A / A / A
Michaël Neuman

Director of studies at Crash / Médecins sans Frontières, Michaël Neuman graduated in Contemporary History and International Relations (University Paris-I). He joined Médecins sans Frontières in 1999 and has worked both on the ground (Balkans, Sudan, Caucasus, West Africa) and in headquarters (New York, Paris as deputy director responsible for programmes). He has also carried out research on issues of immigration and geopolitics. He is co-editor of "Humanitarian negotiations Revealed, the MSF experience" (London: Hurst and Co, 2011). He is also the co-editor of "Saving lives and staying alive. Humanitarian Security in the Age of Risk Management" (London: Hurst and Co, 2016).

Fabrice Weissman

Graduated from the Institut d'Etudes Politiques de Paris, Fabrice Weissman joined MSF in 1995. He spent several years as logistician and head of mission in Sub-Saharian Africa (Sudan, Eritrea, Ethiopia, Liberia, Sierra Leone, Guinea, etc.), Kosovo, Sri Lanka and more recently Syria. He has published several articles and books on humanitarian action, including "In the Shadow of Just Wars. Violence, Politics and Humanitarian Action" (ed., London, Hurst & Co., 2004), "Humanitarian Negotiations Revealed. The MSF Experience" (ed., Oxford University Press, 2011) and "Saving Lives and Staying Alive. Humanitarian Security in the Age of Risk Management" (ed., London, Hurst & Co, 2016).

The Duty of a Head of Mission

Interview with Delphine Chedorge, MSF emergency coordinator in Central African Republic, by Michaël Neuman
This article is based on interviews conducted between May and July 2015. Translated from French by Karen Tucker.

Map of Central African Republic


Landlocked Central African Republic (CAR) has a population of 4 million people and a wholly inadequate healthcare infrastructure. In terms of funding, the country ranks first for MSF-France and third for the MSF movement as a whole, after the Democratic Republic of Congo and South Sudan.“International activity report 2014”, Paris: MSF.
Also one of the most dangerous countries, four MSF employees have been killed in its conflicts since 2007. In 2014, 300 international and over 2,500 Central African staff members worked on some twenty medical projects.

MSF emergency coordinator Delphine Chedorge headed the French section’s operations in CAR from January to December 2014. She spoke with Michael Neuman about the everyday life of a head of mission in charge of team security. The interview is preceded by a summary of the recent events leading up to the bloodshed in CAR.

The 2013-2014 Crisis in the CAR

The CAR has experienced a cycle of violence unprecedented in its postcolonial history. In March 2013, an alliance of armed opposition movements, the Seleka, seized power and installed Michel Djotodia as president. During the months that followed, the new regime’s violent attacks against the population and the previous government’s forces led to the formation of local militias. The result of an alliance between village self-defence militias and members of the former national army, these so-called “anti-Balaka” groups echoed the population’s mounting anger against a government increasingly perceived as “foreign” and “Muslim”.See, in particular, International Federation for Human Rights, “Central African Republic: They must all leave or die”, Press Release, June 2014.

Amid growing tensions and the fear of sectarian massacres, on 5 December 2013 the United Nations Security Council voted to dispatch an International Support Mission (known by the French acronym MISCA) to Central African Republic to re-establish state authority and protect civilians. MISCA was placed under the supervision of the African Union, with backing from French military operation “Sangaris”. The same day, a large-scale offensive by anti-Balaka militias against Bangui failed to bring down the regime and resulted in the flight of ex-SelekaWe will use the term “ex-Seleka” to describe the Seleka forces who are still active after they were disbanded by a decision of President Michel Djotodia in September 2013 following an increase in violent attacks and abuses.
rebels and the stepping-up of the international military deployment. After the anti-Balaka and some of the civilian population looted and committed massacres against Muslims living in Bangui, who had been left unprotected, the routed ex-Seleka launched an onslaught of unrestrained violence.See, in particular, Human Rights Watch report, “‘They came to kill us’: Escalating atrocities in the Central African Republic”, 19 December 2013.

With his back against the wall, President Djotodia gave into international pressure and resigned on 10 January 2014. However, the appointment of a transitional government did not restore the stability that had been hoped for. The ex-Seleka continued their bloody retreat toward the countries on CAR’s north, east and west borders with anti-Balaka groups in hot pursuit. Meanwhile, the anti-Balaka encouraged and led massacres against Muslims forced to flee to neighbouring countries or the few enclaves within CAR protected by international forces. A retrospective mortality survey conducted in April 2014 by MSF among CAR refugees in Sido in Chad revealed that 8 per cent of those who had fled died between November 2013 and April 2014, with 91 per cent of these deaths attributed to violent acts committed during the campaign of persecution against Muslim minorities.

At the time of the attempted coup d’etat of 5 December 2013, MSF’s French section was running three projects in the country—primary and secondary healthcare programmes in Paoua subprefecture in the north-west, where it had been working since 2006, and paediatric services in Carnot and Bria. In December, the section set up an emergency operation to care for the victims of the conflict, with particular focus on Bangui. The Spanish and Dutch sections were also working in CAR and were joined in early 2014 by the Swiss and Belgian sections.

Michael Neuman: What was the situation when you first arrived in CAR?

Delphine Chedorge: My first assignment in CAR dates back to the summer of 2007. I’ve been back several times since, for three months in 2012, then for one year in early 2014. I started off as emergency coordinator and in April I became head of mission. The current conflict began in December 2012 and intensified after the Seleka took power in March 2013, which led to the collapse of the country’s security forces. So in a sense, I “missed out” on just over a year of the conflict’s progress. During the first few weeks I had a hard time getting a handle on the security situation. What’s more, my knowledge of the country was very localised as it was centred on the north-west, where most of MSF-France’s programmes have traditionally been concentrated. I took some time to get to grips with what was going elsewhere in the country. And, personally, I wasn’t expecting such a violent sectarian conflict to break out.

What were the biggest security issues in Bangui when you arrived?

When I landed in Bangui in January 2014, there was lots of shooting in the city, including near the Hopital Communautaire, where we were treating the wounded, and our living quarters and offices. Everyone was in the same neighbourhood, right in the middle of an urban war.

It was also very difficult to get to the Muslim enclaves and neighbourhoods. In January and February we made several attempts to fetch casualties from so-called PK12 district where groups of Muslims wanting to flee had assembled and were under constant attack from particularly unpredictable militiamen. PK12 was also notable for its close proximity to an ex-Seleka camp. International forces were stationed there to protect civilians and the ex-Seleka, which created a really tense atmosphere. Sometimes we met with so much hostility we had to turn back.

