ensuring that international aid does not create new division by favouring some groups over others. It also means working with communities – arranging joint security arrangements and local governance – to avert another descent into sectarian chauvinism and revenge that would allow ISIS to re-emerge.

The campaign against ISIS in Syria is yet more complex. Taking back Raqqa, and subsequently Deir al-Zour, would deal major blows to both the movement’s propaganda and operational capacity. Western intelligence sources assert Raqqa is a hub for ISIS external operations planning. Secretary Mattis reportedly has recommended a beefed-up variant of the Obama plan under which the U.S. would deploy additional troops to back an offensive on Raqqa by the Syrian Democratic Forces (SDF), a U.S.-supported group. No other force in Syria offers a better alternative: Turkish troops and their rebel allies fighting with the Euphrates Shield operation do not currently appear capable of taking Raqqa; an offensive by the regime and its allies – Hizbollah, Iran and Russia – or by Iraqi militias would be disastrous, provoking greater Sunni resentment.

The SDF option, however, raises its own problems, not least that its commanders and fighting core, the Kurdish People’s Protection Units (YPG), have direct operational links to the PKK, which Turkey and the U.S. designate a terrorist group. Turkish President Recep Tayyip Erdoğan – his repeated, if ambivalent, acquiescence to previous aspects of the U.S. plan notwithstanding – is already furious at Washington’s backing for the YPG and apparent failure to keep it east of the Euphrates.

Arming the group to assault Raqqa would further anger a NATO ally that has a critical role in the fight against ISIS and al-Qaeda. Turkey would fear that the YPG could win political capital from the West (indeed, for the Kurds, Raqqa’s only strategic value is as leverage) and divert U.S.-supplied weapons to the PKK in Turkey after the fight. Even leaving aside the longer-term pitfalls of alienating Ankara, Turkish forces and their rebel allies could escalate against the YPG and its local allies elsewhere in Syria, such as in Tel Abyad, which would force the YPG to redeploy fighters away from Raqqa. The U.S. has already felt the need to intervene around Manbij to prevent clashes escalating between Turkish-backed and YPG-backed forces, which both benefit from U.S. military support.

Nor should the U.S. enlist the YPG as the occupying force in Raqqa after an assault, particularly in light of its reported reprisals in some of the non-Kurdish towns already taken from ISIS. The city could not be handed over to the regime or its allies without further outraging its Sunni Arab inhabitants.

U.S. generals should also deepen the coordination they appear to have begun with Turkish and Russian counterparts to avoid clashes among their forces or proxies in Syria.

An alternative would be to slow the battle tempo to minimise the risk of aggravating other fronts in Syria’s war and push for the type of consensus the U.S. built ahead of the Mosul operation. However, if the White House presses ahead – motivated by fear of operations planned against U.S. interests by Raqqa-based militants – steps to mitigate Turkish concerns would be crucial. As U.S. officials have suggested, they could, for example, guarantee not to give the YPG heavy weapons, particularly advanced anti-tank systems; offer to help police the Turkish border; reiterate opposition to linking the three Kurdish cantons in northern Syria; and/or press the YPG to disassociate itself from the PKK militarily. U.S. generals should also deepen the coordination they appear to have begun with Turkish and Russian counterparts to avoid clashes among their forces or proxies in Syria. A realistic plan is also needed for holding Raqqa once ISIS is ousted; local tribes should police inside the city, even if the YPG provides perimeter security.

III. How to Fight Other Jihadists Without Creating Further Chaos?

Beyond fighting ISIS in its heartlands, the new administration confronts al-Qaeda-linked groups elsewhere. Even in Syria, al-Qaeda may pose a graver threat over time than ISIS. There, as in other Arab war zones, its “long game” strategy – embedding within popular uprisings, forming alliances with other armed groups and displaying some pragmatism and sensitivity to local norms – could prove more sustainable than that of ISIS. Picking a fight with everyone, as ISIS has discovered, travels badly outside Iraq.

Syria’s al-Qaeda branch, Jabhat Fatah al-Sham (JFS), formerly known as Jabhat al-Nusra, is the largest force in the newly-formed Hay’at Tahrir al-Sham (HTS) and among the most powerful armed groups in the north west.[fn]Though JFS has formally broken with al-Qaeda, it retains a close link to the movement.Hide Footnote The war’s evolution has worked in its favour. As violence escalated, it forged alliances, if often uneasy, with rebels. Its discipline and suicide bombers have meant that it often serves as shock troops during rebel offensives. Syrian President Bashar al-Assad’s policies – sectarian rhetoric, pitting Alawites and other minorities against the Sunni majority and collective punishment that razes cities – have also played into its hands.

After the rebels’ defeat in Aleppo at the end of 2016, the balance within the opposition, particularly around Idlib, has shifted further toward jihadists, as HTS has encroached on areas controlled by other rebels. To portray the north west as entirely al-Qaeda-run is wrong, however. Much of it is outside HTS control: rebels, particularly Ahrar al-Sham, a large Islamist force that confines its goals within Syria’s borders and has, in principle, accepted political and religious pluralism for a future Syria, controls parts. HTS, though JFS-dominated, includes diverse factions, some non-jihadist. Even JFS is heterogeneous, comprising a core of al-Qaeda and fighters with local motives.

Attacks that kill civilians […] bolster local support for al-Qaeda and undercut non-jihadist groups that portray the U.S. as a potential ally.

No present option against al-Qaeda in Syria is good. The new U.S. administration can continue targeting its leaders and external planning capacity, even as it supports the Raqqa offensive against ISIS further east. Strikes against targets in Idlib, which accelerated in the last months of the Obama administration, have killed dozens of al-Qaeda operatives but cannot reverse that movement’s expanding influence while conditions on the ground enable it. Attacks that kill civilians, as those in al-Jina, near Aleppo, on 16 March appear to have done, bolster local support for al-Qaeda and undercut non-jihadist groups that portray the U.S. as a potential ally.[fn]The U.S. military’s Central Command (CENTCOM) disclosed it conducted a raid against “a meeting location” in Idlib, Syria on 16 March, “killing several terrorists”. “U.S. forces strike Al Qaeda in Syria”, CENTCOM press release no. 17-104, 17 March 2017. A Pentagon spokesman initially told reporters officials thought there were “zero” civilian casualties, but after reviewing further information, the Pentagon launched a casualty “credibility assessment” to evaluate claims of civilian casualties. “Pentagon launches probe after strike near Syria mosque”, Agence France-Presse, 20 March 2017. Local media, activists and human rights groups estimated some 50 civilians were killed and dozens injured, with many more possibly trapped under rubble. Michael R. Gordon and Hwaida Saad, “U.S. military denies reports it bombed mosque in Syria”, The New York Times, 16 March 2017.Hide Footnote

Turkey can shape rebel dynamics in the areas held by Euphrates Shield forces east of Aleppo, where it intervened directly in part to prevent the YPG from connecting the non-contiguous cantons it controls. It has less interest in doing so within and adjacent to HTS strongholds in north-western Idlib province, which, with no Kurdish presence, have less strategic value and where an intervention could provoke an al-Qaeda backlash in Turkey.

An assault by the regime and its allies around Idlib is no solution either. It could weaken HTS temporarily but would ultimately play into al-Qaeda’s hands, stoking resentment and leaving the regime facing a war of attrition against a jihadist insurgency able to recruit from an angry population. Overall, the regime is no counter-terrorism partner in Syria. Even with Russian and Iranian support, it cannot secure the whole country, as shown by its inability to control Palmyra while simultaneously fighting to retake Aleppo. More importantly, its methods of prosecuting the war (use of indiscriminate weapons and targeting of civilians, hospitals and doctors, among others) bolsters the appeal of jihadists it claims to be fighting.

While the Assad regime, Iran and Hizbollah seem inclined to press their advantage, Russia appears to recognise that it and its allies cannot destroy all rebel forces.

Ultimately, the only way of sustainably eroding al-Qaeda’s influence in Syria is through a settlement between the regime and a non-jihadist opposition that has some ability to end violence on the ground. While the Assad regime, Iran and Hizbollah seem inclined to press their advantage, Russia appears to recognise that it and its allies cannot destroy all rebel forces. Shifting the balance in the north west away from HTS would require strengthening more pragmatic rebels and, where possible, peeling fighters with national goals away from al-Qaeda-linked groups. In other words, progress toward settlement, or at least sustained de-escalation, would require deeper U.S. cooperation with Turkey to get the opposition’s house in order and engagement with Russia and Iran. Though this seems remote for now, the Syrian war drives radicalisation across the region, and abandoning efforts to end it would leave a big gap in U.S. counter-terrorism strategy.

In Yemen, as in Syria, al-Qaeda has been a main beneficiary of the war.[fn]See Crisis Group Middle East and North Africa Report N°174, Yemen’s al-Qaeda: Expanding the Base, 2 February 2017.Hide Footnote Its local branch, al-Qaeda in the Arabian Peninsula (AQAP), was dangerous to the West but a sideshow in Yemeni politics until the state collapsed. In the aftermath of the 2011 Yemeni uprising, it established Ansar al-Sharia, parallel but aligned militias, to popularise the movement and lower the bar of entry for recruits. As fighting escalated in 2015 between the rebel Huthis and ex-President Ali Abdullah Saleh, on the one hand, and President Abed Rabbo Mansour Hadi and the Saudi-led coalition on the other, AQAP seized the Gulf of Aden port Mukalla and surrounding areas. It governed Mukalla via a council of local elders, placing less emphasis on enforcing its variant of Sharia (Islamic law) and more on providing water, electricity, dispute resolution and security. Conditions in Mukalla under AQAP rule were better than in many other Yemeni towns, helped by the fact that it was among the few the Saudi-led coalition did not bomb. Throughout the south, AQAP has positioned itself as protector against the Huthis.

Together with Ansar al-Sharia, AQAP now comprises thousands of fighters, embedded in the fabric of the anti-Huthi/Saleh alliance. It has acquired heavy weapons from Yemeni military camps or indirectly from the Saudi-led coalition, whose arms, supplied to a range of anti-Huthi groups, seep into al-Qaeda’s arsenal. Control of the port and emptying banks during its tenure in Mukalla have fed its coffers. While an Emirati-led, U.S.-supported campaign forced AQAP to withdraw from Mukalla in April 2016, the group still exercises on-again, off-again control of areas in Abyan and Shebwa governorates.

The Obama administration in 2016 killed dozens of AQAP members, including its leader, Nasir al-Wuhayshi, then al-Qaeda’s global number two. The recent uptick in U.S. operations, including a special forces raid in late January involving a firefight that left women, children and one U.S. marine dead, suggests the new administration will be still more aggressive.[fn]See, for example, Basma Atassi, Laura Smith-Spark and Angela Dewan, “Yemen raid: The plan, the operation, and the aftermath”, CNN, 9 February 2017.Hide Footnote

This approach carries risks. However many al-Qaeda members are killed and whatever intelligence is captured, harming civilians and deploying U.S. forces on the ground, particularly if they engage in sustained fighting, tend to be counterproductive, alienating communities and generating further support for AQAP and Ansar al-Sharia. Counter-terrorism operations risk complicating the Yemeni war and, ironically, strengthening the Huthi/Saleh bloc in areas where AQAP or Ansar al-Sharia are part of the fighting front against them. Regional and local allies may also try to exploit U.S. support for the fight against AQAP to target local opponents and, in the south, the mainstream Islamist movement Islah.

A few steps could help. Narrowing the range of targets to known AQAP leaders (rather than local Ansar al-Sharia fighters) and training camps, ensuring that each attack complies with domestic and international law and making further efforts to avoid harm to civilians would reduce chances of local backlash. In this respect, for the U.S. to loosen policies on the use of force for such operations would be a mistake.[fn]President Trump could do so in different ways. For instance, he could designate new “areas of active hostility (AAH)” under Obama’s May 2013 Presidential Policy Guidance (PPG). While all U.S. uses of force are subject to applicable law, including applicable international humanitarian and human rights law, the PPG’s strict targeting rules – which include high-level approval procedures – do not apply to strikes in AAH. See “Presidential Policy Guidance on Procedures for Approving Direct Action Against Terrorist Targets Located Outside the United States and Areas of Active Hostilities” (PPG), 22 May 2013; “Executive Order – United States Policy on Pre- and Post-Strike Measures to Address Civilian Casualties in U.S. Operations Involving the Use of Force”, 1 July 2016.  Some reports suggest Trump has already declared parts of Yemen AAH. Missy Ryan, Thomas Gibbons-Neff and Ali al-Mujahed, “Accelerating Yemen campaign, U.S. conducts flurry of strikes targeting al-Qaeda”, The Washington Post, 2 March 2017. A second way to relax policy would be to eliminate, or very loosely interpret, all or some of the PPG’s standards for use of lethal force outside AAH, including the requirement for a determination of “near certainty” that civilians will not be injured or killed, that capture or other non-lethal options are not feasible and that the target poses “a continuing, imminent threat to U.S. persons”. A last option would be to simply scrap the PPG and the associated executive order on pre- and post-drone strike procedures, including annual Pentagon reporting of strikes and civilian casualties outside AAH, and devise new policy. It should be noted that even though Obama’s PPG is widely regarded as providing important protections for civilians, it has also met criticism, including from human rights groups. See, for example, “US: Counterterrorism Report Sets Standards”, Human Rights Watch, 6 December 2016.Hide Footnote The fight against AQAP could focus on areas freed from Huthi/Saleh control, where local forces no longer rely on AQAP/Ansar al-Sharia militias as fighting partners. Ideally, recently formed militias that operate outside the law and often abuse local populations (like Aden’s Security Belt forces or Hadramout’s Elite forces) would be integrated into police and military units.

Most important, though, is not to abandon diplomatic efforts to end the war. Prospects in Yemen are better than in Syria, given U.S. influence on Saudi Arabia and the existence of a realistic UN roadmap that offers a framework for compromise. Helping Riyadh find a way out of an unwinnable war that empowers jihadists, increases Iran’s influence across its border and provokes humanitarian disaster should be the priority.

Steps likely to prolong the war, by contrast, should be avoided. Direct U.S. strikes against the Huthi/Saleh bloc or increased U.S. military assistance for operations against them, for example, would likely push the Huthis – who benefit from Iranian arms shipments but are potent on their own, are not Iranian proxies and have largely parochial interests – further into Iran’s orbit.

That would lead both sides toward greater escalation, with Iran upping support for the Huthis, dragging Saudi Arabia into a deepening quagmire, while feeding the illusion that the Saudis and their Emirati allies could end the conflict by heightening pressure on the Huthis.

A peace settlement along the lines of the UN plan would offer the Huthis a legitimate role in the country’s future

Such an escalation, by heightening sectarian polarisation and prolonging the war, would also play to jihadists’ benefit over time. The anti-Huthi/Saleh alliance is too internally fragmented and weak, even with more U.S. support, to decisively reverse Huthi/Saleh gains in the north while holding territory recaptured from al-Qaeda in the south. A peace settlement along the lines of the UN plan would offer the Huthis a legitimate role in the country’s future; that, plus the promise of Saudi and Gulf reconstruction assistance, would do more to pull them away from Tehran than a conflict that reinforces their mutual dependence and utility.

In Libya, jihadist groups are dangerous but for now less potent than in the Iraqi, Syrian and Yemeni war zones. Ansar al-Sharia groups, with loose ties to transnational jihadists, emerged after the 2011 war and ouster of Muammar al-Qadhafi; some members later joined ISIS, others joined militias that fought ISIS. Between August and December 2016, militias from the western town of Misrata ousted ISIS from a 120km coastal stretch it controlled around Sirte, killing many foreign fighters and scattering others, while locals mostly melted back into communities. The extent to which militants have drifted south to groups in the Sahel or southern Libya is unclear.

Critical in Libya is to resist the idea, promoted in part by Egyptian President Abdelfattah al-Sisi and the Emiratis, that General Khalifa Haftar can eradicate radical groups. While Haftar enjoys considerable support in eastern Libya, he – like the various forces in Syria and Yemen – cannot conquer the whole country, even with international backing. His opponents are too powerful and his support base too narrow.

