Misalignment of International Security Interests and Intelligence Sharing: The U.S.-Türkiye-SDF Dilemma During the Syrian Civil War

The decade-and-a-half Syrian conflict served as a backdrop of competing international influences, proxy forces, and outright interventions, to say nothing of the various Syrian factions struggling for their own version of self-determination. Russian, Iranian, Israeli, Turkish, and U.S. forces all played various roles in both direct and proxy conflicts in the battlefields of the Syrian conflict over the future of the state (Phillips 2020). Further complicating the Syrian war, Islamic State of Iraq and Syria (ISIS) developed as the emerging regional threat, requiring coordination between otherwise belligerent actors. This required intelligence agencies across the conflict to coordinate in ways to eliminate the preeminent threat, while also maintaining a foothold in their own interests in the region.

Through this lens, the Syrian Civil War was not just a battlefield, but an intricate set of intelligence sharing objectives and obstacles. Formalized allyship between NATO countries and partnerships of regional or tactical importance needed to work together to eliminate threats and gain a strategic advantage. What then happens when the disparate parts of this spiderweb of intelligence contain coordination between actors who would, under normal circumstances, act as belligerents towards each other? In formalized intelligence-sharing relationships, fellow-NATO members like the United States and Türkiye had overlapping – but divergent – regional interests that allowed intelligence to be shared between the two countries. But one of the U.S.-supported proxy forces, the Syrian Democratic Forces (SDF), was seen as a threat to Turkish national security initiatives (Lindenstrauss and Shavit 2019). To what extent did competing or misaligned national security interests complicate intelligence sharing operations between de jure allies and de facto partners in the U.S.-Turkish-SDF triangle during the Syrian Civil War and adjacent conflicts? This conflict demonstrated a gap in intelligence sharing frameworks: among allies, intelligence sharing is essential for combating terrorism but poses operational and strategic threats among misaligned partners or competing national security interests. In the case of the U.S.-Turkish-SDF dynamic, this misalignment emphasizes the need for enhanced compartmentalization to protect U.S. (or other) national interests, partner-specific safeguards for source protection and intelligence dissemination, and clearly defined intelligence-sharing authority and allowable actions that allies or partners may pursue with that intelligence when nations may face competing national security concerns.

To better understand the divergent intelligence landscape, one should understand the role of the Kurdish people in the region. The Kurds are an ethnic minority spanning across a region encompassing Syria, Iraq, Iran, and Türkiye. The Kurdish people are the largest ethnic population without a dedicated state (McDowall 2004), which has given rise to an ethnonationalist movement to establish a Kurdish state in the region. This movement gained traction with the establishment of the Kurdistan Workers' Party (PKK) in 1978 by Abdullah Öcalan. In the mid-1980s, Öcalan’s PKK began militant action against Turkish military outposts as a way of establishing legitimacy (Marcus 2007). The Turkish government sees the PKK and their regional and offshoot groups as a national security threat and the PKK have been branded a terrorist organization by dozens of governments including the United States and Türkiye (Lindenstrauss and Shavit 2019).

While the PKK is classified as a terrorist organization by the United States, Türkiye, and others, the United States has utilized Kurdish militias as proxy forces in the past. During Operation Iraqi Freedom, the U.S. employed the Peshmerga – distinct from the Kurdistan Workers’ Party and financed by Iraqi Kurdish politician, Masoud Barzani – as a northern force in Iraq that could flank Iraqi forces while the U.S. invaded. While the Peshmerga and PKK are both Kurdish, their political objectives are vastly different and the two militias have had outright armed conflict in the past. So again, when the situation in Syria deteriorated, the U.S. turned to regional Kurdish forces, the Syrian Democratic Forces to further security interests (Biermann 2017).

With the emergence of ISIS as a threat to regional security beyond the borders of Syria, the United States’ tacit entry into the Syrian conflict came in the form of a temporary partnership between the United States and the SDF. This de facto partnership was established through training, funding, and materiel to support Kurdish resistance to ISIS rather than a formal, written treaty. Turkish history with armed Kurdish militant groups loomed over the ordeal. Türkiye perceived that this funding, training, and supply to Kurdish militants in addition to Öcalan’s call for worldwide Kurdish resistance from Turkish prison (Bianet 2014), could mean the establishment of a U.S.-backed Kurdish state in an already contested region. Bluntly, the establishment of an autonomous Kurdish state was a major national security concern in the eyes of Turkish intelligence and defense services (Lindenstrauss and Shavit 2019).