National and international staff alike were highly exposed to danger. They often had to negotiate with armed, aggressive individuals entering the hospital to look for a particular patient or demanding to be treated ahead of the others. They witnessed first-hand the ferocity of the violence and its consequences in the numbers of casualties and types of wounds requiring treatment. Because of fighting nearby, the hospital team repeatedly had to seek shelter in the bulletproof operating theatre—which the in-patient tents weren’t. Staff members were experiencing high levels of physical and mental fatigue. Nobody was hurt, but the risk was high. No one asked to leave, which would’ve been perfectly understandable. We brought in psychiatrists and psychologists to debrief the teams.

What steps did you take to reduce the personnel’s exposure to danger?

When I first arrived, the head office security focal point was already there helping the teams to protect themselves, for example, from stray bullets entering houses, which had happened several times since December. He set up safe rooms that the staff used when the fighting was close-by. The hospital team were also afraid of staying at the hospital, especially overnight. So we decided that they would only work there from 8am to 4pm. We had to assume responsibility for patients having less access to medical treatment.

We sometimes designed our activities in a way we thought would increase our security while trying to establish more trust with armed groups and civilians (it was sometimes hard to distinguish between the two). For example, while treating victims of the violence that occurred in the Fatima neighbourhood in May and June and which caused fifteen to twenty casualties among the displaced population, we also ran mobile clinics in neighbouring Christian districts. Obviously, these clinics did serve a useful medical purpose—such as treating infantile malaria—but the primary motivation was to avoid being accused of working only with Muslims, even if our health centre in mainly Muslim PK5 district did, of course, provide care for Christians as well.

We also worked hard on getting information. The primary sources were the national staff, most of whom I had known for a long time. They described to me what was happening in the various districts, the groups, their weapons, what they were saying, the rumours and threats they were spreading. They also helped me identify which streets were dangerous. I wasn’t familiar with Bangui’s roads because MSF had no programmes there. We had to make a micro-political analysis of the dynamics in each of the city’s districts. We drove all over the city observing the situation. We used a driver who said he felt comfortable driving around certain neighbourhoods and he provided a running commentary on what was going on.

We worked quite well with the other MSF sections present in Bangui. One had established a working relationship with ex-Seleka while another had more recent ties with anti-Balaka groups due to their work in the M’Poko displaced persons camp.Following violent attacks in December 2013, tens of thousands of displaced persons originally from Bangui assembled at the city’s airport, Bangui-M’Poko.
During the early days of my assignment, we were fairly dependent on these contacts. We worked on the basis of trust—let’s say, not blind, but informed trust. It was hard for me, but it made sense. There were only so many contacts I could handle. Beginning in April, I regained control of contacts in cooperation with my colleagues from the other sections.

Could you also rely on outside information, from journalists and other NGOs working in the country?

Our information mainly came from three networks: missionaries, Ministry of Health medical staff and national Red Cross staff. All were very active in protecting civilians or providing relief when the fighting was at its most intense. We were also in regular contact with old acquaintances—all kinds of political officials, former rebels and district leaders. In addition, the person who had the job of head of mission when I arrived had developed her own network of officials working for some of the other NGOs operating in CAR and we were also in contact with several UN agencies and some of their staff.

At the beginning, there were few organisations delivering aid and travelling around the city and the rest of the country. The United Nations and the French army, followed by the European force EUFOR and later INSO (an NGO specifically responsible for security), gradually set up systems to provide information—usually incomplete and unreliable—to humanitarian organisations. The organisations responsible for keeping others from harm—which is more especially the military’s job—would say “avoid going here or there” or “take an armed escort”. There was some value in giving their advice serious consideration, but it was also important to maintain our decision-making autonomy. In the final analysis, and this was instructive, it was less the information itself (sometimes no more than rumours with no attempt at objectivity) than what it taught us about how much we could trust those providing the information and what they were willing or not willing to share.

Concerning the national staff, were there any specific security issues or requests?

Most of the Central African staff in Bangui lived in neighbourhoods severely affected by the conflict and they were very afraid of moving around the city. In December 2013, many of them stopped coming to the office. Coordination had set up a shuttle system to pick them up. This system closed down in early February 2014 as there was much less fighting in the city and taxis were running again. Despite this, employees regularly stayed overnight in our offices and houses because they couldn’t get home. As of September, being identified as an MSF employee no longer afforded protection and actually posed more of a risk, as employed people have money. Because of the gang culture invading Bangui, their security was jeopardised far more than ours. And, even more dramatically, all our Muslim employees had left the city, and most of them probably the country. We still don’t know what’s happened to many of them.

What were the main security problems outside Bangui?

Until October, we were occasionally able to travel by road in the interior of the country, despite a number of incidents. Of course, NGO and UN employees were sometimes targeted, but it was more for the equipment. In January 2014, a group of ex-Seleka stole a car from us. They stopped us, explained they needed a car for a day or two, took away our radio and MSF stickers and unloaded the car. We got it back after putting pressure on their commanders. The car had been used in combat. The same thing happened when an anti-Balaka group “confiscated” our truck and its crew before returning it a few days later. The truck was also used in combat. This type of “respect”, albeit relative, gradually disappeared over the course of the year. The risk was greatest on the roads, with anti-Balaka roadblocks manned by drunk, drugged-up fighters with no real chain of command. We had to limit travel by road and hire an extra plane to relieve staff and supply our programmes. This seemed to be the only option that would allow us to work in security conditions that we deemed acceptable. This decision was the subject of regular and exhausting discussions with head office who felt that the plane cost too much.

During this period, MSF-France was working in three locations, in Paoua in the north-west, Carnot in the west and Bria in the east. How did the security situation in these three areas evolve over time?

We were expecting Bria and Paoua to be the most vulnerable because they had been affected by the conflicts of the early 2000s. But Carnot ended up suffering the worst violence. There were numerous clashes between civilians, anti- Balaka, ex-Seleka and then Cameroonian MISCA forces, who were acting as a buffer between, on the one hand, the anti-Balaka militias and Carnot’s inhabitants, and on the other, Muslims trapped in the enclave who had barricaded themselves inside the church. The team witnessed massacres of Muslims on several occasions, especially in January, when we had to call on Cameroonian MISCA forces based a few hours away by road to the north to intervene to prevent Muslims being driven out of their homes and killed.

Displaced Muslims attempting to get medical treatment were at great risk and many refused to go to hospitals because of the extreme danger involved in getting to them. However, the team were able to negotiate with the anti-Balaka militias and some inhabitants a safe passage for the MSF ambulance transporting wounded Muslims and MISCA soldiers so they could be evacuated by plane to Bangui.

After clashes between international forces and anti-Balaka, in July 2014 a Fula patient was lynched inside Carnot hospital. This was one of the most serious incidents that had ever occurred in one of MSF-France’sprogrammes. You then began a “mobilisation campaign"—first local, then national—calling for the protection of healthcare facilities. What were you hoping to achieve with public statements about a security incident?