Haftar’s track record against jihadists is also mixed. ISIS, for example, was ousted from Sirte not by him but by his Misratan opponents, who were closer and provoked by ISIS first. His forces did rout Ansar al-Sharia groups from Benghazi and inflicted a blow on ISIS militants there, but he alienated many non-jihadists in the process. Like his Egyptian and Emirati backers, Haftar tends to portray all Muslim Brotherhood-linked groups as terrorists, even though he aligns in some areas with more conservative Salafi militias. Haftar and his constituencies cannot be excluded from Libya’s political order, but backing him militarily in the hope that he can dominate it by force would be a mistake. Given the strength of his rivals and the support they enjoy from their own external backers, particularly Qatar and Turkey, it would escalate conflict, further destabilise the country and potentially open new opportunities for ISIS and al-Qaeda-linked groups that for now are largely contained.

Most dangerous is the Saudi-Iran rivalry, which has fed sectarianism and extremism on both sides of the Sunni-Shiite divide.

The region’s power rivalries overshadow its wars and complicate U.S. operations against jihadists. Most dangerous is the Saudi-Iran rivalry, which has fed sectarianism and extremism on both sides of the Sunni-Shiite divide. Iranian leaders, their perspective shaped by the traumatic war with Iraq in the 1980s – in which almost all Arab states and the U.S. backed Saddam Hussein – and the U.S. invasions of and continued military presence in neighbouring Afghanistan and Iraq, believe their country is encircled. Their rivals’ conventional military capacity dwarfs their own. Backing non-state actors and proxies across the region, in Tehran’s view, is a way to keep threats from its immediate borders.

Yet, what Tehran portrays as defensive appears as anything but to rivals. Major Sunni Arab states see Iran as a revolutionary power and reject the regional role to which Iran aspires and the influence it now wields, thanks largely to the U.S. invasion of Iraq and chaos in the aftermath of the Arab uprisings. The violence Sunnis have suffered at the hands of Iranian-sponsored governments and militias in Iraq and Syria has fed a profound and dangerous sense of victimisation among the region’s Sunni Arab majority and been a recruitment boon for jihadists. It has also incurred high costs for Tehran, by deepening regionwide Sunni animosity toward Iran, its allies and its proxies.

Gulf powers and Turkey, too, bear much responsibility. Their oversight of arms poured into Libya, Syria and Yemen has been inadequate, much ending in jihadists’ hands. Sunni militants of all stripes – not just jihadists – have committed their own atrocities against Shiites. Sectarian rhetoric has been far too common. Exclusionary and repressive policies in Bahrain inevitably have also exacerbated sectarian tensions. Ultimately, all prioritise enemies other than jihadists: in Saudi Arabia’s case, Iran; in the UAE’s, Iran and the Muslim Brotherhood; in Turkey’s, the PKK/YPG. All this has opened space for ISIS and al-Qaeda.

The U.S. appears set to deal with this unfavourable regional context by bolstering ties to traditional Gulf allies – augmenting weapons sales and working in concert with Gulf states on a more muscular approach toward Iran. Providing extra hardware would carry drawbacks, given the weapons proliferation in the region, the economic challenges faced by Gulf monarchies in a time of lower oil prices and the often indiscriminate conduct of the Yemen campaign. Any more confrontational stance would also risk an asymmetrical Iranian response through non-state allies across the Middle East and Afghanistan, a dangerous dynamic that could provoke a military conflagration. It also could put Iraqi Prime Minister Abadi in a bind, as he could ill afford to side with the U.S. in a confrontation with his powerful neighbour.

Diplomacy, by helping to pacify the region’s conflicts, would do as much or more to counter jihadism as any military operation.

That the Trump administration would seek to shore up alliances with traditional Gulf partners in the wake of relative estrangement under Obama is reasonable. But backing should neither be unconditional nor enable a Saudi quagmire in Yemen or a risky escalation with Iran, both of which could further destabilise the region. An alternative would be to use the leverage of improved relations, first, to ensure the Saudi-led coalition prosecutes the war in compliance with international law and, secondly, to press for de-escalation of Iranian-Saudi hostility, in particular through a Yemeni settlement, lessening of sectarian rhetoric, a more inclusive approach in Bahrain and resumption of dialogue between Riyadh and Tehran. Diplomacy, by helping to pacify the region’s conflicts, would do as much or more to counter jihadism as any military operation.

IV. What Direction for South Asia Policy?

Outside the Middle East, South Asia is the region most critical for U.S. counter-terrorism policy, particularly as the centre of gravity of global jihadism over past decades has swung between there and the Arab world. Bar brief mentions of Afghanistan and Pakistan during Secretary Mattis’s confirmation hearing, the new administration has given little sense of its direction for the region.

In Afghanistan, the Taliban is stronger than at any point since its ouster in 2001. Internal UN estimates suggest it controls more than half the countryside.[fn]Crisis Group interview, UN official, March 2017.Hide Footnote In summer 2016, it briefly captured Kunduz, a provincial capital in the north east. As weather warms, it will again threaten that town and other provincial capitals. It mounts sophisticated offensives, deploys mobile columns across front lines in Humvees and confronts Afghan army and police units directly.

The Taliban’s ties to al-Qaeda stretch back decades. According to U.S. officials, al-Qaeda operatives use Taliban training camps to plot operations across South Asia.[fn]See, for example, Brian Dodwell and Don Rassler, “A View from the CT Foxhole: General John W. Nicholson, Commander, Resolute Support and U.S. Forces-Afghanistan”, CTC Sentinel, 22 February 2017.Hide Footnote Senior Taliban leaders, however, have distanced themselves from global jihadism in dealings with the U.S. and states in the region. Their focus is on regaining power in Afghanistan.

A local ISIS branch operates in remote eastern districts. It is deeply anti-Shiite, conducts attacks that kill many civilians and comprises mostly former Pakistani tribal militants, with some local recruits. Since its establishment in 2015, its growth has largely been checked by Taliban operations and U.S. airstrikes, though attacks, including on Shiite Hazaras in 2016, suggest growing potency. The Taliban, however, is by far Afghanistan’s largest armed opposition group.

Pakistan hosts the Taliban’s leadership. Afghan-Pakistani relations are badly strained: President Ashraf Ghani initially tried to strengthen ties to Pakistani leaders hoping they would bring the Taliban to peace talks but now accuses Islamabad of conducting war in Afghanistan.[fn]At the June 2016 NATO summit, Ghani said that Pakistan had imposed an “undeclared war” on Afghanistan. “Afghanistan’s Ghani urges Pakistan to expel insurgents from its soil”, Voice of America, 11 July 2016.Hide Footnote Closer Indian-Afghan ties appear to have deepened the Pakistani military’s long-held view that the Taliban safeguards Islamabad’s national security interests. Successive Afghan governments’ failures, indiscriminate U.S. counter-terrorism operations and local strongmen’s manipulation of those operations to defeat rivals have helped fill its ranks, but the Taliban could not maintain its potency without Pakistani sanctuaries.

The Taliban has built ties to other governments, too. Iran bitterly opposed its rule in Afghanistan in the 1990s but more recently has backed Taliban insurgents, initially to pressure U.S. forces in western Afghanistan and now to support their fight against ISIS. The Russians talk to its leaders also, to share, in Moscow’s words, intelligence against ISIS and in the hopes of laying the groundwork for future talks between the Taliban and Afghan government.[fn]See, for example, Mehmet Ozturk, “Exclusive interview with Russian diplomat Zamir Kabulov”, Anadolu Agency, 31 December 2016.Hide Footnote

Troop increases requested in February by the U.S. commander in Afghanistan, General John W. Nicholson, would help the Afghan army hold the line against insurgents but not decisively tip the balance. The Taliban has weathered far larger numbers of U.S. forces during Obama’s first-term surge. Here, too, diplomacy is as vital as military support. The Afghan power-sharing government is dysfunctional, with friction mounting between President Ghani and Chief Executive Officer Abdullah Abdullah. U.S. engagement with and support for the government would help avert a crisis ahead of parliamentary elections expected in 2018 and a 2019 presidential contest that could further fracture the country and facilitate Taliban gains.

… the new administration should prioritise reopening publicly acknowledged lines of communication to Taliban leaders and rethinking a format for regional engagement.

Nor can the U.S. exit without diplomacy. The only way of withdrawing forces without leaving a haven for al-Qaeda or other transnational groups is through a settlement with the Taliban that, first, requires it to announce it has severed links with international jihadists and respects the Afghan constitution and, secondly, meets neighbours’ core concerns. Though recent Russian-brokered talks brought together neighbours and the Afghan government, serious progress is unlikely without a U.S. lead: the new administration should prioritise reopening publicly acknowledged lines of communication to Taliban leaders and rethinking a format for regional engagement. Sending more U.S. troops only makes sense as part of a political strategy that pushes toward a settlement, however remote that currently seems.

Pakistan poses further dilemmas. Not only does peace in Afghanistan hinge on its military establishment helping bring the Taliban to the table; the country also faces its own multipronged threat from tribal, sectarian and anti-India jihadists, some with old al-Qaeda ties.[fn]Crisis Group Asia Report, N°217, Revisiting Counter-terrorism Strategies in Pakistan: Opportunities and Pitfalls, 22 July 2015.Hide Footnote Anti-Shiite groups such as Lashkar-e-Jhangvi have recently forged an alliance with ISIS (reportedly its Middle East-based leadership), with both claiming credit for sophisticated attacks.[fn]See, for example, Mubashir Zaidi, “IS recruiting thousands in Pakistan, govt warned in ‘secret’ report”, Dawn, 8 November 2014.Hide Footnote

The military has in recent years cracked down on militants that attack the Pakistani state. Operations in the tribal areas along the north-west border with Afghanistan have dispersed disparate tribal militants and foreign jihadists sheltering there since fleeing Afghanistan in 2001. Offensives have often been brutal and displaced the problem rather than resolved it; militants have already begun to regroup and resume attacks countrywide, claiming hundreds of lives in 2017. Introducing civilian governance and policing is the only way to stabilise the tribal areas. Together with years of U.S. drone strikes, however, operations have meant they no longer serve as a base for al-Qaeda’s leadership to the same degree as a decade ago.

The two main anti-India groups, Lashkar-e-Tayyaba (LeT) and Jaish-e-Mohammad, enjoy considerable operating space, with their relief wings distributing aid, madrasas functioning and leaders preaching openly. Though neither has formal links to al-Qaeda, their fighters rub shoulders with other militants and global jihadists in Afghanistan and Pakistan. The gravest danger they pose for Pakistan and the U.S. is another strike on India. Prime Minister Narendra Modi’s response to attacks last year on Indian forces in Kashmir suggests his reaction would be calibrated carefully, and public opinion would weigh only so far on that calculation. But it would be difficult to show restraint in the event of an attack like that which killed large numbers of civilians in Mumbai in 2008.

Pakistan’s jihadist problem, if largely of its own making, is deeply entrenched. That Afghan Taliban leaders who talk to the U.S. or Afghan government without Pakistani blessing are promptly jailed or disappear shows how the military can clamp down. Only a strategic rethink of relations with India, however, would lead it to dismantle the LeT’s and Jaish’s Punjab-based infrastructure.

The main challenge for the U.S. is to persuade the military establishment to push the Taliban toward talks and act against anti-India groups. Inducements to military leaders, including strategic dialogue and extra aid in the early years of the Obama administration, did not shift its strategic calculation. Wielding a larger stick, for which there is some support in Congress, would be a new tactic, though U.S. military leaders would likely have little appetite to exert significant pressure on Pakistani counterparts. Blank checks in the past, however, have produced at best selective counter-terrorism cooperation. U.S. national security interests would be best served by a multipronged policy: conditionality on aid to the military; technical assistance for civilian law enforcement and intelligence agencies; and continued support for a democratic transition that is incrementally empowering a Pakistani political leadership less prone to see jihadists as strategic assets.

Pressing and persuading Pakistan to do more against its militant proxies also requires U.S. cooperation with China. Beijing fears jihadism as much as the U.S., and its proximity to and growing economic cooperation in the region give it more to lose from Afghan instability. The web of trade routes it funds across South and Central Asia could be a geopolitical game changer for the region. Without its support, the U.S. will struggle to extract more constructive policy from Islamabad. This makes the administration’s initial hostility to China all the riskier.

V. Defining the Enemy

A last question for the new administration is whom to fight. Where will it draw the line on which Islamists are the enemy? Secretary Mattis and National Security Advisor H.R. McMaster in the past have been pragmatic, particularly in Iraq, where they dealt with diverse politicians, including Islamists. Secretary of State Rex Tillerson, on the other hand, has argued that defeating the Muslim Brotherhood is as much a priority as defeating al-Qaeda.

Defining the enemy applies first on the battlefield, particularly where jihadists fight beside other militias, whether in Libya, Syria or Yemen. These alliances tend to be tactical: jihadists provide extra firepower against a shared enemy. They rarely signal wider support for aims to strike the West or establish a caliphate. U.S. interests would be best served by defining the enemy narrowly and aiming to change conditions on the ground to prompt other armed groups to break ties with jihadists. Ideally this would involve de-escalating the conflicts that motivate those alliances, but even without that, there may be ways to pull groups with national goals and a willingness to coexist with rivals away from transnational jihadists. Outreach to such groups by the U.S. or its allies – similar to outreach to Sunni tribal leaders ahead of the Awakening and U.S. surge in Iraq – could occur even alongside attacks on al-Qaeda leaders.

Identifying the aims of militias across the Muslim world’s war zones is, of course, hard. Fighters with links, however loose, to jihadists pervade armed groups of all stripes. Few powerful militia leaders champion liberal values or tolerance, even where they espouse national goals or accept power sharing. The perceived failure, over past decades, of secular ideologies and the flow of Gulf funding, combined with severe violence and repression, have empowered few moderates. But the Trump administration should be realistic. Many militants have now rubbed shoulders with al-Qaeda; many espouse anti-U.S. sentiment. The U.S. cannot declare them all beyond the pale if it hopes to influence decisively the wars they fight in.

A sensible position on mainstream Islamists is especially critical. Designating the Muslim Brotherhood terrorist, for example, would backfire. The movement espouses some illiberal and intolerant ideas. Since President Mohamed Morsi was deposed in Egypt, younger Brotherhood members, facing a brutal crackdown, have been implicated in attacks against the Egyptian state, even if the movement’s leaders reject their violence.

Overall, however, the Brotherhood has explicitly distanced itself over past decades from the thinkers that inspire al-Qaeda and ISIS. Its political Islam is perhaps jihadists’ main ideological competitor; ISIS and al-Qaeda propagandists reserve particular venom for its gradualism and electoral participation. They portray that strategy’s failure as vindication of their violence. Over recent years, jihadists’ fortunes have tended to wax as those of mainstream Islamists have waned.

There are other challenges, too. Members of Muslim Brotherhood offshoots sit in the cabinets and parliaments of staunch U.S. allies like Morocco, Tunisia and Jordan, whose support is critical against ISIS and al-Qaeda. Elsewhere – in Syria and Yemen, for example – militias linked to the Brotherhood fight beside U.S. allies. Other allies, like Turkey and Qatar, host exiled leaders. Designating the movement would also play dangerously into rivalries between Turkey and Qatar, which are sympathetic to it, and the UAE and Egypt, which view it as a threat. Where those rivalries play out through proxies, designation would pick a side, encouraging anti-Islamist forces, like those of Haftar in Libya, to double down.

Designation would not necessarily impel Muslim Brotherhood leaders toward violence, but it would narrow the movement’s options and potentially increase the anti-U.S. sentiment of members. It would play into jihadist narratives, already reinforced by some of President Trump’s rhetoric and his immigration policies, that peaceful resistance and accommodation with the West are futile. While little suggests the new administration has either the leverage or the inclination to shift the Egyptian or Emirati line on the movement, it should at least not buy into the same logic. Picking a fight with the Muslim Brotherhood makes no strategic sense for the U.S.

VI. Conclusion

Since the 9/11 attacks in 2001, the war against jihadists has dominated U.S. national security policy. The aggressive operations that look set to mark President Trump’s foreign policy do not in themselves signal major departures. Reversing al-Qaeda’s and ISIS’s gains and protecting U.S. citizens from their attacks should, of course, be imperative for U.S. leaders. But for the last decade and a half, too great a focus on counter-terrorism has often distorted U.S. policy and, in many cases, made the problem worse.[fn]Crisis Group Report, Exploiting Disorder, op. cit.Hide Footnote The new administration’s elevation of the threat, combined with the damaging anti-Muslim language of some in Trump’s inner circle and immigration policies that appear discriminatory, makes risks all the graver.