The Turkish skepticism was not unfounded or based solely along ethnic lines. The SDF is not a monolithic organization. It is largely a collection of militant groups aligned with U.S. interests, led by the People’s Protection Units (YPG) and Women’s Protection Units (YPJ), Kurdish militias who formed in the void left by retreating Syrian regime forces. Although the United States spent considerable political capital lobbying Ankara that the SDF was a distinct entity from the PKK, the firewall was not as cut and dry as those in Washington would have liked. The YPG and YPJ are considered distinct entities from the PKK based on the aims of their engagements; the PKK has engaged in asymmetrical, ethno-nationalistic hostilities with Turkish state actors, infrastructure, and civilians, and the YPG and YPJ have focused their attention towards self-determination in the Kurdish autonomous region surrounding Rojava. In the eyes of state officials in Ankara, that difference amounted to splitting hairs between two militias established for the sole purpose of Kurdish autonomy. Further complicating the picture, after the SDF defeated ISIS in Raqqa, Syria, the YPG and YPJ raised a flag bearing the portrait of PKK leader and Turkish prisoner, Abdullah Öcalan (Middle East Eye 2017).

Viewed through a principal-agent framework (Walsh 2010), the United States’ relationship with its partners can be viewed as supplying intelligence to on-the-ground forces in furtherance of national or regional security objectives. In the U.S.-Turkish-SDF triangle, the United States supplied signals intelligence (SIGINT) and geospatial intelligence (GEOINT) to the SDF on the whereabouts of ISIS militants. The SDF, in return, supplied the United States with on-the-ground human intelligence (HUMINT) and battlefield updates (Asharq Al-Awsat 2022; Katz 2020). The Pentagon officially acknowledged this intelligence partnership due to the SDF’s knowledge of the terrain and proximity to the fight (Youssef 2021). This semi-informal or temporary partnership created inherent friction with Türkiye, which saw the SDF’s relationship to the PKK as a national security threat. The United States also maintained a military intelligence cooperation program with Türkiye targeting PKK militants, including U.S. drone missions that collected GEOINT for Turkish operations (Pamuk and Stewart 2020). What happens when a principal shares intelligence in pursuit of a common objective – elimination of ISIS – while the recipient possesses independent incentives to use that intelligence toward a different objective? The United States maintained simultaneous intelligence relationships with actors whose objectives overlapped in counterterrorism – the SDF and Ankara – but diverged regarding Kurdish armed groups: Washington partnered with the SDF/YPG against ISIS while maintaining intelligence cooperation with Ankara against the PKK. Further, The New York Times reported that Türkiye may have used the information supplied to target not only the PKK, but also the SDF due to their proximity to and relationship with the PKK (Hubbard and Gall 2019). Again, Ankara did not necessarily see the intelligence shared by the United States as isolated to counterterrorism, but rather used it in furtherance of what it saw as an existential threat (Pamuk and Stewart 2020).

How was this gap allowed to exist? Lowenthal (2025) discusses the role of policymakers in the intelligence world. Policymakers can establish intelligence collection priorities, methods, uses, and, ultimately, treat intelligence sharing as a policy tool. Still, there should be a dividing line between policymakers and the actual intelligence analysts. The risk is that intelligence itself becomes politicized or somehow molded by the overarching political landscape. However, when policymakers are not abreast of the intelligence realities on the ground, misalignment of national security policies and information sharing obligations can mean that intelligence shared – even with formal allies – can take on ramifications all its own. Once intelligence is disseminated, the collecting party is no longer unilaterally able to control how that information is acted upon, creating operational and strategic risks for the United States, its allies, and regional stability. That is not to say that Türkiye acted “correctly” or “incorrectly,” but rather that Türkiye acted in its own best interests based on the information available, regardless of the plans or intentions from Washington D.C.

The Syrian case therefore offers a historical framework for examining a recurring problem: how can the United States share intelligence with de facto partners while accounting for the independent national-security priorities of its formal allies? This paper explores a three-pronged approach to better protect the United States and its partners and allies through compartmentalization, the establishment of partner-specific protocols, and through an agreed upon framework of usage restrictions that would prohibit the weaponization of shared intelligence against U.S.-aligned or U.S.-backed forces in the region.

