On 27 May 2026, a number of Sweida residents being held in Adra Central Prison announced an open-ended hunger and thirst strike in protest against their prolonged detention without being referred to the judiciary or formally charged with specific offenses.
In an audio recording smuggled out of the prison, one of the detainees said: “We are twenty young men from Sweida. We’ve stopped eating… Either we die here, or we get out.”
The strike came just three months after a “prisoner exchange” carried out on 26 February 2026, which resulted in the release of the last known captives detained during the July 2025 assault and massacres carried out by forces of the Transitional Administration and affiliated tribal militias.
The new strike has brought renewed attention to the issue of the detention and “abduction” of Druze residents from Sweida. Over recent months, incidents involving the detention of Sweida residents at checkpoints operated by the General Security forces of the Ministry of Interior under the Transitional Administration have increased significantly.
In most cases, these detentions have been arbitrary, ranging from the abduction of civilians along the Damascus–Sweida highway to the disappearance of travellers after their arrival at Damascus International Airport.
This article is based on three in-depth interviews with former captives from Sweida who were held at different stages of detention following the July 2025 assault, from the initial gathering points and temporary detention centres to the Military Security Branch in Daraa and Adra Central Prison.
Their testimonies reveal the emergence of systems of detention, negotiation, and release operating outside the judicial framework, in which detainees from Sweida are treated as a group, subject to a distinct security and political file, governed by a logic of control, exchange, and negotiation.
The July Assault and the Emergence of the Prisoner File
The July 2025 assault on Sweida was accompanied by a widespread wave of killings, abductions, enforced disappearances, and mass arrests that took place alongside the military operations carried out by forces of the Transitional Administration and allied tribal militias between 13 and 20 July.
Within a matter of days, hundreds of residents of the governorate disappeared into a complex network of checkpoints, temporary detention centres, security headquarters, and prisons. For months, many families remained unaware of their relatives’ fate, whereabouts, or the authority responsible for holding them.
The detentions began within the context of an open war zone marked by chaos, the presence of multiple armed actors, and overlapping authorities, operating outside any formal security or judicial process. Most of those who became “prisoners” were not arrested as a result of investigations or arrest warrants. Rather, they were detained while attempting to move between areas, flee the violence, evacuate their families, or simply while passing through checkpoints and control points.
In this context, Druze identity itself became grounds for suspicion and detention, regardless of whether the individual had actually participated in the fighting.
During the first hours and days following the incursion, the prisoners were transferred between various groups affiliated with the Ministry of Defence, General Security, and the Military Police, as well as tribal militias. This multiplicity of actors created a persistent state of uncertainty regarding which authority ultimately held decision-making power over the prisoners’ fate.
At this stage, the first temporary detention centres emerged in petrol stations, schools, halls, and civilian buildings that were rapidly converted into collection points for prisoners. Within these spaces, the initial mechanisms for managing the prisoner file began to take shape.
Some armed groups treated detainees as spoils of war or as targets for direct retaliation, while other actors began registering the names of groups of prisoners and holding them for potential use in future exchange and negotiation processes.
The testimonies of former prisoners reveal that the idea of prisoner exchanges emerged from the very first days of the incursion, before detainees were transferred to the main prisons.
In one case, a Military Police unit kept a record of several Druze prisoners who had been gathered at a petrol station west of Sweida and refused to hand them over to other groups that were demanding their execution. The unit argued that the prisoners would later be used in exchanges for captives affiliated with the Transitional Administration who were being held by armed Druze factions.
After several days of being moved between temporary detention centres, the majority of male prisoners were concentrated in the Military Security Branch in the city of Daraa. There, the first phase of reorganising the prisoner file in a more structured manner began.
Names were recorded, photographs were taken, and preliminary interrogations were conducted. Detainees were also assigned to shared cells and solitary confinement, while remaining completely cut off from the outside world. Yet this transfer did not transform the prisoners into detainees held under clearly defined judicial cases. Instead, they remained a single detention group linked to the Sweida battle itself.
The interrogations conducted at the Military Security Branch did not focus on specific criminal allegations. Rather, they centred on detainees’ personal and sectarian backgrounds, their relationship to the National Guard or armed groups in Sweida, and their participation in the fighting during July. The interrogations also continued to be conducted on ordinary paperwork, without formal case files or clear referrals to the judiciary.
During this phase, the first instances of selective release also began to take place, driven by mediation efforts, personal connections, and local affiliations. Meanwhile, the majority of prisoners remained within a detention system shaped by ongoing negotiations, exchange arrangements, and political pressure.
The brief phone calls that some prisoners were allowed to make also played a crucial role in confirming that they were being held in facilities run by the Transitional Administration. These calls came after weeks of complete uncertainty, during which reports of mass graves and missing persons in Sweida had spread widely.
With the subsequent transfer of prisoners to Adra Central Prison near Damascus, the prisoner file had effectively moved beyond the framework of temporary, battle-related detention. It had evolved into a distinct security and political file, managed by the Transitional Administration through its own mechanisms of detention, negotiation, and release.