Some early signs are particularly troubling. Loosening procedures that protect against civilian casualties during targeted killings would be a serious mistake. Such killings in any case have a mixed record: repeated strikes against al-Qaeda commanders in Somalia, Syria and Yemen have not inhibited their movements’ growth; often harder-line leaders replace those killed.[fn]Ibid.Hide Footnote Invariably they are counterproductive, and potentially illegal, if they kill civilians and, with that, anger local communities as well as partners and allies. Even small numbers of civilian casualties can complicate the fight against jihadists. Overlooking allies’ harmful policies or their potential misuse of counter-terrorism operations against rivals is also a danger and could deepen chaos in the region or even provoke a wider conflagration. So, too, could an escalation against Iran.

Especially troubling is the apparent neglect of diplomacy, which is critical for navigating the rivalries among states in parts of the world most affected and forging solutions to the wars jihadists feed off. Staffing the State Department’s top levels; maintaining a deep bench of expertise at both State and the U.S. Agency for International Development (USAID), which plays a vital role in preventing and mitigating violence and helping communities recover; and maintaining both their budgets are critical to U.S. soft power and should be priorities. Cutting support to the UN would hinder efforts against jihadists, potentially undermining its critical peace-making and peacekeeping, coordination of reconstruction funds in places like Iraq, humanitarian support to sustain communities in war zones and its forum for counter-terrorism coordination.

In the words of the U.S. Counterinsurgency Field Manual that Secretary Mattis co-authored: “The military contribution to countering insurgency, while vital, is not as important as political efforts for long term success”.[fn]“Counterinsurgency Operations”, Joint Chiefs of Staff Joint Publication 3-24, 5 October 2009, p.III.3.Hide Footnote Or in his own words as still a general, “… if you don’t fund the State Department fully then I need to buy more ammunition ultimately”.[fn]Hearing, Armed Services Committee, U.S. Senate, 5 March 2013, p. 16.Hide Footnote Fighting terrorists without diplomats, in other words, is a fool’s game.

The new administration’s focus on degrading groups that plot attacks against the U.S. and its citizens is understandable. But in doing so, it must avoid inadvertently creating further disorder that plays into jihadists’ hands.

Washington/New York/Brussels, 22 March 2017

Appendix A: Map of North Africa, the Middle East and South Asia

Map of North Africa, the Middle East and South Asia International Crisis Group, March 2017. Based on UN Map No 4170, Rev. 13, 2012.
Shiite Huthi rebels drive a truck past a flag of Ansar al-Sharia in Almnash, Rada, which was once the main Yemeni stronghold of this local arm of Al-Qaeda in the Arabian Peninsula. 22 November 2014. REUTERS/Mohamed al-Sayaghi
Report 174 / Middle East & North Africa

Yemen’s al-Qaeda: Expanding the Base

Thriving on conflict, sectarianism, and local opportunism, al-Qaeda’s affiliates are stronger than ever in Yemen. To shrink their growing base will require better governance in vulnerable areas, not treating all Sunni Islamists as one enemy, and above all ending Yemen’s civil war.

  • Share
  • Save
  • Print
  • Download PDF Full Report

Executive Summary

The Yemeni branch of al-Qaeda (AQ) is stronger than it has ever been. As the country’s civil war has escalated and become regionalised, its local franchise, al-Qaeda in the Arabian Peninsula (AQAP), is thriving in an environment of state collapse, growing sectarianism, shifting alliances, security vacuums and a burgeoning war economy. Reversing this trend requires ending the conflict that set it in motion. This means securing an overarching political settlement that has buy-in from the country’s diverse constituencies, including Sunni Islamists. As this will take time, steps must be taken now to contain AQAP’s growth: improving governance in vulnerable areas, disaggregating Sunni Islamist groups and using military tools judiciously and in coordination with local authorities. These efforts will be imperilled if states interested in fighting AQAP and Yemen’s nascent Islamic State (IS) branch, such as the U.S., take military actions that ignore the local context and result in high civilian casualties, like the Trump administration’s 29 January 2017 raid on AQAP affiliates in al-Bayda, or fail to restrain partners who tolerate or even encourage AQAP/IS activities.

Prior to Yemen’s 2011 popular uprising against President Ali Abdullah Saleh, AQAP was a small yet lethal branch of AQ, focused primarily on Western targets. With at most several hundred members, it had limited local appeal and was both sustained and constrained by complex and sometimes contradictory relationships with the governing authorities and tribes. A primary security concern for the West and especially the U.S., AQAP was a sideshow for most Yemenis, at times tolerated by the government and routinely used by local elites for financial and/or political advantage. It was far less threatening to state stability than growing regime infighting, southern separatist sentiment or Huthi militancy in northern areas.

AQAP and, later and to a much lesser extent, a new outcrop of IS, emerged arguably as the biggest winners of the failed political transition and civil war that followed. AQAP adapted to the rapidly shifting political terrain, morphing into an insurgent movement capable of controlling territory and challenging state authority. Its main success derives from its demonstrated pragmatism: working within local norms, forging alliances with Sunni allies, assimilating into militias and embedding itself in a political economy of smuggling and trade that spans the various fighting factions, including the Huthi/former President Saleh alliance. It has at times controlled territory in the country’s south and appears ever more embedded in the fabric of opposition to the Huthi/Saleh alliance, dominant in the north, that is fighting the internationally recognised, Saudi-backed interim government of President Abed Rabbo Mansour Hadi.

IS, with its more brutal tactics, has been less successful in gaining recruits or capturing territory, but war has opened space for it to operate in places that have experienced sectarian-tinged violence, such as the southern port city of Aden. There, the group has turned its sights on the Hadi government and local security personnel through assassinations and bombings that have, indirectly, benefited the Huthi/Saleh front by weakening its common enemies and repeatedly underscoring the lack of security in Aden, the government’s temporary capital.

Virtually all local and foreign fighting parties in Yemen claim to be enemies of AQAP and IS, yet all have contributed to their rise. The Huthis, who as Zaydi/Shiites are AQAP’s primary ideological enemies, strengthened their foes through their February 2015 military push into predominantly Shafai (Sunni) areas, allowing AQAP to present itself as part of a wider “Sunni” front against Huthi/Saleh expansion. The Huthi/Saleh bloc’s willingness to conflate the Sunni Islamist party Islah and southern separatists with AQ and IS does not help. Their opponents, especially a gamut of Salafi fighting groups that the war has pushed to the foreground, as well as their Gulf backers, have poured fuel on the fire, at times crudely labelling Huthis as Iranian proxies who are part of a “Shiite agenda” in the region.

The logic of “the enemy of my enemy is my friend”, coupled with a long legacy of politicians using jihadists in power struggles against foes, has allowed AQAP to forge tacit alliances with a range of anti-Huthi/Saleh forces. The Saudi-backed coalition’s almost single-minded focus on defeating the Huthi/Saleh bloc has been a boon to AQAP, which has controlled territory unimpeded for stretches of time, in the process indirectly gaining weapons from the coalition and mining new funding streams by raiding banks and controlling ports. The United Arab Emirates dislodged AQAP from its Mukalla stronghold in April 2016, but such successes are fragile and could easily be reversed in the absence of more effective and inclusive governance.

The evolution of AQAP into an insurgent force with the ambition and capacity to govern territory, showing pragmatism and sensitivity to local concerns, does not negate the international risk posed by the group. AQ’s long-game strategy, combined with the immediate benefits from Yemen’s war, means that it, along with its local affiliates, will likely outlast the swift global rise of IS and its Yemeni subsidiary, which has pursued a more aggressive approach. The continuation of an increasingly fractured conflict greatly enhances AQAP’s unprecedented ability to expand local support and amass financial and military resources. Countering its gains poses a complex long-term challenge and will require an urgent yet measured response, focused on bringing the civil war to a negotiated end.

Recommendations

To reverse AQAP/IS gains

To all Yemeni and regional belligerents:

  1. End the war by agreeing to a ceasefire followed by negotiations toward a political settlement that contains: 
    1. buy-in from a full range of Yemeni stakeholders, including Sunni Islamists (the Islah party and Salafi groups willing to participate in politics) and groups with a regional base, such as Hiraak in the south;
    2. recognition of the need for regional autonomy, particularly for the south, and creation of a mechanism to determine the future state structure; and
    3. interim security arrangements in various war-torn localities under the state umbrella but with local buy-in.
       
  2. Avoid sectarian language and end media campaigns and mosque sermons that label adversaries in sectarian terms. 

To donor governments assisting Yemenis in combatting AQAP/IS:

  1. Engage in regular assessments of local and regional partners who may at times tolerate or even encourage AQAP/IS activities for political or economic gain, and press them to change course, threatening to suspend counter-terrorism cooperation if they do not.
     
  2. Decouple development from counter-terrorism assistance to reduce the incentives for the (current or future) Yemeni government to benefit financially from AQAP/IS’s presence.
     
  3. Enhance security measures at ports and border crossings with an increased maritime security focus on AQAP/IS sea supply routes along vulnerable coastlines.
     
  4. Encourage and support Track-II and local civil society efforts to heal inter-confessional divides, building on Yemen’s history of tolerance.
     
  5. Where there are opportunities to open lines of communication with AQAP leaders independent of tribal or political elites, those should be explored and if possible used to help de-escalate violence.

To states and groups operating in areas previously under or vulnerable to violent jihadist control, especially, but not limited to, the Hadi government, government-linked militias and the United Arab Emirates:

  1. Prioritise basic security, justice – particularly quick and transparent dispute resolution – and service provision.
     
  2. Disaggregate rather than conflate various Sunni Islamist groups by:
     
    1. including Islah in local governance and security initiatives; and
    2. communicating and negotiating with supporters of Ansar al-Sharia (AQAP’s local insurgency arm), who may not adhere to AQAP’s global ideology, and work to separate them from AQAP by addressing their legitimate locally- grounded grievances.
       
  3. Use military and policing tools judiciously and in compliance with local laws and norms by:
     
    1. avoiding heavy-handed military campaigns in cities and, when possible, working with local leaders to negotiate violent jihadists’ exit from urban areas, as happened in Mukalla; and
    2. using local forces against AQAP/IS when possible, but without creating legally unaccountable militia structures outside the state’s umbrella; bringing local militias, including popular committees, the Security Belt forces and the Elite forces in Hadramout, fully under government authority and under a legal system that ensures transparency and protects human rights.

To the Huthi/Saleh bloc:

  1. Disaggregate rather than conflate various Sunni Islamist groups, and work with those willing to engage in peace talks and operate within the political process.
     
  2. Refrain from military advances into predominately Shafai/Sunni areas that can only further inflame growing sectarian tensions and provide fodder to AQAP/IS propaganda.

Brussels, 2 February 2017

I. Introduction

This report examines the nexus of regional and local factors fuelling al-Qaeda in the Arabian Peninsula (AQAP) and Islamic State (IS)’s gains in Yemen. It builds on Crisis Group’s comparative study of the evolving global jihadist landscape, Exploiting Disorder: al-Qaeda and the Islamic State, by exploring the case of Yemen as a subset of this milieu.[fn]This report will follow the use of the term “jihadist” as explained in Crisis Group Special Report N°1, Exploiting Disorder: al-Qaeda and the Islamic State, 14 March 2016: “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’…. 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 …. 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”. As in the Exploiting Disorder report, this report examines only a subset of jihadist groups, namely Sunni jihadists, in this case, AQAP and IS in Yemen.Hide Footnote In many ways, AQAP’s and IS’s rapid growth in Yemen follow regional trends. The collapse of Yemen’s Arab Spring transition and the chaos that followed have catalysed their expansion, providing them with new political opportunities, money, weapons and recruits. As in Syria, Iraq and Libya, growing enmity between regional states, mainly Saudi Arabia and Iran, has fuelled sectarian tensions and led them to prioritise traditional rivals over violent jihadists, in some cases leveraging the latter as proxies.

Yet, the challenge that AQAP and IS present in Yemen is embedded in the country’s unique history and local political dynamics that both sustain and limit them. A record of regime co-optation of and collaboration with jihadist groups means that AQAP in particular is already intertwined with political actors and integrated into the economy. This creates obstacles in suppressing the group in that these actors may have incentives to use it to advance their own political and economic interests. Yet as AQAP is a Yemeni organisation with legitimate local demands – justice provision, services, jobs – efforts could be made to co-opt it and weaken its transnationally focused leadership by addressing these local grievances. Devising effective policy options for countering AQAP – or IS – cannot be based on a cookie-cutter approach but requires attention to both regional factors and local idiosyncrasies, lest the problem be aggravated, not overcome.

II. Al-Qaeda in Yemen

Yemen has long grabbed headlines as a hotbed for al-Qaeda (AQ) activity, and indeed it holds a special place in jihadist eschatology.[fn]According to a hadith (saying of the Prophet Mohammed), during the end of days, “Out of Aden-Abyan will come 12,000, giving victory to the [religion of] Allah and His Messenger. They are the best between me and them”. Musnad Imam Ahmad bin Hanbal hadith collection, vol. 3, no. 379.Hide Footnote Its common caricature is as the Middle East’s “wild west”, where gun-toting tribesmen, rugged mountains, weak governance and a deeply religious, rural population offer a breeding ground for outlaw groups. This stereotype not only risks oversimplification but can result in incorrect assumptions about AQ (that tribal areas necessarily provide safe haven, government actors are automatically the group’s foes or AQ cannot thrive equally in urban areas) and problematic, even counterproductive, policy prescriptions.[fn]“A False Foundation? AQAP, Tribes and Ungoverned Spaces in Yemen”, Combating Terrorism Centre, September 2011.Hide Footnote

Western analysis tends to explore AQ’s relationship with local tribes but less often examines the group as a tool for Yemen’s political elite to resort to subterfuge for financial and military gain.[fn]Gregory D. Johnsen, The Last Refuge: Yemen, al-Qaeda, and America’s War in Arabia (New York and London, 2012), and Sarah Philips, “The Norm of State-Monopolised Violence from a Yemeni Perspective”, in Charlotte Epstein (ed.), Against International Relations Norms (forthcoming, 2017), are notable exceptions.Hide Footnote Yemenis, by contrast, view domestic political dynamics as fundamental to understanding and countering AQ and similar jihadist groups.[fn]Yemenis generally identify three factors in facilitating Sunni radicalisation and AQAP expansion. The spread of ideas from the Muslim Brotherhood in Egypt in the 1960s and 1970s, and from Saudi Arabia through support for religious schools in the 1980s and the return of Yemeni migrant workers in the early 1990s, allegedly formed an ideological foundation for groups such as AQAP. The second factor is poverty. Most important for many interlocutors is state orchestration of jihadist groups for political and financial gain. Crisis Group interviews, Yemeni politicians, journalists and analysts, September 2010-August 2016.Hide Footnote

The history of AQ and related movements in Yemen is tied to both domestic politics and shifting trends in global jihadism. In the early 1990s, fighters from the anti-Soviet Afghan jihad, known as Afghan Arabs, returned – as part of the first wave of global jihadist violence after the end of the Cold War – just as north Yemen (the Yemen Arab Republic) unified with the socialist south (the People’s Democratic Republic of Yemen, PDRY) to form the Republic of Yemen.[fn]“Afghan Arab” is used to describe the non-Afghan Muslims from Arab, and some non-Arab, nations who travelled to Afghanistan to fight against the Soviets. Those who subsequently travelled to Yemen included non-Yemenis as well as returning nationals.Hide Footnote While most Arab states were turning against Islamists, Sanaa continued to align with them.[fn]Crisis Group Middle East Report N°8, Yemen: Coping with Terrorism and Violence in a Fragile State, 8 January 2003.Hide Footnote Islah, a Sunni Islamist party created in 1990 and encompassing the Yemeni Muslim Brotherhood, provided a political outlet for many returnees as it formed a governing coalition with President Ali Abdullah Saleh’s General People’s Congress (GPC) following elections in 1993. More militant Afghan veterans – such as Tariq al-Fadhli – joined with southerners to form the Islamic Jihad Movement (IJM). On 29 December 1992, IJM leader Jamal al-Nahdi attempted to kill U.S. marines in Aden, the first AQ-linked attack targeting the U.S.[fn]IJM was the first formal Islamist group in Yemen with ties (financial and personal, not organisational) to Osama bin Laden and AQ in Afghanistan, known as “al-Qaeda Central”. Bin Laden later took credit for the failed bombing. Al-Nahdi was arrested but escaped from prison.Hide Footnote