Understanding the United States Intelligence Community’s (USIC) intelligence sharing obligations, guidelines, and complications is crucial to explaining the intelligence asymmetry of the Syrian conflict. The National Security Act of 1947 established the Director of Central Intelligence – amended in 2004 with the Intelligence Reform and Terrorism Prevention Act of 2004 to establish the modern Director of National Intelligence (DNI), which this paper will henceforth refer to – as the coordinating head of the USIC. Domestically, the DNI is the primary coordinator of cooperation and communication between agencies, a presidential liaison and advisor about U.S. national security issues and interests, and statutory authority on intelligence collection and sharing with the Joint Chiefs, the President, and executive branch advisors, among others (Lowenthal 2025). The role of the DNI extends beyond the borders of the United States, however. Under Title 50, the DNI has the authority to oversee intelligence-sharing relationships between the USIC and foreign governments. This authority also comes with obligations to keep sources and methods of intelligence collection balanced against operational success and national security interests (50 U.S.C. § 3024). Title 50 grants the DNI the institutional authority to coordinate counterterrorism intelligence sharing; the U.S.-Turkish collaboration demonstrates the complexities involved when national security interests are asymmetrical. Further, the authority of the DNI grants more than state-to-state intelligence sharing.

Title 50 also establishes the authority and oversight mechanisms for “activities of the United States Government to influence political, economic, or military conditions abroad, where it is intended that the role of the United States Government will not be apparent or acknowledged publicly” (50 U.S.C. §§ 3091–3093). Stated plainly, this statute provides the USIC with the ability to perform covert actions in support of national security interests. Providing materiel, financial support, and narrowly scoped intelligence to nonstate actors could fall under the purview of this statutory provision. Under this provision, tacit support for Kurdish militias in pursuit of ISIS could be considered a covert action, keeping the United States military at arm’s length from the battlefield.

Rarely though are such neat lines drawn between intelligence agencies and the U.S. military. The complementary Title 50 statute, Title 10, provides the U.S. military with its own statutory obligations and guidelines for operational authority. On the ground, conventional military activities and intelligence operations can be indiscernible, but their tasks, funding, and operational authority come from discrete sources, even if they overlap in practice. The U.S. military has the ability to operate under both mandates, complicating the intelligence-military distinction (Wall 2011). Syria provides a case study in the legal ambiguities inherent in a hybrid approach when intelligence agencies–who may or may not have their own hardware–rely on the U.S. military to provide them with the operational runway. As a hypothetical, sharing GEOINT gained from a U.S. military MQ9 Reaper with proxy SDF forces comes with authoritative and jurisdictional ambiguities: Who owns the collection? Who can disseminate what is collected? And ultimately, who is the operational authority? Wall (2011) points out that a gap–real or perceived–can lead to competition between both the U.S. military and its relationship with the USIC and among USIC constituents jockeying to protect their own funding or hard won intelligence.

At least part of the friction around sharing intelligence, domestically and with foreign governments, stems from what happens after information is shared. Once intelligence is shared, the metaphorical genie is out of the bottle; it cannot be put back in. The intelligence lives on, is referenced, analyzed, and synthesized into new forms beyond the boundaries of the originating party. To alleviate the interagency territorialism over intelligence sharing, the USIC has established rules and norms through the “Third-Agency Rule” outlined in Director of Central Intelligence Directive (DCID) 1/7 for interagency intelligence sharing and “originator control” established in Intelligence Community Directive (ICD) 403 for foreign dissemination. Under DCID 1/7, the “Third-Agency Rule” establishes that a USIC agency (party one) that shares information with another agency (party two) must be consulted before intelligence is forwarded onward (party three) (Director of Central Intelligence 1998).

In 2013, under James Clapper, the Office of the Director of National Intelligence put forth Intelligence Community Directive (ICD) 403 establishing the originator control doctrine for intelligence sharing to foreign intelligence agencies. Under ICD 403, there are three principles the USIC must take into account when sharing information with a foreign body: Does it serve a national security purpose; can the receiver reasonably protect the intelligence it receives; and will the receiving party adhere to not passing the intelligence along to third parties (Office of the Director of National Intelligence 2013). With the USIC sharing information with a formal, state actor like Türkiye, safeguards, such as treaties or formalized alliances, establish operational guidelines. If Türkiye passed the intelligence along to another state actor, Iran, for instance, they would be staring down serious diplomatic pressure from the United States and other NATO allies. Safeguards, formal and informal, however can only be trusted when national security interests are not categorically misaligned.

