Syrian President Bashar al-Assad (left) talks to government soldiers. (SANA/AP Photo)

The Islamic State’s spectacular killings of American journalists, persecution of various non-Sunni religious groups and victories against Kurdish forces in Iraq have overshadowed its activities in the lands where it claimed its first victories — northeastern Syria. Yet the struggle continues there, and the manner in which forces on the ground are resisting the Islamic State is laying bare alliances among local forces and the Syrian government. Many of these alliances accord with political science orthodoxy about civil wars, while others raise important questions about how civil war processes should be understood.

Groups organized around regional identities ­– between the national level narrative of sectarian conflict and town or neighborhood level enmities ­– play a central role in the fighting in Syria’s northeast, a role that has not been extensively studied by political scientists. The continued importance of these regional identities during the conflict suggests a degree of structure and continuity to the fight for Syria’s northeast that should surprise observers of the conflict and political scientists studying civil wars alike. My observations about conflict in Syria’s northeast and the enduring importance of sub-national regions are based on a year spent interviewing Syrians residing in neighboring countries.

Any policy calculation about arming opposition groups within Syria or other forms of intervention must take into account regional identities and the local relationships underlying them. Many actors fighting alongside the Syrian government are doing so only to defend their locality, and the same can be said of groups pledging loyalty to the Islamic State or the Free Syrian Army. This point should give policymakers pause in expecting aid to localized forces allied with a particular side in the conflict to produce desired change on a broader scale.

The political science literature on civil wars has begun to examine local dynamics. In his pathbreaking work on violence in civil wars, Stathis Kalyvas has suggested that the “master narrative” of a conflict ­– that is, what the two major sides in a conflict stand for and what they are fighting over ­– often maps poorly onto local patterns of violence. These local patterns have a highly endogenous character, meaning that they do not follow directly from preexisting factors like the ethnicity or economic class background of the actors, but take shape in response to events occurring on the ground. Fotini Christia has pushed the importance of local, material interests one step further, arguing that the alliances animating civil wars, as well as the groups comprising them, are “not merely imagined but rather constantly reimagined communities” reflecting the material self-interest of forces on the ground.

In the case of Syria, the master narrative is that the conflict pits Sunni Muslims ­– whether those desiring liberal governance like the Syrian National Council or those championing religious rule like the Islamic State ­– against a government controlled by one particular religious minority, the Alawites. If this master narrative was operating on the ground, we would expect to see all Sunni Muslim groups fighting together against the Alawite government (and, potentially, its allies in the country’s other minority groups).

The cracks in the master narrative are manifest. For one thing, the Islamic State has come into conflict with virtually all of the largely Sunni Muslim groups fighting the Syrian government, including the Free Syrian Army and the al-Qaeda affiliated Jabhat al-Nusra. In addition, there can be little question of the endogenous character of events in Syria’s northeast. The Islamic State itself emerged from elements of rebel groups fighting the Syrian government and groups coming from abroad. Neither the identities of, nor the alliances between groups fighting the government, could have been imagined before the beginning of the uprising, as none of them existed in Syria until fighting commenced.

Also emergent in the conflict are several new actors formed in response to weakened state control over much of the northeast. First, the most powerful Kurdish party, the Democratic Union Party (PYD) has set up its own army, the People’s Protection Units (YPG). The Syrian government withdrew from most of the Kurdish areas of the country in July 2012, which allowed the PYD to control the territory. Members of the other principal non-Arab ethnic group in the northeast, Syriac Christians, have set up a security organization called Sutoro. Far smaller than the Kurdish forces ­– unsurprising given the relatively smaller size of the Syriac population and its lack of a political apparatus equivalent to the PYD ­– Sutoro coordinates its activity closely with the PYD.

In addition, National Defense Forces (NDF) militias operate in areas outside of PYD and Islamic State control. These militia groups defend the locality from which they are recruited, rather than venturing to faraway frontlines, and are organized under the umbrella of the Syrian government. Their emergence has been a gradual response to increasing violence in Syria. When violence first began to threaten their neighborhoods, residents of many Syrian cities set up popular committees (lijan shabiya) to defend their local areas. As fighting deepened, these groups were turned into NDF militias. The militias receive uniforms and arms from the central government.