During this time, jihadists and the Saleh regime were aligned against the Yemeni Socialist Party (YSP). Prior to the 1994 north-south civil war, Afghan Arabs allegedly killed YSP cadres with the help of northern-linked security services.[fn]The killings – at least 150 between 1990 and 1994, according to the YSP – were a key instigator of the war. Brian Whitaker, The Birth of Modern Yemen (e-book, 2009); and Noel Brehony, Yemen Divided: The Story of a Failed State in South Arabia (London, 2011).Hide Footnote In the war, Saleh and his senior military commander, Ali Mohsen al-Ahmar (who married al-Fadhli's sister and today is Yemen’s vice-president aligned with the Saudi-led coalition confronting the Huthi/Saleh alliance), used the Afghan Arabs as a proxy. Following a quick and decisive northern victory, some IJM members were given positions in the GPC and security services.[fn]During the 1994 war, al-Fadhli was made a colonel in the Yemen army. After the civil war, al-Nahdi joined the GPC’s governing organisation, the permanent committee, and Saleh appointed al-Fadhli to the upper house of parliament. Gregory D. Johnsen, The Last Refuge, p. 51.Hide Footnote  

In the mid-1990s, a second wave of global jihadist violence saw AQ focus on attacking what it called the “far enemy”. In Yemen, AQAP’s precursors started doing the same. IJM remnants and others who refused co-optation by the state gathered under Yemeni Afghan veteran Zain al-Abidin Abubakr al-Mihdar to create the Aden-Abyan Islamic Army (AAIA), the first jihadist group with a transnational agenda. It pledged support to Osama bin Laden and engaged in international messaging.[fn]The AAIA, probably created between 1994 and 1997, was known under various names inspired by a hadith, including the Army of Aden-Abyan and the Islamic Army of Aden. In a message to Agence France-Presse (AFP) in August 1998, it replicated bin Laden’s declaration calling for “total war” on U.S. interests in Yemen. AQAP propaganda referred to the AAIA in 2010.Hide Footnote

Although AAIA’s influence dwindled in the late 1990s, jihadist attacks on Western interests increased. On 12 October 2000, an explosives-laden skiff rammed into the USS Cole, a U.S. warship docked off Aden port, killing seventeen marines. AAIA claimed responsibility, but the mastermind behind the attack was Abd-al-Rahim al-Nashiri, a Saudi AQ member who was bin Laden’s head of operations in the Gulf. The attack propelled Yemen into the spotlight as a critical state in the U.S. fight against AQ. Under U.S. pressure, Yemeni authorities rounded up scores of suspects – but not al-Nashiri.[fn]Four months later, Saudi officials reportedly spotted al-Nashiri in Sanaa with the deputy director of the PSO. “Yemen, an uneasy ally, proves adept at playing off old rivals”, The New York Times, 19 December 2002. Ahmed Abdullah al-Hasani, head of the navy at the time of the Cole attack, said its perpetrators were “well known by the regime and some are still officers in the national army”. “Britons’ killers ‘linked to Yemeni army chief’”, The Sunday Times, 8 May 2005.Hide Footnote Following the 9/11 attack in the U.S. and to avoid political isolation, Saleh moved decisively against AQ, largely defeating it by the end of 2003.

That year, the U.S. invasion of Iraq inspired a new generation of fighters (a third wave of global jihadism) that revived and altered AQ’s Yemeni branch. In February 2006, 23 AQ members escaped Sanaa PSO prison. Among them were Nassar al-Wuhayshi and Qasim al-Raymi, both of whom later became AQAP founding members.[fn]Several prison officials, including warden Salah al-Muradi, were arrested on suspicion of facilitating the escape. “Yemen: Al-Qa'ida escape”, U.S. embassy Sanaa cable, April 2006, as published by WikiLeaks, https://wikileaks.org/plusd/cables/06SANAA973_a.html. The timing raised questions, as Yemen had just lost $300 million of aid from the U.S. and the World Bank.Hide Footnote The event once again focused U.S. attention on the country as a front-line state against AQ. In 2007, funding from the U.S. Defense Department to Yemen increased to $26 million from $4.3 million the previous year.[fn]U.S. funding soared to $67 million in 2009 following two attacks on the U.S. embassy in Sanaa in 2008, and to $155.3 million in 2010, which U.S. officials said was largely due to AQAP’s attempted bombing of the U.S.-bound passenger jet on Christmas Day, 2009. “Yemen: Background and U.S. Relations”, Congressional Research Service Report for Congress, 1 November 2012. “Report to Congressional Committees”, U.S. Government Accountability Office, February 2012. This pattern of increased funding in response to attacks demonstrated the financial and military gains the state could make from AQ’s activities.Hide Footnote Yet, between 2007 and 2009, Sanaa and Washington were distracted by, respectively, the conflict with the Huthis and the war in Iraq, allowing a new generation of AQ leaders under al-Wuhayshi to rebuild the organisation from scratch.[fn]Crisis Group email interview, Gregory Johnsen, 10 January 2017. Al-Wuhayshi became the emir, or leader, of AQ in the land of Yemen (AQLY) in June 2007. On 13 March 2008, AQLY renamed itself AQ in the Southern Arabian Peninsula (AQSAP). AQLY/AQSAP carried out attacks against Western interests, including oil facilities in Hadramout and Marib in 2006, the U.S. embassy in 2008 and South Korean tourists in Marib in 2009.Hide Footnote In January 2009, they formed AQAP from the merger of AQ’s Yemeni and Saudi branches. AQAP launched high-profile attacks against Western interests and the Yemeni security and intelligence forces, especially in the south.[fn]Prior to 2009, there were two AQ groups: AQSAP (the renamed AQLY) and a splinter group, the Yemen Soldiers Brigade, quickly wiped out by Saleh in 2008. Following attacks in Saudi Arabia between 2003 and 2006, a crackdown pushed many Saudi members into Yemen, resulting in the franchises’ coalescence. The group’s founders included non-Yemenis and individuals with close ties to AQ Central’s leadership. Al-Wuhayshi’s leadership – as former secretary to Osama bin Laden and a native Yemeni – combined the valuable attributes of someone with both local ties and a strong connection to AQ’s leadership and its transnational aims.Hide Footnote

By 2011, AQAP was seen in the U.S. as AQ’s most lethal branch, but its influence in Yemen was still circumscribed.[fn]The U.S. put its membership at “several hundred” in 2010 and at “a few thousand” in 2011. “Briefing by Homeland Security Secretary Napolitano, assistant to the president for counterterrorism and homeland security, Brennan, and Press Secretary Gibbs”, White House Office of the Press Secretary, 7 January 2010. “Foreign Terrorist Organizations” in “Country Reports on Terrorism 2011”, U.S. State Department’s Office of the Coordinator for Counterterrorism, 31 July 2012.Hide Footnote Saleh walked a fine line: balancing the U.S. drone campaign against AQ’s leadership and the local population’s resentment toward perceived violations of national sovereignty and the causalities from these strikes.[fn]Notably, a missile attack on al-Maajala in Abyan governorate on 17 December 2009 killed 41 civilians. The Yemeni government claimed responsibility, but photographs of remnants of U.S.-made bombs disputed this. In 2010, Saleh reportedly told General David Petraeus, then U.S. Central Command chief: “We’ll continue saying the bombs are ours, not yours”. “General Petraeus’ meeting with Saleh on security assistance, AQAP strikes”, U.S. embassy Sanaa cable, January 2010, as published by WikiLeaks, https://wikileaks.org/plusd/cables/10SANAA4_a.html.Hide Footnote AQAP was a relatively small component of the domestic balance of power, used by the state to win financial and military support from the U.S. While a top priority for the West, AQAP was far less important for the state and most Yemenis than the growing strength of the Huthis (a revivalist movement in the north based on Zaydism, a version of Shiite Islam), separatist sentiment in the south and an increasingly brittle regime in Sanaa.

III. The Fourth Wave

A. Fertile Ground for Jihadism

Yemen’s fourth wave of jihadist violence is its most potent because of the underlying currents propelling it, including state collapse and sectarianism.[fn]Crisis Group Report, Exploiting Disorder, op. cit.Hide Footnote While IS’s ideological innovations and territorial gains dominate the fourth wave in other parts of the Middle East, in Yemen, AQAP has taken the lead, evolving and adapting in the rapidly shifting post-2011 environment. Its expansion has roughly occurred in two phases: the first in the wake of the 2011 popular uprising against Saleh and the second in the context of a failed political transition and the subsequent war.

1. Uprising

During the uprising, AQAP evolved from a primarily internationally focused jihadist organisation to one with a significant local insurgency component, seeking to strike deeper roots into Yemeni society and establish territorial control. In 2011, it created a parallel group, Ansar al-Sharia (AAS, “Supporters of Islamic Law”), to widen its domestic appeal and separate its local component from its international brand, which many Yemenis view as a regime instrument, infamous for its attacks against the West and likely to trigger a military backlash, especially from the U.S., against communities that support it.[fn]Sheikh Abu Zubayr Adel bin Abdullah al-Abab, a senior AQAP religious official, explained: “The name Ansar al-Sharia is what we use to introduce ourselves in areas where we work to tell people about our work and goals, and that we are on the path of Allah”. Recorded 18 April 2011, posted on YouTube 25 April 2011, https://www.youtube.com/watch?v=js_fbKJN23s.Hide Footnote The move also attempted to address the organisation’s long-running “double-bind” of balancing global objectives – targeting the West and expelling unbelievers from the Arabian Peninsula – with its need to address local grievances, such as corruption and lack of effective justice.[fn]Vahid Brown presents this double-bind as the interplay between two inter-related tensions: the global/local dichotomy, which illustrates the friction between the global range of AQ’s stated plan and the local concerns of those it seeks to win over; and the global/classical dichotomy, in relation to AQ’s concept of violent jihad as the only course of action, an extremist interpretation contested by more authoritative and influential clerical proponents of a classical definition of legitimate jihad. Vahid Brown, “Al‐Qa’ida Central and Local Affiliates”, in Assaf Moghadam and Brian Fishman (eds.), “Self-Inflicted Wounds Debates and Divisions within al-Qa’ida and its Periphery”, Combating Terrorism Center, 2010, pp. 69-100.Hide Footnote In practice, AAS acts as a local insurgent arm and domestic diffusion brand, while AQAP continues to call for strikes against the West, particularly by encouraging “lone-wolf” attacks.[fn]AQAP is hierarchical, structured around a senior leadership whose members are dispersed over committees and councils (such as the media, security and military committees and the Shura council, which advises AQAP’s emir and reports back to al-Qaeda Central). This structure, along with creating a consultative mechanism, allows it to better absorb the impact of assassinations. From 2013 until his death in June 2015, AQAP emir al-Wuhayshi played an additional role as AQ Central’s “general manager” and second-in-command under Ayman al-Zawahiri. The military committee functions as the insurgency’s command structure. Mid-level leaders are province (wilaya) commanders, known as emirs, along with district and city commanders under provincial command. AQAP propaganda regularly refers to these commanders, who are not limited in movement and operations as their title might suggest, as AAS. Jalal Baleedi, killed in February 2016, was both a mid-level AQAP commander and a senior AAS leader. AQAP and AAS have separate media wings for propaganda releases, but these outlets often collaborate. At entry level, AAS recruits are not required to pledge allegiance (bay’a) to AQAP, but senior commanders are. Crisis Group consultant interviews in former capacity, Yemeni analysts and journalists, AQAP members, Yemen, 2011-2016. For the purpose of this report, AQAP and AAS are used interchangeably except when AAS has expressly profiled itself separately. AQAP’s calls for “lone-wolf” attacks underline its continued desire to carry out international operations but may also indicate frustration at its lack of international prowess in recent years. According to a U.S. government official, “AQAP’s ability to balance both local and global objectives has been the key to its sustained relevance and a persistent threat to U.S. interests. AQAP has prioritised a ‘Yemen-first’ approach as the societal instability is far too advantageous for them …. That said, there has been an increase in AQAP’s calls for lone operatives to target the west in its social media, Telegram Channels, and newer traditional media such as Inspire Guides”. Crisis Group email interview, January 2017.Hide Footnote

Under the AAS banner, AQAP captured several towns, including the capital of the southern province of Abyan in May 2011, and governed these areas for more than a year.[fn]This went against Osama bin Laden’s advice in a letter to AQAP leader al-Wuhayshi. Bin Laden also expressed concerns about getting bogged down in the focus on the “internal enemy”, in this case the Yemeni government, rather than on the “far enemy”. SOCOM-2012-0000016, in “Letters from Abbottabad”, Combating Terrorism Center, 3 May 2012.Hide Footnote In keeping with AQ’s global strategy, it has increasingly pursued a gradualist approach, beginning with establishing acceptance among the local population, with the aim of gaining its active support and having civilians join in defending AQAP-controlled territory. This local backing and territorial control – with the aim of creating multiple emirates that should ultimately lead to the creation of a caliphate – would subsequently offer potential for AQAP to launch attacks outside Yemen.[fn]Ayman al-Zawahiri, “General Guidelines for Jihad,” As-Sahab Media, 14 September 2013. This is AQ’s blueprint for a more restrained strategy with an emphasis on localism. AQ groups in Syria, most notably Jabhat al-Nusra, have pursued such tactics. And like its Yemeni counterpart, in July 2016 al-Nusra disassociated itself from the AQ brand and changed its name to Jabhat Fatah al-Sham (JFS). See Crisis Group Middle East Report N°155, Rigged Cars and Barrel Bombs: Aleppo and the State of the Syrian War, 9 September 2014.Hide Footnote

Initially, AQAP suffered a setback under Abed Rabbo Mansour Hadi, Yemen’s transitional president, who took power from Saleh as part of a political deal known as the Gulf Cooperation Council (GCC) initiative in February 2012. In May 2012, a combination of Yemeni security services and local militias, known as Popular Committees, ousted AQAP from Abyan, ending its first experiment with governance.[fn]Although fighting on the same side as government troops, the “Popular Committees” were mostly southern secessionists and openly opposed Hadi. By 2014, Huthis used the same name for their militias.Hide Footnote Reeling from its defeat, it switched back to asymmetrical attacks, which became more sophisticated and larger in scale than before 2011.[fn]These attacks included: a 21 May 2012 suicide bombing in Sanaa that killed more than 100 soldiers; a 30 September 2013 storming of the military’s Zone-2 headquarters in Mukalla, which AQAP militants held for two days; and a similar attack on the defence ministry in Sanaa on 5 December 2013 that left more than 50 dead and included an assault on al-Urdi hospital inside the compound that killed seventeen patients and staff, for which AQAP later apologised.Hide Footnote The interim president also gave the U.S. carte blanche to pursue its drone campaign against jihadists. Yet, military moves against AQAP proved woefully inadequate in the face of expanding political opportunities.

2. Derailed transition

By 2014, Yemen’s political transition was buckling under the weight of corruption and political infighting. The national dialogue conference, a transition cornerstone aimed at constitutional reform, failed to resolve pivotal issues, including the future state structure. In this environment, the biggest winners were the Huthis, a Shiite movement and militia that had previously fought six rounds of conflict with the Saleh regime (2004-2010). They presented themselves as political outsiders opposed to the GCC initiative, which had divided power between established political parties.

Over the course of the transition, the Huthis upended the military power balance in the north by defeating Sunni Islamist and tribal opponents, including an alliance of Salafi fighters, Islah members, Ali Mohsen-aligned military forces and the powerful al-Ahmar clan (no relation to Ali Mohsen al-Ahmar) from the Hashid tribal confederation, in a series of battles in 2013 and 2014.[fn]Crisis Group Middle East Report N°154, The Huthis: From Saada to Sanaa, 10 June 2014.Hide Footnote They also forged an alliance with their former enemies, Saleh and his GPC supporters, who felt marginalised by the GCC-led transition and sought revenge against the Mohsen/Ahmar/Islah alliance that had joined the 2011 popular uprising.