Adra Prison and the Management of Prisoners Outside the Judicial Framework
After spending approximately a month in detention at the Military Security Branch in Daraa, the prisoners from Sweida were transferred to Adra Central Prison near Damascus.
When they arrived at Adra on 18 August 2025, the number of prisoners from Sweida stood at 114, including minors. The total number of Druze detainees in the prison later rose to 120 after six additional prisoners were added, most of them from Sahnaya.
This transfer marked a fundamental shift in the nature of their detention. The prisoners moved from temporary detention spaces exposed to the violence of the battlefield to a central prison with a more stable administrative and institutional structure. Yet the detention remained governed by the same logic that had emerged during the first days of the July assault.
The prisoners entered Adra not as detainees referred to the judiciary under specific legal cases, but as a group collectively associated with the battle of Sweida.
Inside Adra, the prisoners were subjected to a tightly controlled system of isolation and daily management. They were distributed among overcrowded dormitories, small cells, and solitary confinement units, and were prevented from maintaining normal contact with other inmates or with the outside world.
Communication with their families remained limited and closely monitored, while information itself became a key instrument of pressure. Prisoners learned about their fate through rumours, promises made by guards, visits from committees, or fragmentary news about prisoner exchanges that reached them intermittently.
At this stage, time itself became part of the machinery of detention. Promises of imminent release were repeatedly made, and on more than one occasion prisoners were instructed to pack their belongings and prepare to leave, only for the process to be abruptly halted and everyone sent back to their cells.
This cycle of promise and postponement created a prolonged state of waiting, tying the prisoners’ daily lives to the fate of negotiations between the Transitional Administration, dominated by Hay’at Tahrir al-Sham, and the National Guard in Sweida.
The prisoners from Sweida were never brought before a court, nor were specific individual charges filed against them. The interrogations conducted inside Adra remained an extension of the security investigations that had begun in Daraa, focusing on participation in the fighting, ties to the National Guard, views on Sheikh Hikmat al-Hijri, and affiliation with Sweida, a space perceived as opposed to the authority of the new government in Damascus.
In this sense, the purpose of the interrogations was not to build criminal cases against individual detainees, but rather to classify the prisoners within a collective security and political file.
Committees and official visits played a central role in managing this ambiguity. Members of the fact-finding committee established by the Transitional Administration following the events of July 2025 visited the prisoners inside Adra and repeatedly promised their release, yet these visits never evolved into an actual judicial process.
Representatives of the International Committee of the Red Cross also visited the prison. Their visits helped reveal the presence of Druze detainees being held in solitary confinement, whose existence was unknown even to the other prisoners. However, the influence of these bodies remained limited in the face of the authority exercised by the security services and prison administration over the prisoners’ day-to-day fate.
The same ambiguity was reflected in the official terminology used to describe the prisoners. In statements and announcements issued by the Transitional Administration, the term “abducted” was consistently avoided and replaced with varying labels such as “detainees,” “those in custody,” or “arrestees.”
Similarly, Mustafa Bakour referred to those held in Adra as being under “temporary detention” or “preventive custody.”
Hunger Strikes
In response to this situation, hunger strikes became one of the few means available to the prisoners inside Adra to exert pressure on the prison administration and the security bodies overseeing their case. The prison witnessed four major hunger strikes between the autumn of 2025 and the beginning of 2026, all of which followed discussions among prisoners in the dormitories regarding their timing, effectiveness, and the prisoners’ ability to sustain them.
The strikes were driven by the length of detention, uncertainty over their fate, repeated promises of release, and the absence of any judicial process that either the prisoners or their families could pursue. During the initial strikes, the prison administration attempted to contain the protests through repeated assurances that release was imminent and by summoning prisoner representatives for negotiations.
Over time, however, the administration began combining conciliatory measures with direct pressure. Some prisoners were summoned to the office of the duty officer and prison director, known as Sheikh Abu Bayhas, and were informed more explicitly that their case was a security matter linked to prisoner exchange arrangements with the National Guard in Sweida.
The fourth hunger strike, which began on 26 December 2025 under the name “Strike Until Freedom,” marked a particularly sensitive turning point after news of it emerged from the prison and spread across social media pages run by activists from Sweida. At that point, the prison administration could no longer keep the strike confined within the walls of the dormitory.
The disciplinary unit entered the dormitories, beat prisoners, and placed some of them in isolation before a new round of negotiations and promises of release began.
In this sense, the hunger strikes became part of the management of the prisoner file inside Adra. They did not dismantle the detention system, but they forced the prison administration and the security bodies overseeing the prisoners to reveal aspects of how that system operated and to open direct channels of negotiation, either with the prisoners themselves or through committees and intermediaries.
As detention dragged on, the strikes increasingly became a means of reaffirming the prisoners’ existence as a single collective group and of tying their fate to processes of exchange and political pressure rather than to individual judicial proceedings.