Members of the opposition often speak of NDF brigades interchangeably with shabiha (a term used to deride Alawite thugs looting and plundering parts of the country), but NDF forces are at least as much a regional as a sectarian phenomenon. Many branches of the NDF are comprised of members of the Alawite religious sect to which President Bashar al-Assad belongs and some fight away from their home area. The great majority of NDF brigades, however, were created to defend local communities and many of the local communities remaining loyal to the Syrian government are Alawite. NDF brigades thus reflect the ethnic make-up of the population from which they come. In northeastern cities like Hasakah and Deir al-Zour, which have significant populations of Sunni Arab tribal background, NDF brigades come from Sunni Arab tribes and are organized along tribal lines.

The “master cleavage” of the Syrian conflict suggests that the Kurds and tribes (both of which, religiously speaking, are nearly all Sunni Muslim) should align against the incumbent, minority-held Syrian government. Syriac Christians, by contrast, should seek the state’s protection. The micro-level theories suggest that, because no group is powerful enough to dominate the others independently, alliances between fighting groups should be fluid and the identity-based justifications for them (e.g. “We are Arabs combating a Kurdish threat,” “We are Sunnis fighting unbelievers”) ephemeral. On the ground in Syria’s northeast is something between these extremes; identities below the level of the Alawite-Sunni division and above the level of local grudges between neighborhoods are a driving force in the current conflict.

The fighting in Hasakah, the epicenter of the Islamic State’s bid to enlarge its area of control, exemplifies this dynamic. Hasakah, the capital of Syria’s far northeastern province of the same name, is the intersection of several social groupings. The majority of the population is of Arab tribal background, and the city has significant Kurdish and Christian populations as well.

Historically, the Baathist Syrian government has ruled Hasakah as a colonial power might: By working through intermediaries and treating the groups the intermediaries represent differentially. The Syrian government never sought a direct relationship with its subjects in the northeast, preferring to deal with local leaders and important tribal figures. The name of the longtime head of security for Hasakah province, Mohamed Mansoura, immediately calls to mind for many Hasakah residents the government’s tactic of pitting local groups against one another. This pattern of control endures during the current conflict; Mansoura, who had been promoted to head of General Security for all of Syria, was called back to serve in Hasakah during the conflict, this time as governor.

In addition, the Syrian government marginalized and discriminated against Kurdish people while cultivating good relations with many of the tribes of the northeast, their Sunniness notwithstanding. The official ideology of the Syrian government is Arabist, denying the existence of Kurds as a component of the nation. Symbols of Kurdish identity, including flags and teaching of the Kurdish language, were prohibited, and Kurds were denied all but menial positions in the military and civilian administration. By contrast, a significant number of the Arab tribes were loyal allies of the government in the confrontation with the Muslim Brotherhood in the 1980s and stood again by the government while it put down a Kurdish uprising in the region in 2004. The Syrian government continues to cultivate the support of the tribes of the northeast. It recently dropped leaflets over tribes it is trying to keep from siding with the Islamic State, with the following written on them: “A note from the Syrian Arab Army to the heroic men of the tribe, to whose place in the Euphrates valley history testifies. Respect and esteem are sent to you from your Army for the heroism you have shown and we call on you to continue in it.”

In the present conflict, this intermediate, regional level of social relations between the state and local populations continues to structure the flow of political events, as a recent battle for control of a major military base in Hasakah shows. In late July, the Islamic State began an assault on the headquarters of the Maylabiya Battalion (Fawj al-Maylabiya). Army forces in the base were unable to withstand the assault and withdrew, leaving a number of soldiers stranded in the base. With its nearest reinforcements also under siege, the government turned to the NDF to liberate the remaining soldiers, and the PYD sent its military forces (YPG) to fight the Islamic State there as well. Once the Islamic State took full control of Fawj al-Maylabiya, NDF and YPG forces surrounded the base to cut off supplies to the Islamic State fighters controlling it. In the aftermath of the events, the government set up a shared operations room coordinating the operations of the army, the state’s four security agencies, Sutoro, YPG and NDF forces, and a parallel tribal militia called the Dignity Army.

These surprising constellations of actors fighting the Islamic State are marriages of convenience, not based on ideology or personal ties of elites. There is no natural affinity between the parties. First, the official Baathist, Arabist doctrine denies that Kurds and Syriacs are legitimate residents of the ostensible Arab homeland they are defending alongside the Syrian “Arab” Army. Second, though the PYD is widely accused of collaborating with the Syrian government, there is no direct evidence of any agreement and the organization has all the reason to despise the government. The government had long hosted Abdullah Ocalan, the founder of the Kurdistan Workers’ Party (PKK) out of which the PYD emerged, and expelled him in 1998 when seeking rapprochement with Turkey. He was captured by Turkey soon thereafter, and the Syrian government began arresting and harassing Syrians associated with the PKK.