With support from the GPC, the Huthi militia captured Sanaa in September 2014. By February 2015, a dispute over the constitution prompted them to oust the Hadi government. One month later, alarmed by the Huthis’ military advances, which they attributed to Iranian meddling, Saudi Arabia and a coalition of Sunni Arab states, including the United Arab Emirates (UAE), backed by the U.S., UK and France, launched a military intervention and imposed a naval and air blockade to reinstate the Hadi government.[fn]For an overview of the Huthi coup, the group’s expansion southward and the Saudi-led military intervention, see Crisis Group Middle East Briefing N°45, Yemen at War, 27 March 2015.Hide Footnote

As Yemen plunged into war, AQAP and IS flourished. AQAP declared war against the Huthis in early 2011, but then rarely acted on its strong rhetoric, carrying out only a handful of attacks.[fn]The most notable AQAP attack against the Huthis was the car bombing of a Huthi festival on 24 November 2010 that killed 23 people. AQAP claimed responsibility, stating it had killed the Huthis’ spiritual leader, Badr-al-Din al-Huthi. They also claimed a follow-up attack on 26 November targeting a convoy on its way to al-Huthi’s funeral. AQAP’s Sada al-Malahem magazine, 15 February 2011.Hide Footnote This changed in 2014, as Huthi forces broke out of their Saada stronghold. By mid-December 2014, AQAP had claimed responsibility for 149 attacks against the Huthis in fourteen governorates in less than 90 days.[fn]Oren Adaki, “AQAP claims 149 attacks in Yemen since late September”, The Long War Journal (www.longwarjournal.org), 19 December 2014.Hide Footnote  Al-Bayda, a crossroads governorate between north and south that borders eight provinces, was a focus of these assaults, as the Huthis moved into the area in October under the pretext of fighting Daesh (the Arabic acronym for the Islamic State in Iraq and Syria), a term the Huthis have used loosely to characterise a wide range of their opponents.

At the same time, IS, an AQAP competitor, took advantage of increasing sectarianism and violence. It made its Yemen debut on 20 March 2015 in four coordinated suicide attacks against mosques frequented by Huthis in Sanaa, a day after fighting broke out between Saleh loyalists and Hadi-aligned fighters in Aden. The IS bombings provided justification for the Huthi push into Aden as a necessary fight against the growing security void, which the Huthis viewed as intentionally created by their political rival, President Hadi, and filled by AQ and other violent jihadists.[fn]Crisis Group interviews, Huthi supporters, August and September 2015.Hide Footnote

Only a week after the Saudi-led coalition began attacking Huthi-Saleh forces from the air, AQAP moved once again to capture territory, this time in the eastern governorate of Hadramout. Unchallenged by local military units, it easily took control of the provincial capital, Mukalla, and large swathes of the province’s coastline.[fn]AQAP denied it had seized Mukalla, saying instead it had been a “Sunni tribal takeover” by the “Sons of Hadramout”. However, local residents and the international community regarded Mukalla as an AQ-controlled city from April 2015 to April 2016. This report refers to AQAP as having seized the city, which it then controlled through local administrative bodies, some of which included non-AQAP members. The name AAS is not referred to in this context, as it was used by neither residents nor the militants to describe themselves during this period.Hide Footnote There, it exhibited improved governance skills by applying lessons learned from its previous experience in Abyan. AQAP held Mukalla for over a year, as the coalition was fighting Huthi/Saleh forces elsewhere.

After nearly two years of war, AQAP/AAS is now deeply enmeshed in the on-going battle against the Huthi/Saleh bloc on a number of fronts, including al-Bayda, Shebwa, Marib, Jawf and Taiz governorates. Its numbers, while difficult to assess, have swelled, reaching approximately 4,000 by 2015 according to U.S. State Department estimates.[fn]Compared to 1,000 in 2014. “Country Reports on Terrorism 2015”, U.S. State Department, 2 June 2016.Hide Footnote Equally importantly, its staying power has grown through a vast war chest. It also has acquired a wide range of new weaponry, including heavy weapons from Yemeni military camps or acquired indirectly from the Saudi-led coalition, which has been supplying arms to a range of anti-Huthi fighters.[fn]AQAP has reportedly plundered thirteen army units across Yemen since March 2015. “Midterm update of the Panel of Experts on Yemen established pursuant to Security Council resolution 2140 (2014)”, unpublished report, August 2016. Eyewitnesses reported seeing tanks seized by AQAP from Mukalla’s two military bases being transported toward northern Hadramout, destination unknown. Crisis Group consultant interviews in former capacity, Mukalla, March 2016.Hide Footnote While AQAP was forced to withdraw from Mukalla in the face of a UAE-led, U.S.-assisted military campaign in April 2016, the group is far from defeated and has instead relocated to adjoining governorates or blended into the local population. It continues to exercise on-again, off-again control of areas in Abyan and neighbouring Shebwa.

IS has not seized territory, and its following remains small.[fn]In May 2016, a U.S. State Department official estimated IS numbers in Yemen at around 150. Crisis Group interview, Washington, May 2016. A suicide bomber from Aden said in a phone call with friends shortly before his June 2016 death in Mukalla that IS’s Yemen membership was as low as 70. Crisis Group consultant phone interview in former capacity, friend of the suicide bomber, 27 June 2016.Hide Footnote Yet, it has found fertile ground in cities such as Aden, which suffered sectarian-tinged violence in the aftermath of the Huthi takeover and subsequent removal. IS has taken credit for a number of high-profile attacks against both Huthi forces and the Hadi government and its allies in the south.[fn]Including for two suicide bombings against Yemeni security personnel in Aden that killed 57 on 10 December 2016 and over 50 on 18 December 2016.Hide Footnote

B. Drivers of AQAP and IS Expansion

1. Opportunity in chaos

An important factor in AQAP’s steady gains has been its ability to capitalise on the near-collapse of weak state institutions, particularly the security services. In the past, the Yemeni state at times lent tacit support to violent jihadist groups for political or financial gain. Yet, prior to 2011, there were limits to AQAP’s activities. The group never, for example, held or governed territory. This changed in 2011 when state security services split, one side remaining loyal to Saleh and the other, led by Ali Mohsen, joining protests against him. This fracturing accelerated during the 2015 war. Now, members of the security services are either fighting with the Huthi/Saleh front or the Saudi-led coalition or staying at home. In this environment, not only is there no unified effort to put AQAP on a leash, but the group can step into local political and security vacuums.

Each time AAS has captured significant territory, there has been little or no resistance from the security services, regardless of the latter’s political alignment. This was the case, for example, on 29 May 2011 when militants took over Zinjibar, Abyan, and later five more towns across Abyan and Shebwa provinces. Some observers suggest that Saleh’s allies were preoccupied with the battle for Sanaa as the uprising unfolded and the army split. Local residents in Abyan, who witnessed AQAP’s territorial gains, said security forces abandoned their positions and handed over municipal buildings to barely a handful of militants.[fn]Zinjibar residents said that fewer than a dozen men took control of the town with minimal, if any, effort to repel them. Crisis Group consultant interviews in former capacity, Aden, Jaar and Zinjibar, 30 and 31 May 2011, 22 and 23 May 2012 and 13 and 14 June 2012.Hide Footnote Saleh’s opponents viewed the Abyan events as a ruse by the besieged president to persuade his international allies to support him and to draw attention from events in Taiz, where, on the same day, his forces had razed an anti-government protest camp, killing and injuring more than 270 demonstrators.[fn]Crisis Group consultant interviews in former capacity, Yemeni political analyst, Sanaa, June 2011; Yemeni politician who defected from Saleh’s GPC party, Sanaa, 12 June 2011. Defected general Ali Mohsen said Saleh had “handed over Abyan to terrorist gunmen”. Quotes in al-Hayat, June 2011. For details of the Taiz crackdown see, “No Safe Places: Yemen’s Crackdown on Protests in Taizz”, Human Rights Watch, 6 February 2012.Hide Footnote Whatever the reasons for the swift takeover of southern cities, security services failed to act and thus AQAP became the major beneficiary.

In Mukalla, too, state security forces were either unwilling or unable to oppose an AQAP takeover in April 2015. Local residents reported that they failed to put up a fight when the group entered the town and even prevented tribal fighters from stopping it.[fn]Army commanders fled the Zone-2 military headquarters in Mukalla to join the 27th mechanised brigade at al-Riyan airbase, north east of the city. When local tribal fighters attempted to enter the city to fight AQAP, they were blocked by the military, who preferred negotiating their safe withdrawal; clashes ensued between soldiers and tribal fighters. At the time of the takeover, soldiers’ and commanders’ loyalties were murky. The military commanders of Zone-2 and the 27th mechanised brigade were Hadi appointees. Yet, many of the troops in the area are affiliated with Saleh or Ali Mohsen. AQAP allowed the soldiers to leave as long as they deposited all but their personal weapons; it provided transport and cash handouts to them; and it encouraged them to continue to collect their government salaries in Mahra, a neighbouring province, and northern Hadramout, but made them pledge not to fight AQAP or AAS in the future. Crisis Group consultant interviews in former capacity, soldier and army officer present on the 27th mechanised brigade base in April 2015 and evacuated by AQAP, Hadramout, 15 March 2016; Mukalla resident party to the negotiations between AQAP and the army, 14 March 2016; Sheikh Amr bin Habrish, head of the Hadramout Tribal Confederacy, PetroMasila, 15 March 2016.Hide Footnote

Equally important, once in control of cities, AQAP/AAS has presented itself as a viable and indeed better alternative to the state by providing more reliable services and dispute adjudication. In Abyan between May 2011 and 2012, AAS provided services such as water and electricity, as well as education and an efficient justice system based on Sharia (Islamic law), and went as far as compensating families which had lost their homes to U.S. drone and airstrikes. AAS’s popularity was clearly based on its comparatively efficient governance more than on its ideology.[fn]Crisis Group consultant interviews in former capacity, imam of Jaar central mosque, local lawyer from Jaar, teenage girls and their mothers, male Jaar residents, Jaar, Abyan, May 2012.Hide Footnote

After being evicted from Abyan, the group applied lessons from its experience there to Mukalla. It further softened its approach by socialising with residents and refraining from draconian rules.[fn]AQAP executed spies, but permitted women to be outside their homes after dark. AQAP leader al-Wuhayshi wrote to the leader of al-Qaeda in the Islamic Maghreb (AQIM) after the group’s withdrawal from Abyan in 2012 to give advice on how to govern. This included tips on the implementation of Islamic punishments. He wrote: “Try to avoid enforcing Islamic punishments as much as possible, unless you are forced to do so”. “The al-Qaida Papers”, Associated Press, February 2013.Hide Footnote As part of this effort, it put in place a local ruling council – the Hadramout National Council (HNC) – rather than instituting direct rule.[fn]The fifteen members of the HNC also included local dignitaries (some of whom had served on the pre-existing local Council of Sunni Scholars) and prominent non-AQAP Hadramis. According to a resident, “The council [HNC] is widely viewed as a front to legitimise AQAP’s hold on power, but locals see it as an acceptable way to deal with the outside world”. Crisis Group interview, April 2015; consultant interview in former capacity, Mukalla, March 2016.Hide Footnote AQAP appointed local members as religious police. It also launched infrastructure projects, provided social services, such as food distribution for families in need and medical supplies and equipment for hospitals, and staged community events and street festivals.[fn]Crisis Group consultant interviews in former capacity, AQAP officials, residents, Mukalla, March 2016. AAS’s propaganda channel, al-Atheer, released pictures and videos of its community work. These efforts extended to areas under its control in Abyan and Shebwa.Hide Footnote The infamous rayat al-sudaa’, AQ’s black banner, was hard to find in the city, and by order of AQAP’s then-leader, al-Wuhayshi, it was not displayed during the militants’ takeover.[fn]AQAP’s then-spokesman spent considerable time trying to persuade international media that it was not AQAP or AAS seizing the city, but a Sunni tribal takeover led by the so-called “Sons of Hadramout”. This may have been an attempt to avoid drawing U.S. airstrikes, which later killed the spokesman and other senior AQAP leaders in Mukalla. Crisis Group consultant multimedia interview in former capacity, AQAP spokesman Muhannad Ghalleb, April 2015.Hide Footnote

Residents in AQAP-controlled areas, while not supporting the group’s ideology, have regularly praised its prioritisation of security, basic services and a mechanism to resolve grievances, such as long-running land disputes.[fn]Crisis Group consultant interviews in former capacity, civilians living in AQAP-controlled territory, Abyan and Hadramout, 2011-2016.Hide Footnote According to a Mukalla resident in 2015:

We view the [Hadramout National] Council positively, because it has managed to continue to pay government salaries …. It has kept public services at a much better level than what is available in the rest of the county …. The AQAP judicial system is fair and swift and therefore preferred over the government’s corrupt system. Many prominent cases that had lingered for years were resolved in a single day.[fn]Crisis Group email interview, Mukalla resident, April 2015.Hide Footnote

AQAP also avoided a bloody fight to hold territory once it became clear that Saudi-led forces, especially the UAE, were determined in May 2016 to drive the group out. Its experience in Abyan in 2011-2012 had taught it that defending territory in a conventional conflict against foreign-backed forces was costly and risked alienating local populations it had spent months winning over. AQAP reportedly arranged its exit from Mukalla in coordination with coalition-allied forces prior to the UAE/Yemeni government assault.[fn]Members of the Saudi-led coalition and supporters of the Hadi government argued there was no agreement with AQAP to allow its fighters to leave the city. Instead, they said, AQAP fled, possibly with local help in the face of the impending UAE/Yemeni military onslaught. Crisis Group interviews, August, September 2016. However, AQAP officials said they received ample advance warning of the coming offensive and the fact that it would be preceded by airstrikes, and began sending their fighters out a month in advance. A former HNC member said AQAP coordinated its withdrawal with the coalition in Riyadh to allow a westward retreat. The eventual push by coalition-supported troops beginning on the night of 23/24 April 2016 was preceded by several days of airstrikes targeting empty AQAP military camps and facilities. Crisis Group consultant interviews in former capacity, AQAP officials, former HNC member, Mukalla, March 2016; AQAP official, April 2016; and phone interviews, residents of Mukalla, witnesses in coastal Shebwa, April 2016.Hide Footnote

2. Huthi expansion, sectarianism and new alliances

Arguably, AQAP has most benefited from a combination of Huthi military expansion and growing sectarianism, as these have opened new opportunities for forging local alliances. Huthi inroads in predominantly Shafai/Sunni areas south and east of Sanaa sparked a series of opportunistic alliances with a range of anti-Huthi forces, who called themselves “resistance fighters”.[fn]When anti-Huthi forces pushed south in 2015, local guerrilla forces became known as the “Southern Resistance” in the fight for Aden and southern governorates. In Taiz, these fighters identify as “the Resistance”. Although all fighting on the Yemeni government’s side, these paramilitary forces are not by default pro-Hadi government. In the south, the name specifically denotes the armed separatist movement Hiraak. The southern resistance consists of local residents taking up arms to defend their homes, as well as former soldiers of the PDRY.Hide Footnote This happened on key battle fronts such as Taiz, Marib and al-Bayda, all provinces of the former north Yemen. It also happened in territories of former south Yemen, including Aden, before Huthi/Saleh forces were pushed out in July and August 2015.

The majority of anti-Huthi/Saleh fighters do not share AQAP’s ideology but reject northern/Zaydi domination. In the south, fighters are predominantly separatists, often leftist in orientation. Their tacit alliance with AQAP broke down as soon as Huthi/Saleh forces had been driven out of Aden. Indeed, both AQAP and IS started attacking their erstwhile tactical friends. In response, southern security forces, with the help of the UAE, have launched a number of military operations against them.

More importantly, growing sectarian sentiment has provided political/social space for groups such as AQAP and IS to recruit and establish more durable footholds in local communities. Yemenis are quick to point out that Zaydism and Shafaism (a Sunni school of jurisprudence) are relatively tolerant, have converged over time to the extent that Zaydis and Shafais pray in each other’s mosques, and do not have a history of violent relations. In the past, even AQAP acted pragmatically in light of these social constraints by avoiding direct confrontation with the larger Zaydi community and instead focusing its critique on Twelver Shiites, predominant in Iran, Iraq and Lebanon.[fn]Barak Barfi, “AQAP’s Soft Power Strategy in Yemen”, CTC Sentinel, November 2012.Hide Footnote Yet, Yemen’s legacy of tolerance is becoming a casualty of the civil war, and AQAP is taking advantage.