This process reached its culmination in the prisoner exchange carried out on 26 February 2026, which resulted in the release of the last known prisoners detained in Adra following the July assault.
The final exchange, conducted under the supervision of the International Committee of the Red Cross, involved the release of 61 Druze prisoners in exchange for the National Guard’s release of 25 members of the Transitional Administration’s forces.
The exchange itself confirmed the nature of the file: the prisoners were released through a negotiated agreement, not through a judicial process, nor as a direct result of the hunger strikes. In this way, one chapter of the July prisoners’ file came to a close, while leaving behind an established mechanism that could be reactivated in future waves of detention, abduction, and enforced disappearance.
After the Exchange: The Return of Detention and Abduction as Tools of Pressure
The prisoner exchange of 26 February 2026 brought the chapter of the known July prisoners held in Adra Prison to a close, but it did not end the underlying logic of captivity itself.
In the months that followed, the issue re-emerged in new forms: detentions at General Security checkpoints, the disappearance of travellers after arriving at Damascus International Airport, and abductions carried out by tribal groups operating outside the formal prison system.
In this sense, the file evolved from a form of mass detention directly linked to the July assault into a broader pattern of detention, abduction, and pressure directed at Sweida.
The new checkpoint detentions appear to be a direct extension of the logic of suspicion that emerged in the aftermath of July. On 7 May 2026, for example, the General Security checkpoint in al-Matouneh detained Hatem Moeen Nasr after forcing him to unlock his phone and submit its contents for inspection.
According to local testimonies, phone searches at some checkpoints have come to focus on images of Sheikh Hikmat al-Hijri, material related to the National Guard, or songs celebrating the people of Sweida during their resistance to the July assault.
In such cases, detention is not based on a specific criminal act. Rather, the phone is treated as a personal archive through which an individual’s political stance or local affiliations can be inferred.
The same pattern is evident in cases of disappearance following arrival in Damascus or contact with its security institutions. On 28 May 2026, General Security detained Talie Ghazi al-Tawil at Damascus International Airport after his arrival from Saudi Arabia, after which all communication with him ceased.
His family was initially informed that the detention was linked to unpaid instalments owed to Bank Al-Ibdaa. However, the bank later denied having issued any request for his arrest, stating that its procedures in such cases are limited to placing executive or precautionary liens on a client’s assets.
This case illustrates how suspicion has expanded from the checkpoint to the airport, and from the road to official state institutions, while the identity of the authority responsible for the detention and the detainee’s place of confinement remain unclear.
Alongside this track involving the security services, another distinct pattern emerged that must be carefully distinguished from it: tribal abductions carried out outside the formal prison system.
On 25 May 2026, the family of Raed Kamal Zein al-Din, his wife Saba Zreifeh, and their two children, Fajr and Matar, together with taxi driver Safwan al-Qudmani, were abducted while travelling from Sweida to Damascus to obtain documents related to passport applications.
The abductors did not transfer them to Adra Prison. Instead, they took them to the countryside of Homs and identified themselves as a party seeking the release of Ali Hassan al-Sameer, who had been detained by the National Guard in Sweida after being accused of participating in the July assault.
On 3 June 2026, Saba Zreifeh and her two children, Fajr and Matar, were released. Raed Zein al-Din and Safwan al-Qudmani, however, remain abducted in Homs Governorate.
This incident reveals another layer of the broader issue. The abducted family did not enter the detention system of Adra Prison or General Security. Instead, they became hostages of a tribal group linked to a wider armed milieu.
The abduction took place along a route controlled by checkpoints and patrols affiliated with the Transitional Authorities in Damascus, making the absence of any security intervention a central element in local accusations that the checkpoints either colluded with the abductors or deliberately turned a blind eye to their actions.
In this sense, after February 2026 the prisoner file assumed two parallel forms.
The first operates through General Security, checkpoints, the airport, and Adra Prison, where residents of Sweida appear as detainees or disappeared persons within an official or semi-official security apparatus.
The second takes the form of direct tribal abductions, in which civilians are used as bargaining chips outside the prison system, as in the case of the Zein al-Din family and their driver.
The distinction between these two forms is important. Yet they converge in the same outcome: the transformation of Sweida residents into instruments of pressure within an ongoing security and political relationship with the Transitional Administration.
The hunger strike announced by a new group of detainees in Adra on 27 May 2026 emerged within this context. These detainees are not among the remaining known prisoners from the July assault who were released in the 26 February exchange. Rather, they belong to a new wave of detainees and abductees taken during the preceding two months.
The new strike draws on the legacy of earlier hunger strikes as one of the only means of self-defence available to prisoners confronting unlawful detention. What continues to endure is the treatment of Sweida residents as part of a security and political file subject to bargaining, negotiation, and pressure outside the framework of the judiciary.
In this sense, the 26 February prisoner exchange did not bring the Sweida prisoner file to a close. The underlying mechanism of captivity itself remains available for reuse whenever tensions between Sweida and the Transitional Administration escalate once again.