On the other hand, the Syrian Democratic Forces, as an amalgamation of militias, operates under a different framework. Passing intelligence to the SDF required a measured approach against the national security interests of the United States, of which ISIS was deemed one of the highest threats. Pentagon Press Secretary John Kirby acknowledged the ambiguity around the United States direct intelligence sharing with the SDF, but was “mindful of our obligations… about the rule of law” (Youssef 2021). The SDF’s ability to protect U.S.-provided intelligence could be reasonably debated given that they were located in the middle of a multi-front warzone consisting of ISIS and the broader Syrian civil war. Additionally, from Ankara’s point of view, the political closeness and operational overlap of the PKK and the YPG looked less like a state-to-state intelligence brokerage and more that the United States was providing intelligence capabilities directly to a group that it saw as its own national security-threatening terrorists (Vugteveen and Farrell-Molloy 2022). Whatever semantic safeguards were put in place, assurances or perception by the USIC that the YPG and PKK were distinct entities, were not enough to convince the security services in Ankara that there were not overlapping ideologies, practices, and leadership structure between the two Kurdish forces. Despite the formalities of Third-Agency or originator control, intelligence shared with SDF was assumed by Turkish intelligence services to have bled across the firewalls back to the PKK. The political and national security complexities of the modern battlefield with the behavior of the YPG/YPJ raising the Öcalan banner all highlight the difficulties of intelligence sharing even between two de facto allies (Center for Strategic and International Studies 2025).

So long as the United States and its allies seek to stay ahead of emerging threats, on the kinetic battlefield and intelligence landscape, intelligence sharing will be a cornerstone of that strategy. How, then, could the United States look to minimize these intelligence sharing asymmetries exemplified by the U.S.-Turkish-SDF triangle? Tear-lines provide an operational tradecraft as outlined by the Office of the Director of National Intelligence (2012). Under ICD 209, tear-lines in intelligence analysis attempt to fence the actionable intelligence from the method of collection. Under these guidelines, the USIC can begin to compartmentalize intelligence data beyond the dissemination authority established by both third-agency and originator control. In the case of the Syrian conflict, the United States providing the SDF raw GEOINT or SIGINT data collected from drones overflying the battlefield could reveal more than intended to potentially command structure overlaps between the SDF and PKK. Additionally, that raw data perhaps codifies the origin of the intelligence more so than the USIC might want, especially knowing that formal allies in Türkiye might have their own objections.

In practice, tear-lines isolate the actionable bit of intelligence from the sources and methods of intelligence collection. “Above the line” data could include collection platform, technical capabilities, and identifiable HUMINT sources. The “below the line” data is the raw, actionable data that could contain coordinates, plans, or other intelligence written in a way that has the ability to be shared. The ODNI’s (2017) updated guidelines instruct intelligence analysts to write with a focus of maximum sharability, moving focus from “need to know” to “duty to share.” ICD 208 recommends that intelligence data be written in a form that is easily redactable through tear-lines, and tailored for maximum utility. Applying this framework to the actors in the Syrian conflicts could provide utility to partners and allies, alike. For instance, providing the SDF with immediately actionable, time-sensitive, tactical data–ISIS fighters are at particular coordinates near Rojava–instead of strategic GEOINT of the area might ease the concerns in Ankara. This concrete intelligence supports the SDF partnership’s immediate counterterrorism goals, but the time sensitivity of it limits the ability for information to travel back across sympathetic YPG/PKK command structures. Similarly, providing Ankara with intelligence with SDF locations “above the line” would help limit unwanted conflict between Turkish security forces and Kurdish partners.

As hinted at in the above example, partner-specific protocols help protect the triangle between the USIC, its local partners, and its formalized allies. Walsh’s (2010) principal-agent model is helpful for contextualizing this, as well. When Washington shares intelligence with both Ankara and the SDF command structure about its primary security concern, ISIS, both have their own immediate security and survival priorities. As such, “defection” from the intelligence alliance becomes a concern. Stated more plainly, the motivations for self-determination override Washington’s primary objectives, stated or understood, with the shared intelligence (Walsh 2010). When states believe their intelligence sharing partners are not using actionable intelligence in the prescribed or intended way, intelligence sharing ceases to become an advantageous element of the relationship and instead drives inter-partner or inter-ally competition (Byman 2025). Partner-specific tear-lines and protocols can reduce the concerns of defection or competition.

Because the goal of intelligence sharing is to maximize both principal and agent value, duty to share should not be interpreted as sharing less intelligence or raw data. Partners and allies both need to have sufficient context in shared intelligence to support the broader shared mission. In the case of stopping ISIS, giving Türkiye or the SDF redacted communications may not be actionable intelligence. Partnership is still needed for local forces to carry out local missions. Tear-lines can compartmentalize individual pieces of shared intelligence, but given a sufficient body of work, the agent can begin to reverse-engineer collection methods. As a hypothetical, if the U.S. shares repeated on-the-ground HUMINT about ISIS with Ankara, it’s reasonable to assume there is an asset somewhere in the proximity. Joined with repeated GEOINT from drone flyovers, the security services in Ankara can begin to combine shared intelligence with their own generated intelligence to paint a more complete picture. Sharing safeguards can attempt to limit the information shared with partners, but tear-lines only limit the intelligence shared, not the follow-up actionability of the intelligence.