We probably should have done it sooner because we realised that some of the health workers were in fact not surprised that sectarian groups were settling accounts inside a health facility. Our message was: the hospital provides care for everyone and we cannot tolerate any violence; otherwise, we’ll have to leave. The team went to ask all the local health and political authorities, armed groups, local people and neighbourhood leaders to get all their contacts to spread the message that this was not normal. We succeeded in getting the message through.

Then, when we talked with the team and the other sections of MSF, we realised that what had happened in Carnot could happen anywhere. That’s why we decided on a national campaign, which included other MSF sites. We used posters and radio broadcasts to call for the protection of our medical activities.

Wasn’t it somewhat futile to call for the protection of health facilities on the basis of humanitarian principles?

It doesn’t hurt to use these magic words, so long as they’re followed up with a conversation and much more tangible negotiation. When an incident occurs, we try to determine the cause of the problem and our role in it. We also try to figure out how we can continue operating and providing relief in the particular environment. In the case of Carnot, it was in everyone’s interests that we stay. But our communications were not limited to the campaign. The local press in CAR published all public statements made in response to security incidents. Despite the lack of public reaction from politicians, some of our contacts did call us—even if only to see how we were coping. By speaking out, we could also counter government rhetoric about the supposed “normalisation” of the situation, which Central African and international officials (with France taking the lead) began claiming in late 2014. In view of the increasing number of robberies at homes and offices belonging to other MSF sections operating in the country, attacks on vehicles on roads, extortion and the seizure of our transporters’ trucks, it was important to make a point.

Central African Republic is where MSF-France’s last international volunteer was killed. It was in June 2007and the victim was Elsa Serfass, a logistician with the Paouaprogramme. You first worked in CAR in the days and weeks following this tragic event. Did this incident affect the way you managed security during your most recent assignment?

My greatest and most constant fear was losing a member of our team. I used to bring up Elsa’s death in my briefings with volunteers. Telling the story provided an opportunity to remind them about all the weapons that were circulating and the state of chaos in the country. This was important because, even in 2014 during a time of extreme violence, as soon as the situation calmed down for a few days, some team members could be quick to forget that we were working in a dangerous country. You also have to be honest with people coming to a programme and give them specific examples, such as murder, rape and lynching of patients.

I believe it is unacceptable to hide serious incidents from people arriving in the field. Even I found myself without information about a number of serious incidents—including sexual assaults—against colleagues in other sections. This led to some tense conversations. Managers sometimes tend to withhold information because they want to protect the dignity of the victims, but this information is vital to assessing the shifting nature of the risks teams are exposed to.

And, there’s a risk of the violence becoming trivialised. People immersed in a dangerous environment where incidents are commonplace can become inured to danger and no longer react to it because they end up seeing exposure to violence as the norm.

What were the circumstances that led you to suspend operations or evacuate staff during your assignment?

In 2014, when we thought there might be a significant deterioration in the situation, we carried out several preventive evacuations to reduce exposure to danger. For example, during the violence that took place in Bangui in October, we decided to evacuate twenty-four people by road and boat to three neighbouring countries over three days. And then there was the attack against Boguila hospital in April, which left nineteen people dead, including three of MSF-Holland’s national staff. We had lots of discussions with the heads of mission of the five sections working in CAR about how we should respond. There were two opposing opinions. The first advocated closing all projects in the country for a set period of time in the slim hope that such an extreme decision would provoke a reaction from the armed groups. The second, more moderate, opinion, mainly supported by MSF-Holland’s head of mission, called for evacuating just the international personnel and staff relocatedThis refers to Central African staff who are mainly hired in Bangui and then relocated by MSF to programmes in other parts of the country.
from Boguila. In the end, we chose the more minimal option of limiting care to emergency cases across all programmes for one week. An exception was made for Boguila, where international and relocated employees were withdrawn for a longer period and replaced with spasmodic visits. We found out who was responsible for the killings—a leader of an ex-Seleka group. But we didn’t speak out and release this information to the public. Instead, we complained to his superiors and waited to see what they would do. But to no avail as he’s still at large.

MSF’s operations in Paoua ended up being suspended for the longest period, even though the area had been the least affected by the war. How do you explain that?

In August, our Central African employees began making a series of demands, which they backed up with a strike. They wanted salary increases and transport allowances. We didn’t agree to these demands so they decided to call a day of strike, while maintaining minimum service. During the strike, which took place in September, picket lines were set up and some employees who wanted to continue working were seriously threatened. The local authorities who had agreed to act as mediators were accused of being traitors, which raised the question of whether we could continue to operate. The team was finally evacuated in December after international staff began receiving death threats. They made a gradual return, but only at the very end of the year.

MSF has been working in Paoua since 2006. How would you explain this deterioration?

The first factor has to do with a context specific to Central African Republic, the deteriorating labour relations resulting from the many years of violence in the region and the absence of State representatives and local government mediators—and all this against the backdrop of an economic crisis. Other organisations also had to deal with very difficult labour conflicts. The second factor is internal to MSF. During the year, five people had successively held the position of project coordinator in Paoua and this lack of continuity definitely impacted our ability to make a clear assessment of the worsening situation, particularly regarding labour issues. What’s more, we were taken up with the other programmes because we felt their teams were at much greater risk so, for sure, the coordination team didn’t monitor the situation closely enough.

In general, how much autonomy do project coordinators have to assess and handle security?

It depends on the person and how our relationship develops. Not everyone has the same amount of experience or the same ability to analyse the situation they find themselves in. For example, when I consider the explanations and precautions aren’t sufficiently convincing to justify a journey, I can refuse to give my authorisation. When you feel that your team leader has a handle on all this, you can give more autonomy.

We partially delegated the security of one team to another organisation: Catholic missionaries, as it happens. This is a very rare occurrence at MSF nowadays. For a few days in late January 2014, we left a small two-person team—an anaesthetist and a surgeon—in Bossemptele to the north-west of Bangui with no car or means of communication. This was at the time the ex-Seleka were taking flight and the anti-Balaka were carrying out violent reprisals against the town’s Muslims, causing many casualties. Wounds were becoming infected because the Central African doctor at the missionary hospital was out of the necessary supplies. So we decided to send two people to give them a hand.

This was a really specific situation. The Catholic mission was actively defending and helping Muslims in the area, the priest was accustomed to interacting with all of the armed groups, and there were missionary nuns there too. The mission compound was relatively well protected. I left the team there without a car. At the time, having a well-maintained MSF car would have attracted the militias’ attention, so not having one was actually safer. The team was almost invisible, but nevertheless, all the political and military groups were aware they were there; we didn’t act surreptitiously.

In Central African Republic as in other places, MSF has in recent years decided to ban some volunteers from working in certain programmes on the basis of their nationality and the colour of their skin. How did we get to this?