The Huthis and their opponents share responsibility for the growth in sectarian sentiment. The Huthis often conflate Islah, Salafi groups, southern separatists and others with AQAP and IS, referring to all of them as takfiris, al-Qaeda or Daesh.[fn]Crisis Group interviews, Huthi representatives, June, August and September 2015; consultant observations in former capacity, Aden, Taiz, Sanaa and Saada, May-November 2015.Hide Footnote They justify their original push south from Sanaa by the need to fill the security void and associated threat posed by Daesh/AQ, which they said were gaining strength and allied with Hadi.[fn]Crisis Group consultant interview in former capacity, Huthi representative, Sanaa, May 2015. Crisis Group phone interviews, Huthi supporters, May-June 2015.Hide Footnote However, at that time, IS did not exist in Aden. It was the war, triggered by the Huthis’ capture of Sanaa, and subsequent chaos that gave rise to the group. In other governorates, such as al-Bayda, the Huthis’ military progress had a similar catalysing effect on sectarianism.[fn]A researcher from al-Bayda noted that, prior to the Huthi advance into the governorate, any differences between local mosques concerned political affiliation: GPC versus Islah. As conflict spread, however, mosques began to segregate along Shafai versus Zaydi lines. Crisis Group consultant interview in former capacity, Abdulsalam al-Rubaidi, Sanaa, 11 November 2014.Hide Footnote

The layering of Saudi-Iranian competition in the Gulf onto Yemen’s civil war has further amplified the sectarian dimension, with the Saudis accusing the Huthis of being Iranian proxies, a charge echoed by Hadi supporters.

In addition, the alliance with Saleh and his networks, which are strongest in the Zaydi highlands, has reinforced a common perception among their opponents that the war has a north/Zaydi versus south/Shafai dimension.[fn]Crisis Group interviews, Salafi politician, Aden, February 2015; Yemeni businessman, Sanaa, March 2015; phone interviews, adviser to Hadi government, October 2015; Yemeni activist from Ibb, November 2015.Hide Footnote Sectarian language is common among anti-Huthi fighters. They accuse the Huthis of harbouring an agenda that either includes the promotion of Zaydis, and particularly Hashemites (a subset of Zaydis claiming to be descendants of the Prophet Mohammed), or converting to Twelves Shiism. Some even refer to them as rawafedh (“rejectionists”) and murtadeen (apostates), terms laden with anti-Shiite connotations.[fn]The term rawafedh is used by some Salafis and Wahhabis to denote Shiites for their rejection of what Sunnis consider the legitimate line of succession from the Prophet – the issue that gave rise to the Sunni-Shiite split. In this line of thinking, Shiites are then also murtadeen, apostates from the true belief.Hide Footnote The layering of Saudi-Iranian competition in the Gulf onto Yemen’s civil war has further amplified the sectarian dimension, with the Saudis accusing the Huthis of being Iranian proxies, a charge echoed by Hadi supporters.[fn]Crisis Group interviews, Yemeni government officials and GCC diplomats, May and June 2015; “Arab states send complaint letter against Iran to the UN”, al-Arabiya English, 13 November 2016. For a review of growing sectarianism see Farea al-Muslimi, “How Sunni-Shia Sectarianism is Poisoning Yemen”, Carnegie Middle East Center, 29 December 2015.Hide Footnote

AQAP has purposefully blurred the lines between its followers and the wider Sunni and anti-Huthi population. In August 2014, AAS leader Jalal Baleedi led an unusually brutal attack in Hadramout against unarmed soldiers he accused of being Huthis, and then warned of a sectarian war if the Huthis were to take Sanaa.[fn]Baleedi led a team of fighters in the beheading of four and shooting execution of ten off-duty soldiers in Sayoun, Hadramout, in August 2014, claiming they were Huthis. In response to journalist Iona Craig’s question, AQAP ideologue Nasser bin Ali al-Ansi later denounced such beheadings in a video release billed by the group as their “first international press conference”, Al-Malahem Media, 8 December 2014. Baleedi’s association with IS is a point of confusion in Yemen. In Exploiting Disorder, op. cit., Crisis Group reported that Baleedi had pledged allegiance to al-Baghdadi based on reports from a Yemeni consultant and Yemeni media sources. However, when he died in February 2016, it became apparent he had remained loyal to AQAP. AQAP released an audio eulogy from its leader, Qasim al-Raymi, who had replaced al-Wuhayshi, and subsequently it produced a lengthy propaganda video of a “Special Forces Battalion” training centre named in Baleedi’s honour.Hide Footnote When that happened, AQAP promptly called all Sunnis to arms.[fn]AQAP chastised Sunni leaders’ support for the “rafidi” Huthi takeover and made a call to arms for Sunnis. “Statement Regarding the Crimes of the Huthi Faction Against the Sunnis”, with the strapline “A Call to Sunnis, AQAP official statement disseminated by AAS social media channels, 23 September 2014.Hide Footnote

As the conflict has unfolded, AQAP has used the pretext of a “Sunni” defence against the “Shiite” Huthis to blend with local tribes and Salafi sympathisers. A senior AQAP official compared the Huthis’ “Shiite” control of Yemen after they captured Sanaa with a Shiite-dominated Iraq after the U.S. overthrew Saddam Hussein; contended that AAS was the only remaining Sunni defence in southern governorates against Huthi expansion after the Hadi government fled into exile and the military collapsed; and that one of AQAP’s main roles was to provide military experience and expertise to local tribesmen, saying: “We are as one with the [Sunni] tribes like never before. We are not al-Qaeda now. Together we are the Sunni army”.[fn]The senior AQAP official repeatedly reiterated AAS’s position as gathering “Sunni tribesmen” against what he called Yemen’s takeover by Shiites. Crisis Group consultant multimedia interviews, AQAP senior official, September 2014 and April 2015.Hide Footnote

The group’s strategy of blurring lines between core membership and sympathisers has also allowed it to extend its reach broadly into local communities. The creation of AAS in particular lowered the bar for would-be recruits by enabling them not to bind themselves to AQAP and its ideology through a formal pledge of allegiance (bay’a).[fn]In AQAP’s case, bay’a is a pledge of allegiance that formally binds an individual to the organisation, itself tied by bay’a to AQ Central. The concept of bay’a goes back to the time of the Prophet Mohammed, who expected bay’a from his followers. By comparison, a mere pledge of support does not imply a binding relationship.Hide Footnote AQAP has similarly used the “Sons of” designation for affiliate groups, as in its takeover of Mukalla. There, it presented the Sons of Hadramout as a local Sunni collective brought together to fight the common Huthi enemy. A senior AQAP official quipped that it was Western obsession with labelling that had motivated the group to use the “Sons of” tag: “You [Westerners] call us all sorts of names and the names may change. But to us, we are all Muslims, we are all brothers”.[fn]According to AQAP, the Sons of Abyan and Sons of Hadramout (in addition to other geographically-linked “Sons of” groups) are not full AQAP/AAS members but “might become so in the future”. They receive basic military training and religious teaching with a specific mandate of fighting the Huthis. They have not pledged allegiance to AQAP. Crisis Group consultant interviews in former capacity, AQAP official, Yemen, April 2015, April 2016. Other analysts have put the difference down to little more than semantics. “The Hadramawt: AQAP and the Battle for Yemen’s Wealthiest Governorate”, The Jamestown Foundation, 10 July 2015. The exact relationship between AQAP and the “Sons of” movements is unclear. The opaqueness of the “Sons of” groups allows greater opportunity for AQAP to act as a sleeper movement, particularly after its withdrawal from territory, to be activated at an opportune moment.Hide Footnote

3. AQ as priority number two for regional actors

In prosecuting the war, the Saudi-led coalition has relegated confronting AQAP and IS to a second-tier priority. Its stated primary rationale has been to roll back Huthi gains and reinstate the Hadi government, which requested the military intervention. Saudi motivations to enter the war were based on their perception that the Huthis, as alleged Iranian proxies, posed an existential threat, and on internal political dynamics in which a successful intervention in Yemen would boost the prospects of its main architect, Deputy Crown Prince and Defence Minister Mohammed bin Salman.[fn]A reason cited less often is that Saudi Arabia and the UAE saw a need to manage a Sunni uprising against the Huthi/Saleh alliance. Crisis Group interviews, Saudi official, Yemeni businessman with Saudi ties, June 2015. According to a UAE official, the UAE and Saudi Arabia were forced to act to prevent a Syrian-style civil war, a scenario in which Sunnis, lacking protection or backing, would turn to groups such as AQAP or IS for support. Crisis Group interview, May 2015.Hide Footnote

This prioritisation, much to the U.S.’s frustration and embarrassment, has facilitated AQAP’s efforts to blend in with the anti-Huthi opposition, which has given it access to weapons and new sources of income. During AQAP’s occupation of Mukalla, it robbed state banks, looted weapons caches and embedded itself in the local economy. AAS has regularly fought alongside Saudi-led coalition forces in their effort to dislodge Huthi/Saleh forces from Aden and other parts of the south, including Taiz, indirectly obtaining weapons from them.[fn]Crisis Group consultant observations in former capacity, Aden, May-August 2015, Lahj, July, August 2015, Taiz, September 2015; consultant interview in former capacity, Adeni doctor taken by UAE troops to treat AAS fighters in a military camp, Aden, July 2015. AAS’s media wing has also released several videos over the course of the conflict purportedly showing AAS members taking part in fighting against the Huthis during the battle for Aden and Abyan in 2015, as well as Taiz in 2015-2016 and al-Bayda in 2016. Wilayat Aden, September 2015: https://justpaste.it/ADEN; Wilayat Abyan, June 2015: https://justpaste.it/ltlz; Wilayat Taiz, January 2016: http://tinyurl.com/jn62seb.Hide Footnote

It is only in southern territories from which Huthi/Saleh forces have been removed that the UAE, in particular, has begun to confront AQAP and IS, working with a variety of southern groups. In Aden, they have worked with some success with the security chief and governor to drive AQAP and IS from the city.[fn]Both AQAP and IS were prepared for the eventual reversal, taking full advantage of the Saudi-led coalition’s fight against the Huthi/Saleh bloc to stockpile money and weapons. Crisis Group consultant interviews in former capacity, AQAP members, Mukalla, March 2016; AQAP supporters, Aden, August 2015. Both groups gathered heavy weapons left behind by retreating Huthi/Saleh forces. In preparation for expected later confrontations with UAE troops and allied forces, IS was also able to acquire UAE-supplied armoured vehicles for suicide bombings against coalition forces. Crisis Group consultant interviews in former capacity, two field commanders of anti-Huthi resistance forces and colonel commanding the Decisive Salman Brigade, Aden, July and August 2015.Hide Footnote The UAE and local allies, supported by a small group of U.S. military advisers, also retook Mukalla in April 2016 without serious fighting and have a continued troop presence there.[fn]Official Saudi-led coalition statements indicated that 800 AQAP militants were killed in the “battle for Mukalla”. AQAP said that twelve of their fighters were killed at a checkpoint on the edge of AQAP-controlled territory (some 100km outside Mukalla), but that no lives were lost in the withdrawal from the city. Crisis Group consultant multimedia interview in former capacity, April 2016. A commander of Hadrami soldiers taking part in the offensive said they killed around twenty AQAP militants at the checkpoint marking the northern entry to their territory and no fighting took place inside the city. Crisis Group consultant phone interviews in former capacity, April 2016.Hide Footnote Similar efforts to evict AQAP/ASS from Abyan and Shebwa have been less successful; AQAP and AAS fighters have repeatedly re-emerged in areas they had vacated or, in some cases, never fully left.

Saudi-led coalition statements that fighting the group is a top priority and announcements of military victories against AQAP in the south are belied by events.[fn]In May 2016, during UN-sponsored peace talks between Yemeni parties in Kuwait, Saudi Foreign Minister Adel al-Jubeir said that the coalition’s priority had shifted from the Huthis to AQ. “Saudi FM: Fighting al-Qaeda is now a priority in Yemen”, al-Arabiya English, 14 May 2016.Hide Footnote In northern Yemen, where the battle against Huthi/Saleh forces continues, the coalition has engaged in tacit alliances with AQAP fighters, or at least turned a blind eye to them, as long as they have assisted in attacking the common enemy. Indeed, three Hadi associates have appeared on a U.S. Treasury list of “global terrorists” for allegedly providing financial support to, and acting on behalf of, AQAP.[fn]These include: Abd al-Wahhab al-Humayqani, one of Hadi’s delegates to UN peace talks in Geneva in June 2015, Nayif Salih Salim al-Qaysi, governor of al-Bayda, and Hassan Ali Ali Abkar, a militia commander and member of the Consultative Council from Jawf. “Treasury Designates Al-Qaida, Al-Nusrah Front, AQAP, And Isil Fundraisers And Facilitators”, U.S. Department of the Treasury, 19 May 2016; and U.S. Department of the Treasury, Resource Center, Counter Terrorism designations, 7 December 2016.Hide Footnote The focus on the Huthis in some ways makes short-term sense for Saudi Arabia, where the threat of Iranian encirclement resonates widely domestically, as opposed to the threat from Sunni extremists, which is a more complicated sell, given local pockets of sympathy and support. Yet, AQAP also seeks to topple the Saudi monarchy, which it views as corrupt and tied to the West, a threat that grows as the organisation gains ground in Yemen.

In the south, where the UAE has shifted its priorities to fighting AQAP/IS, its efforts are complicated by the latter’s ability to find temporary safe havens when needed. Moreover, lack of a unified southern leadership or plan to address security and governance challenges in chronically neglected and impoverished governorates like Abyan and Shebwa continues to provide a receptive environment for violent jihadist groups. According to a resident of Jaar:

A successful strategy for combating al-Qaeda should focus on governance and service provision. Ideologically, AQAP and IS don’t have much in the south. In Jaar, however, many young people have joined Ansar Sharia because they are poor, have little education and see no future for themselves. Ansar provides these young people with an income and a purpose.[fn]Crisis Group interview, September 2016.Hide Footnote

As long as the war, infighting among southern elites and chronic governance challenges continue, the much publicised military initiatives against AQAP and IS will probably not amount to more than temporary victories.

4. The war economy

AQAP’s progress is also in no small part the result of its financial gains. During the evolving conflict, the group has expanded its war chest by raiding banks and controlling seaports and smuggling routes. Its most successful feat was its looting of the Mukalla bank in April-May 2015, which netted approximately 24 billion Yemeni riyals ($111 million).[fn]A U.S. strike sometime between May and August 2015 reportedly destroyed 9 billion Yemeni riyals ($41 million). Crisis Group consultant interview in former capacity, former security official, Mukalla, March 2016. In addition to central bank funds, local bankers estimate that AQAP acquired about 6 billion Yemeni riyals (approximately $24 million) and more than 20 million Saudi riyals (approximately $5.3 million) from commercial banks on 2 April 2015. Crisis Group consultant interviews in former capacity, Mukalla, April 2016.Hide Footnote It also imposed import levies at the Mukalla and Ash-Shihr seaports, collecting a fee for every litre of fuel and every cargo container offloaded.[fn]Crisis Group interviews, Mukalla resident, phone interview, April 2015; north Yemen tribal sheikh, Hadrami politician, September 2016.Hide Footnote Ironically, much of the imported fuel found its way to northern markets, which were largely cut off from the outside world by the Saudi-led coalition embargo, through a chain of local intermediaries who purchased fuel in Hadramout for resale to Huthi/Saleh forces.[fn]Crisis Group phone interview, Mukalla resident, August 2016.Hide Footnote

A UAE official described Mukalla as “al-Qaeda’s lungs”, and indeed the loss of the port was a significant blow to its funding stream.[fn]Crisis Group interview, UAE official, Abu Dhabi, December 2016.Hide Footnote Yet, the effects of controlling Mukalla for over a year will not fade quickly. AQAP’s accumulated revenues enhance its ability to purchase military hardware and attract recruits. It already offered its fighters in 2011 a salary higher than that of government soldiers, and its new financial resources can exercise an even more significant pull on impoverished young men.[fn]AQAP/AAS paid a $200 monthly salary to fighters compared to $140-150 for a regular Yemeni soldier. Crisis Group consultant multimedia interview in former capacity, AQAP member, March 2016. Four male residents in AQAP/AAS-controlled territory said that it enticed many local recruits primarily with the financial reward, not through religious or ideological persuasion. Crisis Group consultant interviews in former capacity, Abyan, May 2012.Hide Footnote

C. IS versus AQAP

Beginning in November 2014, a nascent IS branch put itself on the Yemen conflict map through spectacular attacks against a variety of protagonists, including both the Huthis and the Hadi government. Like AQAP, IS has benefited from state collapse, Huthi expansionism, regional states’ preoccupation with Iran and a burgeoning war economy. More than AQAP, however, IS is also a product of growing sectarianism and extreme levels of violence, which have radicalised young men and made them susceptible to its recruitment. Most importantly, its appearance in Yemen is tied to its successes in Iraq and Syria. At the same time, its influence in Yemen has been circumscribed by AQAP’s long history, military prowess and local support.