Each of the groups allying with the Syrian government needs the latter to protect the group’s population and the land it occupies. The Syrian government similarly needs the groups to help repel the greatest present threat to its power, the Islamic State. Tensions in these alliances ­– coming from the divergent interests and intermediate level identities of the groups ­– are beginning to show through. The newspaper al-Araby al-Jadeed reported that members of the Syrian security service were upset at the Hasakah NDF for its lackluster effort to save stranded government forces at Fawj al-Maylabiya. The security forces’ complaint was that, because the majority of the military officers at the base were of Alawite background, the NDF fighters were happy to have Alawites killed. Similarly, 200 NDF fighters in Hasakah recently turned on their commanders for assenting to the government’s decision to give the Kurdish YPG forces authority over the military facilities of the NDF. PYD checkpoints throughout the city and sweeps inside non-Kurdish neighborhoods have agitated local residents and caused NDF fighters to give up fighting in parts of Hasakah city.

Fighting in Hasakah has continued since the battle for Fawj al-Maylabiya and regional identities continue to structure the conflict. At present, the combination of PYD, NDF and Syrian government forces have succeeded in repelling Islamic State fighters from all but one neighborhood of the city. The government is coordinating with local notables to give fighters three choices: Come to a settlement with the government, leave the neighborhood or have the neighborhood stormed by government forces.

Two aspects of this tactic taken by the government speak to the importance of intermediate identities. First, the government took measures to avoid harming local residents. When rebel groups were similarly taking over neighborhoods of the central city of Homs and entire towns on Homs’s periphery in 2012, the government made a habit of indiscriminately shelling these areas. Second, the government asked local leaders to flush out fighters. This reliance on local social structures ­– including leaders who are ambivalent about both the Islamic State and the incumbent government ­– shows a degree of continuity with patterns of rule from before the uprising.

Two of the processes political scientists have identified in civil wars are clearly operating in Syria’s northeast. First, the set of alliances and tactics in Hasakah bear little resemblance to the ideological, black-and-white narratives offered for the conflict as a whole. Second, the alliances and patterns of contestation emerge out of events in the conflict itself, not just preexisting factors like religious or racial divisions, social classes or agreements among national and local elites.

Yet the dynamics in northeastern Syria sit uncomfortably with political scientists’ theories that assign combatants narrowly individualistic, material motivations. The various groups cooperating with the Syrian government are not maximizing their control of territory and resources but defending their territory, their region. As soon as Fawj al-Maylabiya fell, several groups rushed to prevent the Islamic State from taking control of the city. NDF and Sutoro forces secured their respective neighborhoods, and the YPG set up checkpoints on the periphery of the city and within many neighborhoods. The controversy over whether the NDF allowed Alawite army officers to be slaughtered by the Islamic State at Fawj al-Maylabiya further suggests the importance of region; NDF fighters rallied to hold a military base protecting their area but made little effort to save officers unconnected to their local area.

The regional identities along which these groups form and act ­– between national-level sect and individual-level self-interest ­– reflect not so much groupings made on the fly according to leaders’ self-interest or the peculiar motivations of actors (i.e. a Kurdish or tribal mindset driving group action) as the local networks and relationships compelling fighters to act. A Syriac Christian finds his most important social relations, from the neighborhood he resides in to his religious practice to his intimate life, overlapping with his religious identity. Thus, to defend these networks is to join Sutoro, the militia associated with the Syriac identity, and vice versa.

The first several years of the Syrian conflict have seen numerous surprising alliances and, in the rise of the Islamic State, the emergence of an entirely new social force. Many of the armed groups entering into alliances with the Syrian government arise out of existing social forces; some, like the Kurdish PYD, break with the historical stance of the social group they represent vis-à-vis the state, while others, such as the NDF comprised of tribal Arabs, exhibit continuity in the relationship of local groups and the government. Dynamics within the civil war may yet rearrange the alliances among groups in Syria’s northeast, but the intermediate identities rooted in the country’s regions will have enduring importance no matter who controls the state apparatus based in Damascus.

Kevin Mazur is a doctoral candidate in the Department of Politics at Princeton University. His research was supported in part by a Travel-Research-Engagement grant from the Project on Middle East Political Science.