We decided to do this in two cases. In April 2014, a logistician was attacked in Bria for being white and French. French Sangaris forces present in the area were seen as taking sides against the Muslims and we were at risk of being associated with them. The first step we took was to pull out the volunteer. Then we decided to stop assigning white people there at all, since they might have been thought to be French. We soon realised that this was an isolated incident; the perpetrator had been upset and angry because his son has been killed in the fighting. And, in fact, lots of people had jumped in and defended the logistician.

Nevertheless, there could have been more cases like that, so, after talking with the team, for several months we kept to our decision. Considering our overall operational volume and the number of international staff in the country, we were simply making our jobs easier. But we didn’t prevent visits by the coordination team working out of Bangui. These increased, to such an extent that the restriction lost its meaning. We definitely could have brought white Western staff back in faster.

Then came the issue of staff with Muslim backgrounds. To avoid any problems, we adopted a super-pragmatic position because we thought that North Africans would be seen as white by the anti-Balaka. As for the Africans, some of them changed their first names to less Muslim-sounding ones. This was left up to each individual. On the other hand, I myself refused to appoint a Malian Tuareg as deputy head of mission because the nature of his job would have meant showing his face as he moved around Bangui, which would have posed too much of a risk.

Among the issues relating to security in Central Africa, the extent of MSF’s exposure was a major concern. Many people felt there were too many staff on the ground: 300 international employees, eighty of them in the French section alone, and 2,500 national employees in all MSF sections. What was your position on this issue?

You have to remember that Bangui is the most dangerous city in CAR and it’s where the coordination team is located. It’s also the city with the largest team. If you add the employees working at the hospital and our health centre in PK5 to the coordination team, there can sometimes be over forty-five international employees.

What’s more, head office’s decision to reduce operations in order to limit our exposure to danger contradicted their policy of deploying “first assignments” [or “first missions”, to use NGO-speak] to the field. In an environment that was unstable at best, and frankly dangerous most of the time, jobs were created to meet the need to train new volunteers rather than immediate operational requirements. It was completely contradictory and done without my agreement. In Paoua, for example, this was the case of two volunteers out of a total of eight and, after the violence that occurred in Bangui in October, I had to get them out via Chad in dangerous conditions.

At the beginning of our interview, you mentioned the role of the security focal point—an innovation for the French section instituted in 2013. The focal point’s appointment coincided with the operation department’s introduction of systematised “security management tools“, such as the risk assessment matrix and the system used to record and file information on security incidents known as “SINDY”. What do you think of these measures?

The logbook for recording incidents occurring in the areas where we operate, the guides, briefings and emergency public statements issued after incidents— all these were nothing new. The security focal point helped us make the team aware of the security environment and contributed to briefings, particularly with the logisticians tasked with setting up security measures (communications, safe rooms and tracking travel). This aspect was useful. Then, when he returned to head office in Paris, he insisted that we keep the SINDY database up-to-date. SINDY is a centralised system for filing reports on security incidents affecting only MSF.See Chapter 4, box: “Security Incident Narratives Buried in Numbers: The MSF Example”, p. 67.

 This we disagreed about because I didn’t feel it was of direct benefit to the field. We already had logbooks and incident registers for recording important events to be taken into account in analysing the security environment. I’m sure it’s useful for MSF as an institution to keep a database of reports of the most serious incidents but, given that we were already very busy, I didn’t think it was necessary to do head office’s secretarial work. What is important is working with the team on managing incidents and sharing the information with the other sections. And the other danger with using SINDY in the field is that people will only see the problem from the MSF perspective and overlook incidents affecting other agencies.

To get back to the risk assessment, isn’t there something scary about making an exhaustive list of the threats you might be facing?

Yes, I ask myself that question. But in my experience, when I use the risk assessment during briefings I’ve noticed that the people I’m talking with become calmer and more focused as the conversation goes on. Their eyes are opened and they become more aware of their environment. In the end, after these conversations, people feel prepared and confident because they know the situation has been well thought out.

The idea is for people to be on their guard. There needs to be a balance between trivialising and exaggerating the risk.

At MSF and other organisations, there’s a certain amount of protest against the growing number of security rules in the field. One of your head of mission colleagues who spent a few weeks in Bangui said that “the curfew rules treat volunteers like children and encourage them to flout the rules’’.

Of course this happens; it’s a natural consequence of rules. But the volunteers didn’t seem particularly reluctant to comply with them. When they did flout them, it was in a way that didn’t put them at too much risk. That’s what we ask of people: when you break a rule, make sure you know why and how. If we need to, we’ll talk it through again and, if necessary, we’ll change it.

For example, when you ban your teams from going to the flea market in Bangui, is it because you worry there might be be serious problems or is it because you might have to deal with the theft of a careless volunteer’s bag?

What actually happened was that ordinary petty thieves got themselves grenades and were more and more violent. Also, you don’t manage forty international staff the same way you manage ten; you can’t talk to each one, look into every issue, etc. We definitely wouldn’t have had the same rules if there’d been five or ten of us in Bangui, but we were forty to fifty. It also explains something that was occasionally contested. When the security situation in town settled down enough to allow volunteers to go out, different curfew hours for weekdays (9pm) and weekends (10pm) were introduced. I never thought the city was less dangerous at the weekend. On the other hand, from a personal point of view, a simple question of fatigue, I couldn’t allow myself to be on call every evening of the week to deal with a car stopped at a police checkpoint on the way back from a restaurant. I was okay with being available an hour later at the weekend in case of problems so they could have more freedom. It’s a shame the teams couldn’t manage small incidents like these without outside help but it wasn’t always the case. I was making life easier with these rules. They were geared more toward management of human resources than security management.



BOX: The Case of "Dangerous Patients" in Yemen's Governorate of Amran

Map of Yemen

“I would not want to be a doctor here.”

Interview with the Amran project coordinator.

Already in 2010, well before Yemen became engulfed in all-out war between Houthi rebels and factions supported by Saudi Arabia in 2015, national and international staff working in MSF projects in Amran GovernorateLocated in the north of the country, the governorate has a population of around 1 million inhabitants. viewed the situation as highly dangerous.

Khamer, where MSF has been in charge since 2011 of all the hospital’s departments with the exception of the Ministry of Health-run outpatient Department (OPD), had been a peaceful town where international personnel were free to walk around—except at night, because of stray dogs. However, during the period from 17 April 2010 to 15 June 2013, MSF project coordinators in Khamer and in the nearby town of Huth, recorded twenty-three security incidents, none of which involved the death or kidnapping of any MSF employees. Verbal threats were a daily occurrence and being threatened at gunpoint was commonplace, as were shootings in the hospital compound and car-jackings. International employees were not usually affected, while Yemeni medical personnel working in the emergency room (ER) were more exposed than staff in the inpatient department (IPD). The most serious incident had been a revenge killing in 2011 that resulted in a patient’s death at the hospital.