On 13 November 2014, shortly after the September Huthi takeover of Sanaa, IS leader Abu Bakr al-Baghdadi announced IS in Yemen.[fn]Seven local branches announced themselves shortly afterwards, operating in ten governorates: Saada, Sanaa, al-Jawf, al-Bayda, Taiz, Ibb, Lahj, Aden, Shebwa and Hadramout. A handful of young men posted pictures on social media purportedly declaring an unverified eighth local branch in Mahra governorate in September 2016.Hide Footnote Initially, IS’s meteoric rise in Syria and Iraq produced a similar momentum in Yemen. Numerous AQAP members defected to it, including senior ideologues such as Sheikh Mamoun Hatem and the prominent religious cleric Abdul-Majid al-Hitari.[fn]Hatem was an early vocal IS supporter on social media. His defection was confirmed by AQAP’s spokesman, who said that although Hatem was “no longer one of us [AQAP], he is still our brother”. Crisis Group consultant interview in former capacity, October 2014. Hatem returned to AQAP in 2015 and was killed in an apparent U.S. drone strike in Mukalla on 11 May.Hide Footnote This trend was helped by a number of U.S. assassinations of AQAP ideologues, including Harith al-Nadhari, Ibrahim al-Rubaysh, and Nasser al-Ansi, between January and April 2015, and their leader al-Wuhayshi in June 2015.[fn]These senior ideologues were crucial to providing theological arguments against joining IS. While al-Wuhayshi was not a theologian, he was a very popular, astute and much-respected figurehead of the group. After his death, an undetermined number of AQAP members were said to have defected to IS. Crisis Group consultant multimedia interviews in former capacity, Shebwa tribal leaders, July, August 2015. The senior ideologues were also veteran AQ members with experience garnered in Bosnia, Afghanistan and the Philippines. This old guard appears to still hold significance to AQAP. They were featured posthumously in an AQAP propaganda video in December 2015.Hide Footnote

IS also took advantage of the 2015 battle for Aden. Four months of heavy fighting that left thousands dead fuelled the radicalisation of young men and accelerated the group’s rise. It was quick to deploy its own “journalists” among the youth to spread its ideology through videos and propaganda songs.[fn]Some local youth in Aden and Little Aden (an area to the west of Aden city) referred to IS recruiters as “propagandists” who distributed videos by smart phones and promoted IS social media, and who told the youth in the streets about IS and its work. Others used the word “journalist” to describe these activities. Crisis Group consultant interviews in former capacity, Aden, August 2015.Hide Footnote This tactic proved fruitful, as a significant portion of IS’s young suicide bombers targeting UAE and UAE-supported forces in Mukalla in April 2016 hailed from Aden. These recruits continued to carry out high-profile attacks, including a suicide bombing against a government military recruitment centre in the port city on 29 August 2016 that killed over 60.

AQAP responded to IS’s rise by trying to contain the group. Initially, it downplayed the threat by refusing to respond to journalists’ questions about IS’s very existence.[fn]According to a senior AQAP official, referring to IS: “AQAP is not responsible for Ansar al-Dawla [Defenders of the State] actions. But they are our brothers and we would help them”. The same official (later killed in a U.S. drone strike) refused, on orders from al-Wuhayshi, to comment on IS’s first attack in March 2015 against mosques in Sanaa. Crisis Group consultant multimedia interviews in former capacity, November 2014, March 2015.Hide Footnote Later, AQAP leaders publicly criticised IS, denouncing its attacks on mosques, which they contrasted with their own reputed sensitivity to local norms. They also engaged in a broader media campaign, ridiculing IS and al-Baghdadi, its self-proclaimed caliph.[fn]AQAP denounced “Baghdadi’s group” as being “based on nothing but a lie … sin, and a blend of ignorance, deviation and desire”. AQAP video release, 1 November 2015.Hide Footnote While they have been quick to praise attacks by persons claiming to act on IS’s behalf in the West, there is no evidence that the two groups have collaborated in any way, including by sharing intelligence.[fn]On 23 June 2016, AQAP’s al-Malahem media arm published the first “Inspire Guide” analysing an IS-linked operation on a nightclub in Orlando, Florida, U.S., on 12 June 2016. A second edition on 21 July 2016 praised the IS-linked attack in Nice, France, a week earlier.Hide Footnote

Their [Islamic State] brutal tactics, including mass killings and mosque bombings, and their relatively autocratic style, are at odds with societal and tribal norms.

IS’s Yemen leadership, unlike AQAP’s, consists mainly of non-Yemenis and its members appear to have been with IS in Syria and Iraq; they brought to Yemen the same strategy of embedded networks of informants and local propagandists that contributed to the group’s successes there.[fn]IS uses these networks to increase its sympathisers and determine whom they can trust and whom to eliminate when taking territory. They rise up when the group moves in from the outside to jointly take control of a place. Paranoia among southerners that Huthi/Saleh forces were using this type of infiltration to retake Aden (after they were evicted in July-August 2015) prompted southern military forces to expel many northerners.Hide Footnote Despite, or possibly because of, their leaders’ international credentials, they have struggled to establish a broad base of support in Yemen. Their brutal tactics, including mass killings and mosque bombings, and their relatively autocratic style, are at odds with societal and tribal norms.

In December 2015, IS faced an internal mutiny when fifteen senior figures and 55 fighters accused their leader, the governor (wali) of Yemen province, of violating Sharia. They listed a number of infringements, including the wrongful dismissal of soldiers, failure to provide basic supplies during a battle in Hadramout, committing “injustices against the weak” and refusal to adhere to a Sharia ruling against an IS regional commander. A contemptuous written response from IS’s central leadership in Syria/Iraq demanded obedience.[fn]The defectors included three members of the group’s Sharia committee: Sheikh Abu al-Shayma al-Muhajir, Sheikh Abu Muslim al-Mansour and Sheikh Abu Hajar al-Adani, in addition to the “province’s” military commander, Abu Aassim al-Bika, and the security chief. In their letter titled “A Statement of Defection from the Wali of Yemen” and published online on 15 December 2015, they reaffirmed their allegiance to IS leader in Syria, while announcing their defection from their wali. On 19 December 2015, Abu Ubayda al-Hakim, a member of the Shura Council of the Caliphate in Syria/Iraq, responded to the complaint by supporting Yemen’s wali and stated that by disobeying their local leader the group had renounced their pledge to al-Baghdadi.Hide Footnote Rejecting the letter, all 70 members left the group. On 24 December, an additional 31, including three senior figures, released a statement joining the rebellion and renouncing the IS leader in Yemen.[fn]Three senior figures – Sheikh Salman al-Lahiji and Rawaha al-Adeni from the Security Committee, along with a member of the Preaching Committee, Abu Hafs al-Somali – led the mass defection that included a local security official and 27 members from Abyan, Aden, Shebwa and Hadramout. They stated that they remained loyal to IS leader Abu Bakr al-Baghdadi, but IS in Syria rejected this. “Dissent in the Islamic State’s Yemen Affiliates: Documents, Translation & Analysis”, Aymenn Jawad Al-Tamimi, 29 February 2016, http://www.aymennjawad.org/2016/02/dissent-in-the-islamic-state-yemen-affiliates.Hide Footnote

IS continues to carry out suicide bombings in the government-controlled cities of Aden and Mukalla and also engages in assassinations of local security and intelligence personnel that have a significant impact. After the UAE-led forces retook Aden from the Huthi/Saleh bloc in July 2015, these killings increased, the majority claimed by IS’s media channels. This and repeated IS suicide bombings of military recruitment centres and mass gatherings of soldiers collecting salaries have led many southerners to view IS as part of a historical pattern in which northern political elites use violent jihadists as a tool in asymmetric warfare against the south.[fn]The northern political elite is viewed by many southerners as comprising factions on both sides of the post-2011 divided regime. This includes individuals such as current Vice President Ali Mohsen al-Ahmar, Islah party figures and, in the view of some, even President Hadi. Crisis Group consultant observations in former capacity; and interviews, Hiraak activists, Aden and Mukalla, August 2015-December 2016.Hide Footnote

AQAP’s strategy of gradualism, relative respect for local norms and integration with local Sunni populations has been far more successful. Its longstanding presence and well-established networks across the country have given it a clear upper hand.

D. Salafi Militias

While AQAP and IS dominate headlines, a range of Salafi militias are an increasingly important part of Yemen’s Sunni militant milieu.[fn]Labelling individuals and Yemen’s rapidly expanding number of armed groups is fraught with problems. Fault lines between groups are increasingly unclear and many have no visible structure. Individuals and fighting factions often assume tribal, political and religious identities simultaneously. Attempts at simplification are often misleading while making any future reintegration of armed groups problematic. The label Salafi is similarly vulnerable to these errors and underlines the need for disaggregation. “Salafism” is used here in its broadest sense: a Sunni movement that seeks to revive “original” Islam by drawing on the so-called pious ancestors (salaf al-salih). Increasing fragmentation throughout the civil war has resulted in open conflict between Salafi strands.Hide Footnote Yemen has long housed a variety of Salafi groups.[fn]The Salafi spectrum there has historically comprised three main strands: quietist, jihadist and activist (harakiya). Quietist, scholastic or missionary Salafis are apolitical, reject parliamentary politics and, in theory, give allegiance to existing authority. Salafi-jihadists advocate violence against religious and political enemies. And activist Salafis are more inclined to challenging authorities through the political process. Crisis Group consultant phone interview, Laurent Bonnefoy, July 2016. See Laurent Bonnefoy, Salafism in Yemen: Transnationalism and Religious Identity (London, 2011). See fn 1 for Crisis Group’s definition of “jihadist”.Hide Footnote Prior to the war, most were non-political and non-violent. Some, like the al-Rashad party, embraced politics and were closely associated with Islah. As the Huthis expanded southward, however, many took up arms against them. The earliest indications of this were in 2013, when the Huthis fought Salafis from the Dar al-Hadith religious institute in Dammaj, Saada.[fn]Muqbil al-Waddii, a Saudi-educated Saada cleric and convert from Zaydism, established Dar al-Hadith institute in the heart of Zaydi territory in the late 1970s with funding from Saudi Arabia. Salafi proselytising there, itself arguably a product of socio-economic grievances against advantages given to Zaydi elites, particularly Hashemites (descendants of the Prophet), is a core grievance that sparked Zaydi revivalism in the 1980s and later gave impetus to the Huthi movement. Crisis Group Middle East Report N°86, Yemen: Defusing the Saada Time Bomb, 27 May 2009.Hide Footnote Although the Huthis emerged victorious in January 2014, fighters from Dammaj and another religious institute in Kitaf, Saada, regrouped and are now fighting the Huthis on a number of fronts.

Opportunistic alliances forged by the Saudi-led coalition have propelled Salafis to prominence. In Aden, they act with UAE support as state-sponsored, irregular security forces. As the battle for that city reached its peak in July 2015, the UAE worked with Hashem al-Syed, a former Dar al-Hadith student, to lead Salafi fighters there. After Huthi forces were pushed out, another little-known Salafi, Bassam Mehdhar, became the UAE’s main beneficiary. In 2015, the al-Mehdhar Brigade, based in Sheikh Othman and Mansoura districts, acted as a local security force. In October 2016, the group joined other Saudi-supported Yemeni forces in crossing the Saudi-Yemeni border in an attempt to push into Saada, the Huthi stronghold.[fn]This new battlefront is a very personal fight for many Salafis, who are seeking revenge for being forced out of the Dar al-Hadith institute.Hide Footnote

Another group, the Security Belt forces, a UAE-supported militia established by presidential decree in May 2016 to help secure Aden and led by Nabil Mashwashi, a former South Yemen army commander, appears to have a significant Salafi component. Prominent Salafis, such as Hani Bin Baraik, a minister of state and figurehead of the group, tend to be anti-Islah which, just as their UAE backers, they suspect of collusion with AQAP.[fn]Crisis Group interviews, coalition member, Adeni journalist, Dubai, September 2016. Many Salafis in the south are anti-Islah also because they see the party as northerners opposed to the south’s secession.Hide Footnote

In Taiz, the lines between Salafi and AQAP/AAS fighters are blurred.[fn]A former AQAP member said: “Taiz has been a special case in the Yemen war because AQAP has been able to successfully embed itself in the opposition to the Huthis there from the beginning, but those fighting are not AQAP core members”. Crisis Group interview, September 2016.Hide Footnote As in Aden, Salafis are at the forefront of Saudi-led coalition-sponsored efforts to repel Huthi advances. Among the city’s myriad factions, one of the more notorious groups, acting with UAE-supplied weapons and armoured vehicles, is led by another former Dammaj student, Adel Abdu Farea, also known as Abu al-Abbas. His followers have included the Humat al-Qi’dah (Protectors of the Faith), a group responsible for the destruction of a seventeenth-century tomb in July 2016 and the burning of books belonging to Yemeni Christians. Al-Abbas’s men have clashed with another group led by a Salafi sheikh, Sadeq Mahyoub, which is loyal to the Saudi-sponsored tribal sheikh Hamoud al-Mekhlafi. Saudi Arabia also sponsors the al-Hassam (Decisiveness) Brigade, which was founded by Adnan Ruzayyk al-Qumayshi, who left Dammaj in 2014, and is now led by Salafi figure Ammar Jundobi.[fn]Crisis Group consultant phone interviews in former capacity, local residents, political activists, journalist from Taiz and Aden, August 2016.Hide Footnote

Allegations abound regarding possible AQAP/IS links to a number of Salafi groups, but the exact connection is unclear.[fn]Yemeni observers have linked the al-Hassam brigade to AQAP. Crisis Group interviews, Taiz politician, August 2016; prominent Sheikh from the Hashid tribal confederation, September 2016. Others claim links between AQAP or IS and Abu Abbas, with similar observations about connections with other individuals and factions. Crisis Group consultant phone interviews in former capacity, Taizi activists, journalist, September 2016.Hide Footnote Since December 2016, Salafi and other resistance militias have nominally been integrated into the Yemeni army while remaining separate in reality.[fn]Crisis Group interview, Ammar al-Jundobi, Decisiveness Bridge field commander, Taiz, January 2017.Hide Footnote

Yemen’s Salafi movement is undergoing a rapid transformation, shaped by the war and new sources of patronage. It is unclear how relationships between AQAP and various Salafi groups will develop and what, if any, political ambitions the latter have beyond defeating the Huthis. Their growth into a pivotal player in the civil war elevates the need for their representation in any political resolution, especially if they are to play a role as an alternative, among religious conservatives, to AQAP or IS. Thus far, the growth of Salafi militias appears to be feeding into AQAP’s narrative of a Sunni defence against the Huthi takeover, while contributing to AQAP’s aim of blurring the lines between AAS, its local insurgency arm, and Salafi groups in areas where they have been fighting alongside each other.

The UAE complicates this picture: it supports Salafi groups and, according to numerous Yemeni sources, has tried to either suppress or marginalise Islah because of that party’s links to the Muslim Brotherhood, which it bans domestically.[fn]Crisis Group interviews, GPC supporter, December 2016; Taizi politicians, November 2016; Adeni journalist, September, December 2016. Crisis Group consultant phone interviews in former capacity, Islah-aligned, pro-unity Mukalla residents, October 2016.Hide Footnote Islah, like Islamist groups actively engaged in politics across the Arab world, can play an essential role as firewall against radicalisation in Yemen. While the UAE officially supports its inclusion in any political settlement, its intolerance of Islah in practice risks pushing young men who might have chosen politics into the arms of the very violent jihadist groups the UAE wants to quash.

IV. Reversing the Gains

Yemen’s war has created vast new opportunities for AQAP, a relatively minor player before 2011, and has given rise to the Yemeni wing of IS. While U.S. drone attacks and other military action have dealt repeated blows, AQAP is thriving in the context of state collapse, sectarianism, shifting opportunistic alliances and a war economy, with fresh recruits and more sources of weapons and income than ever before.