In addition to compartmentalization and partner-specific safeguards, intelligence sharing should carry with it specific limits around how and when the shared intelligence can and cannot be used. With formalized partners like the U.S. and Türkiye, NATO obligations such as the Joint Intelligence, Surveillance, and Reconnaissance initiative provide structured frameworks by which both parties share collection and analysis obligations (NATO 2026). The two allies have established a Combined Joint Operation Center for intelligence processing in the Şanlıurfa region of Türkiye (Jones 2019). But despite the third agency, originator control, or any additional constraints put on shared intelligence, ultimately the actionability of the data is up to the security services acting in their own self interest or preservation. The United States’ temporary partnership with YPG-linked militias in Syria does not outweigh the threats against national security posed by the same group in the eyes of Ankara.

The United States faces distinct problems of intelligence sharing with partner organizations, especially those with porous organizational boundaries or command structures such as the SDF. The United States may be able to put extreme guardrails on the use of intelligence with SDF liaisons and the SDF may even comply with those restraints. But those on the ground who are either sympathetic or outright collaborating with the PKK or other Kurdish militias could forward the intelligence onward. In this scenario, the SDF could formally play by all rules laid out in the intelligence sharing agreement, but still have political, ethnic, or opportunistic actors relay the intelligence beyond the agreed upon boundaries, causing ramifications that the United States or the SDF did not intend. As a hypothetical in this scenario, if the SDF were given information that Turkish troops would be in the area as a courtesy to clear, an SDF member could forward that intel to members of the PKK who could then stage an ambush against those Turkish troops. In an organization like the SDF where its member parties are other militias, the constituents of those other militias might have other motives beyond solely the elimination of ISIS.

These compounding factors of intelligence sharing, ethnic tensions, political sovereignty, and national security misalignment came to a head in October 2019. When Turkish forces launched Operation Peace Spring into northeastern Syria, the theoretical and hypothetical vulnerabilities of partnership-based intelligence sharing manifested into a kinetic response by Turkish troops. Strategic and security misalignment allowed the gap between the United States’ formal ally in Türkiye to attack its partner in the SDF. Operating under an understandable threat to national security, Turkish intelligence saw U.S. intelligence and military support to the SDF, and by default the YPG, as structurally inseparable from the PKK. Ankara utilized its own intelligence and military services to directly attack the U.S.-backed SDF positions (Hubbard and Gall 2019). The SDF’s associations with Öcalan and the PKK that Turkish intelligence had long warned about served as justification for an operation designed to subdue the autonomous, ethnically Kurdish region along the border that Türkiye viewed as an existential threat.

The immediate fallout demonstrated the limits of Washington’s dual-track intelligence strategy. The delicate balancing act of sharing counter-terrorism intelligence with the SDF while maintaining cooperative intelligence mechanisms with Ankara collapsed entirely. In the wake of the cross-border offensive, the United States was forced to abruptly halt its secretive intelligence and drone cooperation programs with Türkiye, exposing a profound fracture in a vital NATO intelligence relationship (Pamuk and Stewart 2020). The 2019 offensive laid bare the inherent contradiction of the principal-agent framework: when an institutional ally and an irregular proxy possess mutually exclusive survival imperatives, intelligence sharing ceases to be a tool of collective security and instead accelerates inter-allied conflict.

The United States’ utilization of local forces for proxy warfare is not going away. Local forces have not just a literal “home field” advantage, but also connections with the local population to provide valuable HUMINT back to Washington. The USIC must advance the mechanisms and communication around intelligence sharing through compartmentalization and partner-specific protocols and specific limits on how intelligence can be used in furtherance of national security interests that may or may not be aligned with the United States’ interests in the region. The tension between partners and formal allies risks not just the United States losing credibility with its proxy forces, but rather operational and strategic stability in the region (Honigman 2022). While the focus on this relationship is confined to the Syrian Civil War and offshoot conflicts with ISIS, the lessons learned have ramifications even today. In 2026, the United States publicly encouraged Kurdish militias to support its regime change operation in Iran (Bertrand et al. 2026), while Ankara has staunchly opposed this and provided Iran with the intelligence to identify Kurdish threats (Zaman 2026).