Such incidents led several Yemeni doctors to leave the projects. In 2012 alone, one surgeon left after being verbally threatened by the relative of a patient he had operated on, a doctor after being forced at gunpoint to treat a patient, and a third after he was slapped. A local doctor interviewed in 2013 commented: “There is a 20 per cent chance I get killed in the hospital, 80 per cent chance I stay safe.”

This situation prompted the programme manager for Yemen to request an investigation into the logics of violence and the reactions to this violence of MSF and Ministry of Health staff. Conducted in July 2013, the investigation was based on interviews with patients, personnel and local authorities, mission archives and a review of pertinent social science literature on Yemen. Its main findings are presented below.

“Some Patients are Dangerous, We Know It”Interview with a member of hospital management, Khamer.Referring Patients for Security Reasons

In interviews, most Yemeni and international doctors tended to blame the insecurity on the lack of education of patients and their families and an “archaic tribal system living off the lack of strict regulation of government allowing any member of a tribe to do whatever he wants.”Interview with a hospital director, Sanaa.
People from villages outside Khamer—the primary target population of the project—were perceived to be the main troublemakers. The doctors recognised that this perception influenced their medical practices, as one of them explained:

"When patients come from communities with whom we’ve had problems, it bloats, and then, the therapeutic decision has no longer any medical and scientific rationality. It is quite common to hear comments such as: ‘this one is from this family’, ‘he is the son of that one’, ‘he comes from this region’, etc. It has a significant impact."Interview with an international doctor, Khamer.

In fact, it was common for patients with alleged “dangerous profiles” to be referred to other medical facilities in Amran or Sanaa, even if their medical conditions did not warrant it. There was very little disagreement among Yemeni and international staff that, “if there is a security risk, it is better to refer.”Interview with a Yemeni doctor, MSF, Khamer.
In some cases, the decision was at the discretion of the night supervisor, a non-medical staff member who “knows everything and everybody.”Interview with a Yemeni doctor, MSF, Khamer.

“Promises Are Not Followed by Acts”17. Interview with a Yemeni doctor, former MSF employee, Sanaa.Dealing With the Sheikhs

When a serious incident occurred, MSF frequently reacted by seeking the mediation of local tribal authoritiesThe concepts of “tribe”, “subtribe” and “family” in the context of Yemen are the subject of academic debate. As Paul Dresch puts it in “Tribalisme et démocratie au Yémen” (Arabian Humanities, no. 2, 1994), tribes are “evidently, not [...] very solid group[s]”. Although central to the understanding of social and political dynamics, the tribe is a malleable concept: “a very important flexibility in term of conflicts and alliances potentially exists.” and sometimes suspended its activities to put pressure on them and their tribes. In most cases, after a period of suspension varying from one day to six months, mediation was successfully concluded, compensation paid—money, cows, or guns—and victims apologised to by culprits in a gathering of local leaders. This reactive approach to insecurity was criticised by some staff members for its ineffectiveness. Given that doctors could expect little protection from the various local institutions, staff demanded that MSF play a more assertive role in ensuring safety. A Yemeni doctor formerly employed by MSF commented:

"The only thing we’ve been doing lately is incident, apology ceremony, incident, apology ceremony, incident, etc. We have to think about it in a different way."

The international team seemed to believe that the sheikhs were all powerful, if only the right one could be identified. As a member of the international team said, “He can do whatever he wants with his people”. However, some academics question this assertion, echoing the view of many Yemeni staff. As political scientist Laurent Bonnefoy explains, it is unreasonable to expect the sheikhs to prevent violence from occurring. Controlling violence in North Yemen is based first and foremost on “mitigation” and “regulation”, rather than on “prevention”, in an effort to ensure that conflicts do not get blown out of proportion and stay contained within acceptable limits.Interview, June 2013. See also Nadwa Al Dawsari, “Tribal governance and Stability in Yemen”, Carnegie Endowment for International Peace, 2012.

“Doctors are Parasites which Live on Human Blood.”“Yemeni doctors cause more harm than good”, National Yemen, 18 July 2012.

MSF’s reaction to the violence appeared to overlook as a source of tension the poor relationship between doctors and patients. Generally speaking, Yemeni doctors appeared to suffer from a very poor image, as illustrated by an article published in National Yemen in July 2012 entitled “Yemeni doctors cause more harm than good”:

"Thousands of Yemenis fall victim to medical errors at the hands of doctors, whose unearned and undeserved titles and certificates are the only things which connect them with the practice of medicine. (...) Many Yemenis have expressed their dissatisfaction with Yemeni doctors, who they say are not good at their jobs and have transformed their sacred profession into a way to earn money. Many have gone so far as to liken doctors to “parasites” which live on human blood."Ibid.

Some aspects of the operational set-up appear to have further exacerbated the general distrust. Lack of clarity in ER admission criteria was often mentioned as a factor of tension by both medical staffand patients. The ER saw around half of the total number of patients who arrived in triage, between 1,500 and 2,500 a month, while the other half were referred to the Ministry of Health’s OPD, run by three doctors from the former Soviet Union, where services were not provided free of charge.

Many patients refused to be referred to the OPD and exerted pressure on medical staff to be treated by MSF. As one interviewee explained, “the more vocal the patients are, the better chances they get to be seen by the MSF doctor.” Many people viewed this medically unjustified discrimination as the source of most of the problems encountered by the hospital’s employees, and this was without taking into account, as a Yemeni MSF doctor explained, “our watchmen, our staff, nurses, nurses’ assistants, they are taking their friends, their relatives to treat them. Sometimes we, the doctors, refuse, and sometimes, we don’t.”

Some patients did not understand why MSF delivered mostly emergency services and not, for example, care for chronic diseases and non-urgent surgery,The maternity, paediatric and adult medical departments admit non-emergency patients; patients with leishmaniosis and rickets are also provided with non-emergency treatment.
nor why they would be or needed to be referred to other facilities where they would have to pay. Routine profiling of patients by doctors according to where they came from and their family and tribal affiliations added to the tension. What’s the point of having a hospital if it cannot be accessed?

The layout of the hospital also contributed to tension in and around the maternity ward.

"Part of the problem is that there is no waiting room in the maternity—the building is too small. So the families generally wait outside while the women are in labour. Sometimes it can last for hours, during which the family are left in the dark, uninformed about how things are going if the midwife in charge does not take the time to come out and talk to the families."Interview with an international midwife, Amran.