At a regional level, the undertow of the Saudi-Iranian rivalry drives sectarianism and incites radicalisation on both sides of the war. Without de-escalation between them, it risks becoming an extension of a wider competition between, on one side, Iran and its allies in Iraq, Syria and Lebanon – with Russia also playing a role – and the predominantly Sunni powers of the Saudi-led coalition, backed by Western states, including Israel. Dialling back this regional enmity is a vital priority.

Within Yemen, achieving a sustainable ceasefire and nationwide political settlement should be the priority. To reverse the growth of AQAP and IS will require a political settlement that is truly inclusive, provides a mechanism for addressing demands for local autonomy and outlines interim security arrangements that are accepted by local communities while operating under the umbrella of the state.

Including a range of Sunni Islamists, particularly Islah and Salafi groups prepared to engage in politics, in power-sharing arrangements would give them a stake in national politics and a viable political outlet as opposed to marginalising them and potentially pushing some toward violent jihadism. Many Yemenis have turned to violence because they view a Huthi/Saleh-dominated state as a threat to their survival. Overcoming zero-sum perspectives requires, as a first step, a compromise in which each side can be part of the government and security apparatus.

Addressing demands for regional autonomy would also be crucial to rolling back AQAP. The group has astutely carved out space in political battles between opposing regional sentiments by forging de facto alliances with other Sunni groups against Huthi/Saleh forces and embedding themselves in the war economy. For their part, Huthi/Saleh forces have used AQAP and IS as a convenient excuse to advance into predominantly Sunni territory, thereby exacerbating the problem.

Southerners in particular routinely downplay the AQAP/IS influence in southern areas, viewing them as tools in the hands of powerful northerners, particularly Saleh and Mohsen.[fn]At times, this perception has served as a powerful motive for locals to assist in evicting AQAP from their areas, and the UAE-backed offensive in southern Hadramout in April 2016 utilised it. This local view of AQAP greatly contributed to the success of efforts by intelligence agencies in routing members who went into hiding after the April offensive. Hadramout residents, through information they provided to security forces, contributed to the capture of several leading figures, including AQAP’s emir of Ash-Shihr in May 2016 and a senior ideologue in Hadramout in October 2016, and the discovery of weapons caches. Crisis Group consultant interviews in former capacity, Hiraak activists, Mukalla, March 2016.Hide Footnote Assassinations of southern intelligence and security officials, both before the Saudi-led intervention and after the Huthi/Saleh forces’ removal from southern governorates five months later, have heightened their suspicions of them as an extension of a historical pattern of instrumentalising Islamists for political score-settling.

At the same time, AQAP and IS have exploited opportunities in the south when the Huthis were the proximate enemy. Even after the latter’s defeat there, some parts of the southern resistance fought with AAS against Hadi’s forces, which many viewed as corrupt.[fn]From February to March 2016, AAS and renegade fighters from the southern resistance fought together against Hadi forces in the Mansoura district where AAS had a base. Three of them, who were listed by the Hadi government as wanted AQ members, acknowledged they were fighting alongside AAS but denied being members. They cited corruption and cronyism as primary grievances against the government. The jihadists were eventually pushed out, with some leaving through negotiations. Crisis Group consultant interviews in former capacity, Aden, March 2016.Hide Footnote These dynamics, which allowed AQAP/AAS to capitalise on local animosity toward the central government, are likely to continue as long as demands for regional autonomy remain unaddressed.

Clear interim security arrangements are a critical part of any effective settlement. This issue has been a sticking point in UN-led negotiations. The Hadi government and its GCC backers insist on close implementation of UN Security Council Resolution 2216 (2015), which calls on Huthi/Saleh forces to withdraw from areas they captured and to disarm. The Huthi/Saleh bloc, however, argues for the establishment of a new government first, of which they would form an integral part, and then the withdrawal and disarmament of all militias, not just their own.

After 22 months of war, the Huthi/Saleh bloc emphasises that there is no functioning state led by the Hadi government to hand authority to. Even in areas “liberated” from Huthi/Saleh control, authority is diffuse, often resting with local militias. As such, abrupt withdrawals of Huthi/Saleh forces from areas they control could open a security void for AQAP and IS to exploit. Yet, their continued presence in contested areas and dominance in the north, to the exclusion of other constituencies, exacerbates communal tensions that radical groups could also take advantage of.

In the short term, Yemen needs clear interim security arrangements that are tailored to local political realities. Areas like Taiz will be most difficult to tackle, as warring forces are positioned in close quarters on the battlefield and can each claim acceptance from certain parts of the local community. Bringing together locally accepted combatants under the umbrella of local authorities acting on behalf the state would be ideal, at least until the overarching issues of military-security reform and disarmament, demobilisation and reintegration can be addressed nationally.

Reaching an inclusive overall settlement to end the war offers the best chance to undercut AQAP’s prospects, but such a settlement may not be feasible. Even if the UN is able to broker a deal, it is unlikely to result in a quick end to this multifaceted conflict with regional dimensions. Yemenis and their external backers should, therefore, look for measures, such as those that follow, that could still go some way toward curbing AQAP/IS expansion.

Addressing state orchestration of jihadist groups. AQAP and IS are part of a domestic/regional power struggle marked by shifting alliances in which they tend to be no one’s primary enemy. Following the 2011 political unrest and the Huthis’ 2014-2015 military advance southward in alliance with Saleh’s forces, the jihadists’ traditional channels of influence and co-optation became divided, leaving AQAP activities theoretically open to exploitation by various regime and ex-regime forces in the ensuing civil war. Because of this, Western governments and regional states should continuously re-evaluate the motivations of Yemeni actors and their external backers to counter AQAP/IS. While it is often difficult in Yemen to discern who may or may not be using AQAP/IS to their advantage, states involved in the conflict should be willing to regularly assess their partners and put pressure on them to change course if they are found to be tolerating or encouraging the growth of AQAP/IS to achieve tactical objectives.

In addition to routine evaluation of allies’ actions, it is important in principle to decouple development assistance from counter-terrorism aid to the Hadi government and any other government that may emerge from a political settlement, thereby limiting the extent to which they can benefit financially from AQAP/IS’s presence. Failure to acknowledge and address this state-jihadist interaction would likely further alienate the population, which often views AQAP, or more recently IS, at least partially as political leaders’ tools, from policymakers.

Improving governance in areas previously under, or vulnerable to, militant control. A key aspect of AQAP’s ability to gain initial acceptance has been its proven pragmatism: instituting effective governance and addressing a long-neglected population’s pressing concerns. It has prioritised providing security, basic services and a judicial system able to resolve grievances, such as long-running land disputes, showing itself as a viable, better alternative to the state.

For a post-conflict government to successfully counter AQAP, it would have to provide better governance, with local ownership over decisions, in areas previously under AQAP control and/or that are vulnerable to them. As a priority, this includes quick and non-corrupt dispute resolution, security provision rather than predatory score-settling and basic services such as electricity and water.

Aden stands out in many ways as an example of what not to do. When Huthi/Saleh fighters were evicted in July 2015, the exuberance of military victory was not followed by a similar enthusiasm for instituting governance. More than a year later, local authorities reestablished the city’s central prison and opened unofficial detention facilities, but failed to put in place a functioning court system. Even as some residents are supportive of the governor’s and security director’s efforts to restore order, many complain of corruption by government officials and of continued insecurity, a function of a security service divided along intra-southern lines of competition.[fn]Security services under the Aden governor and security chief function as a state within a state. Tensions are high between them and forces aligned with Hadi and his interior minister, Hussein al-Arab. The Security Belt forces technically fall under the interior minister, though the degree to which he controls them is questionable. All factions are competing for Emirati patronage and support. Overlaying the competition is a historical division from the 1986 civil war in which a group from the current governorates of Dalia and Lahi won over their adversaries, mainly from Abyan and Shebwa. Hadi and al-Arab are associated with the latter, while the Aden governor and his security chief are associated with the former. Crisis Group consultant interviews, more than a half a dozen Aden residents, the Aden governor and Aden security chief, Aden, January 2017.Hide Footnote

The experience in Abyan is a cautionary tale of how working through local militias without a clear plan for incorporating them into the state security forces or deploying them to help stabilise areas retaken from jihadists can backfire. Local militias known as popular committees were central to the U.S.-supported pushback against AQAP/AAS in Abyan and Shabwa in 2012. Their use yielded the short-term gain of driving AQAP out, but as the only force in charge of security they contributed to local tensions, entrenched exclusionary patronage networks and were vulnerable to infiltration by violent jihadists, who used them as a cover to establish networks for future resurgence. As a Zinjibar farmer put it, “I would not trust the Popular Committees even to watch over my goats”.[fn]The farmer said he knew Popular Committee members who had been AAS members. Crisis Group consultant interview in former capacity, Aden, June 2012. There were exceptions, notably in northern Abyan areas such as Lawder, where local militias were well-liked and more representative of local communities. Crisis Group consultant interviews in former capacity, local residents, Lawder, May 2012. For more information, see Iona Craig, “End of Emirate?”, IHS Jane’s Terrorism and Insurgency Centre, September 2012.Hide Footnote

After several cycles of evicting AQAP/AAS from towns in Abyan, only to see them return, residents still complain of the government’s lack of attention to services and governance. According to a resident of Jaar, the UAE and its local partners, most of which seek southern separation, had some success in pushing AQAP into the neighbouring governorate of al-Bayda in 2016, but there are still far too few government-provided services and an unmet desire for governance.[fn]Crisis Group phone interview, Jaar resident, September 2016.Hide Footnote

Mukalla appears to show improvement. In April 2016, locally recruited ground forces and UAE troops, with U.S. support, retook territory through what appeared to be a largely negotiated withdrawal of AQAP forces that spared the population a bloody battle. Some Hadramis say that the population is now working with the UAE and local security forces to identify and apprehend remaining AQAP supporters in the city and that services, such as water and electricity, are working well.[fn]Crisis Group interviews, Hadrami politicians, Abu Dhabi, October 2016; phone interviews, Mukalla resident, November, December 2016.Hide Footnote Still, the success was almost immediately followed by accusations of corruption, cronyism, arbitrary detention and torture by the new authorities and security forces. Non-secessionist residents and those politically aligned to Islah continue to complain of unlawful arrests, torture and disappearances at the hands of the new Elite security forces (a group similar to the Security Belt in Aden that is composed of local fighters trained by the UAE) and UAE troops.[fn]Crisis Group consultant phone interviews in former capacity, Islah-aligned, pro-unity Mukalla residents, May, June and October 2016. See also ‘“We lived in days of hell”: Civilian Perspectives on Conflict in Yemen’, Center for Civilians in Conflict, January 2017, p. 31.Hide Footnote

Disaggregating rather than conflating Sunni Islamist groups. AQAP is an internally diverse organisation with varying layers of support among the local population and shifting alliances. Its efforts to blend in with the larger Sunni community and to ease affiliation requirements (especially the loyalty oath) expand its influence even as they leave it vulnerable to efforts to peel off supporters motivated less by its global agenda than by local political or economic grievances.

Protagonists on both sides of the war have at times been quick to label a wide range of Sunni Islamists – from Islah to various Salafi and other fighting groups – as AQAP, instead of acknowledging clear differences between them. As a southern fighter from the “February 16” militia (one of many southern insurgent groups), listed by Aden’s security directorate as a wanted AQAP member, stated:

We are not al-Qaeda but joined with them to fight [Security Director] Shalal because we have no choice. We fought and died for our city for six months and they offered us nothing in return. They gave positions to their friends and families, stole money meant for us and treated us like garbage to be thrown away or burnt.[fn]Crisis Group consultant interviews in former capacity, “February 16” fighters, Mansura, Aden, March 2016.Hide Footnote

In addition, there is far too little effort to disaggregate AAS rank and file from AQAP core members. As a politician from Abyan noted:

Ansar al-Sharia was born of al-Qaeda but is different. Most Ansar followers in Abyan are local. Many are young men who are very poor with no prospects. You can strike agreements with them and pull them away from al-Qaeda. After al-Qaeda was removed from Abyan, Ansar supporters stayed behind. It is important to give them [political and economic] opportunities.[fn]Crisis Group phone interview, September 2016.Hide Footnote

Understanding who can be negotiated with and convinced to peacefully participate in political and social life is tricky and a shifting target that requires buy-in and expertise from local communities. But in Yemen’s fractured political environment, it is a critical component of limiting AQAP’s growing reach.

Using military tools judiciously. At times military force is a necessary component of confronting AQAP/IS, but it should be used judiciously, in coordination with local actors and in a way that respects local laws and norms, lest it produce a political and social backlash to AQAP/IS’s advantage. Military campaigns, whether carried out by Huthi/Saleh forces or the government, that have targeted real or alleged AQAP/IS operatives with conventional force have often devastated local infrastructure and communities, while arguably setting back the cause of curtailing AQAP/IS influence. A tactic that has proved more effective is the threat of force combined with local negotiations with militants to encourage core AQAP supporters to leave areas, particularly cities, thereby sparing population centres widespread destruction and taking the fight against combatants unwilling to negotiate to less-populated areas. For the most part, this happened in Mukalla, and with considerable success.

The type of force used against AQAP/IS is also important. In Yemen, foreign troops, particularly Western ones, and even fighters from a different region of the country, risk antagonising local populations that view them as invaders. Even when local fighters are used, they can become part of the problem if they are operating outside of a clear legal framework. Many local residents saw as predatory some Popular Committees in Abyan, Security Belt forces in Aden and Elite forces in Hadramout.[fn]Crisis Group phone interview, Aden resident, November 2016; consultant phone interviews in former capacity, Islah-aligned, pro-unity Mukalla residents, October 2016.Hide Footnote

Drones, too, should be used judiciously and in coordination with Yemeni authorities so as not to violate sovereignty. While the Hadi government and the civil war have given the U.S. virtual carte blanche to pursue its drone campaign, tactical success in killing key operatives and ideologues has not stopped the organisation’s rapid growth. Their use raises the additional risk of replacements becoming increasingly hard-line. For example, al-Wuhayshi’s successor, al-Raymi, is widely regarded as far more ruthless than his predecessor.

While drone strikes’ impact is difficult to assess, many Yemenis suggest that they are counterproductive, breeding anti-U.S. and anti-Yemeni government sentiment when civilians are killed, which can radicalise victims’ families. This is especially the case with U.S. “signature” strikes that are based on patterns of behaviour, without knowing the identity of the targeted individual.

What went wrong in the deadly raid on al-Qaeda in Yemen?

Richard Atwood, Director of Crisis Group's New York Office, speaks with Hari Sreenivasan about the 29 January U.S. raid on al-Qaeda in Yemen. PBS NewsHour

The first military actions by the Trump administration in Yemen bode poorly for the prospect of smartly and effectively countering AQAP. A 29 January 2017 U.S. Special Forces raid in al-Bayda governorate, a critical battleground between pro- and anti-Huthi forces, killed a U.S. commando and several prominent tribesmen associated with AQAP, but also according to local sources many civilians, including at least ten women and children.[fn]“Yemeni civilians killed in first US raid under Trump”, Al Jazeera English, 31 January 2017.Hide Footnote The use of U.S. soldiers, high civilian casualties and disregard for local tribal and political dynamics – many killed were local tribesmen motivated by the internal Yemeni power struggle as much as or more than AQAP’s international agenda – plays into AQAP’s narrative of defending Muslims against the West and could increase anti-U.S. sentiment and with it AQAP’s pool of recruits.[fn]

V. Conclusion

AQAP and IS are unique in being the only two of many warring parties in Yemen which are at least ostensibly enemies of the conflict’s two main antagonists: Hadi-aligned forces and the Huthi/Saleh bloc. They are also in the incongruous position of being the conflict’s greatest beneficiaries, thriving in the context of state collapse, growing sectarian polarisation, fluid alliances and an expanding war economy. They are part of a regional trend of religiously-justified violence that is making conflict resolution evermore elusive. Yemen’s regionalised civil war shows little sign of abating. Instead, this multifaceted struggle looks set to deepen confessional divides – not previously a focal point of conflict – to the benefit of AQAP and IS and detriment of the country, its people and global security.  

Brussels, 2 February 2017

Appendix A: Map of Yemen

Map of Yemen International Crisis Group/Based on UN map no. 3847, Rev. 3, January 2004.