Brief analysis of the incidents MSF staff encountered revealed their extreme diversity, as much in origin as in manifestation. Ultimately, the issues confronted by MSF in Amran were framed, for the most part, within a demand wholly comparable to that experienced by MSF and health professionals in hospitals all over the world: a quality relationship between patients and health personnel. At the hospital in Khamer, MSF operated in a setting where this expectation may have conflicted with the reality on the ground, given that the high level of violence in the region appeared to be generally socially accepted and that intimidation is integral to social regulation there. The investigation revealed that, while humanitarian organisations do not have to see themselves as passive victims, neither do they have to view Yemeni patients as inherently dangerous.



Security Issues and Practices in an MSF Mission in the Land of Jihad

Judith Soussan

Map of Syria

Translated from French by Philippa Bowe Smith.

On 11 May 2013, the coordinator of the Qabassin project sent an email to the coordination team in Turkey announcing the opening of the MSF hospital that very morning. He summed up with a downbeat “so far so good.” He was certainly being modest, for there was much to be proud of. This was the first hospital set up by an NGO and with international staff to be located deep inside Syrian opposition-held territory rather than along the Turkish border like MSF’s other projects: the hospitals in Atmah (MSF-France), a few kilometres as the crow flies from the coordination team’s base in Reyhanli in Turkey, Bab al-Salama (MSF-Spain), Bernas (MSF-Belgium) and Tal Abyad (MSF-Holland). It had taken six weeks of extensive refurbishment to turn an empty building into a clean and well-equipped facility with surgery and maternity units, an emergency room and a twenty-five-bed inpatient department.

However, the opening of the Qabassin hospital, a small but significant event, did not get the attention it deserved. A bomb exploded the same day in Reyhanli, killing fifty-one people and injuring over 150, while the previous day armed men in Atmah verbally attacked and threatened to kill a Swedish MSF staff member they accused of spying. He then had to appear before a local Islamic court. In comparison, Qabassin seemed almost a haven of peace and the members of the team, all newly arrived and enjoying a quiet life during their time off, walking around the town, visiting the market and receiving invitations to take tea, had trouble taking the coordination team’s calls for vigilance seriously. “They were too relaxed, they’d forgotten where they were,” recalls the head of mission at the time.

This chapter tells the story of the Qabassin mission from the security perspective. It examines the perceptions and practices of the team in the field (starting with the field coordinators, who came and went at a rapid pace) and the coordination team in Turkey, with whom they were in permanent contact.To avoid any confusion arising from successive changes in staff, we will indicate in pp. [111-116] the footnotes the period during which a particular project coordinator or head of mission was present.
How did these people analyse their situation, the prevailing risks and events as they unfolded? What behaviours did they adopt in the face of danger—from the rules and procedures (introduced, modified or forgotten) to the various strategies designed to “reduce exposure” and “improve acceptance” (to use current parlance)? This chapter will pay particular attention to moments when disagreements arose, during which people’s often complex definitions of the word “security” were revealed and came into conflict.To explore these questions, we consulted all the literature produced by the Qabassin mission: field reports (“sitreps”), security documents (“security guidelines”) and, especially, daily emails exchanged with the coordination team. Interviews with over twenty people involved on the ground, in the coordination team and at the head office in Paris were also conducted between January and June 2015. We thank them for their time and their help. This study does not include any views from the team’s Syrian personnel, fascinating as it would have been to have heard their opinions on the situations they faced, the decisions taken and various people’s attitudes. For reasons of feasibility, this was always going to be the case.

Finding the Right Role in the War (Mid-2011 to Early 2013)

Exploration (How to Find Protection from Bombs?)

With the organisation’s first attempts to take action in Syria in mid-2011, MSF-France’s approach toward its operational positioning took the form of a common dilemma—how to operate within the context of the Syrian civil war and reach out as much as possible to its victims, without exposing teams to excessive risks or compromising quality of care.

It took the emergency desk team a lot of patience, several false starts, exploratory missions that failed to lead anywhere because the risks were considered too high, and finally a few pivotal encounters for MSF-France to open its first project in Atmah in June 2012. MSF was, as it’s now called, “embedded”, as the hospital and house where the international staff were accommodated and their security were all provided by a highly influential Atmah public figure, who was also a doctor and a member of one of the local Free Syrian Army (FSA) brigades. The many months of patient strategising and exploration in an environment where foreigners were strongly suspected of spying resulted in the watchwords: keep a low profile.

The Atmah project now open, MSF started to look for ways to reach out to areas more directly impacted by the conflict. Those behind the Syria programme had their eyes firmly fixed on Aleppo, a city split into a government zone and a constantly bombed FSA zone. But after an exploratory mission in August 2012, operations managers in Paris deemed it too dangerous and rejected the idea of sending in international staff, especially in light of the loyalist force’s targeting of field hospitals.In the end, MSF-Spain opened a project in the “industrial district” on the outskirts ofAleppo in mid-2013.
Two further attempts to deploy operations failed. In October 2012, a hospital project set up in partnership with some Syrian doctors in Kafr Ghan near the Turkish border had to close three weeks after opening due to fundamental differences between MSF and the Syrians on how to run the hospital. Then MSF turned to Al-Bab, a city with 130,000 inhabitants situated some thirty kilometres from Aleppo along the route used to evacuate casualties. The project was well advanced when, in January 2013, the town was repeatedly bombed and the coordination team evacuated the project team to Turkey. In the words of the head of mission: “I told them, ‘we’re in the same situation as the Syrians, we’re not protected. So we need to find a safer place.’”Interview with the head of mission (January-June 2013), 17 June 2015.

The database he had compiled showed that, out of all the places in a ten-kilometre radius around Al-Bab, the town of Qabassin had never been shelled. The team had also heard that its population of 20,000 Arabs and Kurds included a not insignificant proportion of people who backed the regime. And, unlike Atmah, where black-clad foreign Islamists paraded in their pickups, there was no visible armed presence. This was why, so it was said, Qabassin was untouched by the turmoil among opposition groups and bombing by pro-regime forces. On 27 January, the day after they were evacuated, two members of the small team decided to go to the town.

Opening the Project (How to Gain Acceptance ?)

They carried on the work of identifying and meeting with key contacts initiated during the two months spent in Al-Bab. In addition to members of prominent families, they had met with representatives of nascent revolutionary institutions—the civilian-run local council, whose responsibilities included health, and the Islamic court, which handled justice and policing. The project coordinator had also established contact with representatives from local politico-military groups, including brigades affiliated with the FSA and Islamist groups such as Al-Qaeda affiliates Ahrar al-Sham and Jabhat al-Nusra. While maintaining these contacts in Al-Bab, they became acquainted with the small world of Qabassin. They met with influential local leaders, some of them also members of the local town council (there was no Islamic court as yet), as well as representatives of the Kurdish party affiliated to the Kurdistan Workers’ Party (PKK).Bravo [Qabassin] sitrep, week 7-8, 15 to 28 February 2013. “Tea. I drank a lot of tea during this period,” recalls the project coordinator with a smile.

The team driving the Qabassin project wanted to do things differently to Atmah (not to rely on a protector) and to the failed Kafr Ghan project (to avoid co-management). The hospital was to be “100 per cent MSF.” In terms of activities, however, the project was similar to the others, with surgery the focus to provide treatment to casualties evacuated from Aleppo via Al-Bab. The decision was also taken to offer general medical treatment and surgery to local people, a decision that, according to documents written at the time, met two essentially tactical objectives. First, to avoid appearing as a hospital for fighters, thus minimising the risk of being targeted for bombardment by the regime. Second, to gain better “acceptance by the community”Bravo sitrep, week 3-4, January 2013.
by allaying any possible concerns about the risks stemming from setting up a hospital in the town and offering services that people would probably need. The other major feature of the project would be to identify small medical facilities closer to the combat zone (including in Aleppo, which the organisation was not giving up on) that MSF could support as the situation evolved on the ground. These outreach activities would also enable MSF to channel sick and wounded to the hospital in Qabassin while monitoring needs engendered by local politico-military developments.

As preparations were underway, the team wrote an in-depth analysis of the risks and detailed the security rules. According to the Security Guidelines approved in March 2013, the main risks in Qabassin related to road traffic and the “psychological impact”, i.e. stress.Security Guidelines—annex 3—Bravo risk analysis, 26 March 2013.
In addition to these “high” probability dangers, bombing and crossfire were classified as “medium”, and chemical weapons as “low to medium”. The risk of kidnap—despite the abduction in Atmah on 13 March of two members of NGO ACTED—was described as low, “as we are well-known and well-accepted within the community.”Security Guidelines—annex 2—Bravo rules, 26 March 2013.
The security rules included standard procedures, for example relating to travel: in a car, wearing a seatbelt, carrying identity documents, leaving the Syrian driver to handle interactions with people manning checkpoints, and on foot, writing down destinations on the board used to keep track of where staff are going, and not walking alone or at night. But, a number of very strict and detailed rules governing personal behaviour were also laid down:

"1. Behaving appropriately towards staff/local people is the very first security rule. Do not shout or address people aggressively (...) No flirting/sexual relations between international/national [staff]. 2. Laughing out loud can give the impression of being drunk. (...) 4. NO PHYSICAL CONTACT WHATSOEVER between men and women (no hand-shaking, etc.). (...) Do not take ANY photos as people may think we are spies/journalists.Bravo Security Memo, summary of the rules accompanying the Security Guidelines, 25 March 2013.

Women must not smoke in public (...) Alcohol, illicit drugs and marijuana must not be taken or even discussed with Syrians. Do not discuss politics or religion. Dress appropriately outside the hospital at all times (men: no shorts, women: cover the head, arms, and legs no tight-fitting clothes)."Security Guidelines, annex 2.

In short, these rules formalised the need to keep a low profile, something everybody agreed on, which would lead to “acceptance”, as indicated in the words concluding the rules: “Better acceptance by the community = improved security”. It is worth noting that this notion of acceptance, introduced to MSF via security manuals,See Chapter 5, p. 71.
was used extensively by members of the mission to describe a very wide range of practices and behaviours: no smoking in public, providing maternity care and meeting with local authorities.

In early March, just as work on refurbishing the hospital was about to begin, an incident occurred that made it seem that a project would have to be abandoned yet again. A delegation of Qabassin residents informed the project coordinator that they opposed the opening of a hospital because they feared that the town, so far unscathed, would become a target of bombardment by the regime. Remembering the failure of the prematurely opened Kafr Ghan project, the project coordinator took the time to understand and make sure of the project’s backers and, during the period 9 to 16 March, he held numerous meetings. It transpired that the complaints were motivated as much by genuine fear as by frustrations stemming from the unequal distribution of the benefits prominent families stood to gain from MSF’s project. So, after reassuring them, the refurbishment began and the recruitment process was initiated. The incident made the team more determined than ever to be as objective and transparent as possible, and they interviewed 300 candidates, from surgeons to cleaners. At the same time, the team took care to achieve a balance between Arabs and Kurds and between the various families for all posts not requiring any specific skills.

A Peaceful Little Town?

Networking I. (Establishing a Network of Contacts—How and Why)

After the hospital opened on 11 May 2013, the medical team concentrated on setting up activities. Although Qabassin remained calm, bombing and TNT barrel-bombs dropped from regime helicopters were daily occurrences across the Aleppo Governorate and there were sporadic rumours about the use of chemical weapons elsewhere in the country. It was this specific threat that led MSF’s newly appointed (and very first) security focal point to make a visit to Syria in May 2013.

This was the background to the increasingly troubled relationship between the head of mission in Turkey and the new project coordinator, who had arrived in mid-April. The head of mission complained of “poor visibility” about the situation, which he blamed on inadequate communication on the part of the project coordinator.Interviews with the head of mission (January-June 2013), 17 June 2015, and deputy head of mission (June 2013), 6 February 2015.
He therefore asked him, starting with the tools and procedures, to show that he was paying sufficient attention to security matters. A morning and evening “security contact” was set up between the coordination team in Turkey and the project coordinator and initial hiccups further exacerbated tensions.Email exchanges between the logistics coordinator and the project coordinator on 4 June, 6 June, 9 June 2013, etc. This security contact had not been put in place beforehand: “What for? We talked to each other all the time anyway,” said the project coordinator who set up the project (December 2012-April 2013).
The head of mission accused the project coordinator of failing to use monitoring tools (the incidents database he had created and the board used to keep track of each team member’s movements). Above all, he asked him to provide more detail about the context and to improve his network building. This included getting to know the people controlling checkpoints and obtaining their phone numbers, maintaining links with various groups and meeting newcomers, for, as well as Jabhat al-Nusra (the Al-Qaeda affiliate already present on the outskirts of Qabassin), other Salafist groups started setting up offices in June 2013. “You must communicate, build your network,” the head of mission told him. In the Syrian context, where spy-fever was rife, and in the absence of anything specific to discuss, the project coordinator was firmly of the view that it was best to steer well clear:

“I felt it was necessary to stop asking people too many questions. (...) I’m convinced it was the right thing to do. Sometimes I feel it’s inappropriate, the way we turn up and start questioning people (...).

And then there are the things that I know [being Muslim] about people who are a bit conservative, with radical tendencies. You’re a guest; the fewer questions you ask, the greater your chances of being accepted.”Interview with the project coordinator (April-July 2013), 17 June 2015.

The same goes for the checkpoints: “Our laissez-passers worked everywhere; if they were letting us through, why ask questions?” He felt it more appropriate to observe and to “secure the relationship with our close entourage” of three or four regular contacts who had reassured him that “if the team toed the line, no one would harm us.” Far from being settled by the change of head of mission in late June, the disagreement merely intensified, as his replacement was particularly focused on documenting the political and military context.

Disquiet (How to Interpret Information?)

While rumours of a major “battle for Aleppo” had been circulating since late May, June and July were in fact characterised by tensions and incidents within the opposition forces. These included clashes between PKK-affiliated Kurdish forces from Qabassin and police units acting under the authority of the Al-Bab Islamic court, the town falling under the provisional control of these Kurdish forces, and a bomb explosion at the Jabhat al-Nusra base just outside Qabassin. There were also confrontations at checkpoints between FSA brigades and fighters from a group newly arrived on the scene, known to be an offshoot of Al-Qaeda in Iraq and not on good terms with Jabhat al-Nusra: ISIS/ISIL or the Islamic State in Iraq and al-Sham/the Levant.We will use the more common acronym: ISIS (the group took its present name, Islamic State (IS), in June 2014).

At the end ofJuly, the project coordinator reported back on these troubles, noting that, although Qabassin was currently “really quiet”, Islamist groups (which included ISIS and Jabhat al-Nusra) further north in Jarabulus had announced their intention “of establishing an Islamic State” and had “declared that foreign NGOs are infidels and so not welcome in Syria.” He added: “We’ve got all these groups in Qabassin.”Bravo sitrep week 26-27, 25 July 2013.

His mission ended on 30 July and, one week later, his replacement arrived (the project’s third, including the set-up stage). From the start, he was alarmed by the situation he discovered, which contrasted strongly with the impression he had been given during his briefing back in Paris. As well as the heightened tensions between the various opposition groups, there had been two major incidents. The car bringing the project administrator back from the Turkish border to Qabassin after his leave had been held up by a group of armed men on the outskirts of the town. After apparently hesitating to kidnap the administrator, they had eventually left, taking with them the end-of-month wages he had been transporting. And, in Aleppo, an MSF-Spain car had been stopped by an armed group, and its occupants (a Syrian MSF logistician and two non-MSF passengers, said to be a Turkish contractor and his American girlfriend) were still being held prisoner.

By mid-August, the now very detailed emails the field team sent to the coordination team reported one concern after another. These included the team’s lack of preparedness for a possible chemical attack, renewed ISIS statements attacking foreign NGOs in Jarabulus and rumours of a group targeting British citizens for kidnap, etc.Email from the head of mission to the project coordinator, 13 August 2013.
The project coordinator informed the coordination team that he wished to cut back the international staff immediately. The head of mission acquiesced, but not with the same urgency. 15 and 16 August saw clashes between Kurdish forces and the FSA/Islamist groups at various points across northern Syria. On 17 August, fighting broke out in Qabassin and that evening the project coordinator wrote to the head of mission: “everybody is safe at the house (...) ISIS now controls the town.”Email from the project coordinator to the head of mission, 17 August 2013.

An MSF Mission in “The Clutches of Islamic State of Iraq and al-Sham”

The first contact with ISIS took place at the MSF hospital when several combatants showed up after the fighting on 17 August. Two were injured and a third came complaining of stomach pains: “When we asked him to remove his coat, he said it was explosive and he couldn’t take it off.”End of mission report, project coordinator (August-September 2013).

The fighting had lasted just a day and calm was restored on 18 August. At the MSF house, the project coordinator and medical advisor interviewed each international staff member in turn. Were they willing to stay in an environment where “Al-Qaeda, in this case one of its affiliates, ISIS, is now fully in control of Qabassin?”Ibid.
Nine of the fourteen team members chose to leave the mission and, by 19 August, they were all in Turkey.Late that night, having interviewed all the team members, they applied the same process to themselves and interviewed each other in order to avoid prejudging their respective decisions (discussions with the project coordinator, March 2015).
The arrival of ISIS had enabled the project coordinator to get what he wanted and cut back the team in quite drastic fashion.

The five international staff who remained in Qabassin did not rule out following the others to Turkey. “What are we waiting for? For them to start executing our patients against the wall behind the hospital?” asked the nurse.Reported by the project coordinator, interview, 23 June 2015.
Predictions went back and forth between the coordination team and the team in the field. The project coordinator reported that a Syrian staff member said: “They want to establish an Islamic State [in Qabassin].” “That doesn’t necessarily mean there’s no place for MSF,” replied the head of mission, “it depends on the degree of tolerance of the group as a whole as well as on their commander here.” To which the project coordinator replied: “Can you give me one example of a place where they have sole power, where they have proclaimed the State, and where people like us are tolerated for very long?”Exchange ofemails between the project coordinator and head ofmission, 19 August 2013. Based on his twenty-five years’ experience with MSF (including eight years as president), the project coordinator was not optimistic. His research backed up his instinct: “I went online to look into their behaviour in recent years. In terms of civilian casualties, they’re quite simply the most murderous of Al-Qaeda’s branches,” he wrote a few days later.Email from the project coordinator to the head of mission, 25 August 2013. As well as using the Internet, the project coordinator asked the medical adviser to show him how to use Twitter so that he could track statements and rumours circulating on that platform.

Yet, on 19 August, the new ISIS representative in Qabassin paid the project coordinator a visit, assuring him that MSF could stay and work in safety. He agreed to put this commitment in writing:

"In the name of Allah, the beneficent, the merciful,
Thanks be to Allah, peace and blessings be upon his prophet Mohammad.

From now on, the MSF hospital can continue treating all cases without bias to any party. MSF will consult the Islamic State in Qabassin if any problems arise at the hospital. And the Islamic State takes responsibility for protecting the hospital in the event of any danger. All doctors, men and women, can carry on working at the hospital."ISIS letter, cited in email from the project coordinator to the head of mission, 20 August 2013.

Networking II. (What Does “To Drink Tea” Signify?)

MSF also received unsolicited written support from important local contacts such as the local council and the Al-Bab Islamic court, whose various members were wary of ISIS’s takeover of Qabassin. Right from the beginning of his mission, the project coordinator had met with as many people as he could and had even established cordial relations with some of them, such as a military judge at the Al-Bab court with links to the Ahrar al-Sham Islamist group, and a prominent citizen of Qabassin and member of the Free Syrian Army. Several would occasionally turn up unannounced at the house to talk. “I didn’t really make any new contacts, but maybe I treated them differently,” said the project coordinator. So when, for example, the local council or the Islamic court approached him to provide material assistance to displaced people, “I said ‘ok, let’s go and see what we can do’, and that’s how we came to spend a fair amount of time together.” The project coordinator saw this as more about running the project than managing security. But time spent with local contacts did help to get a better grasp of the situation—the needs and the political currents—which he viewed as “the first stage in security.” He and other members of the team would spend evenings and Friday afternoons in the company of these contacts and, in these more informal settings, “we didn’t pretend to be neutral.” They would chat and listen to the Syrians’ accounts of “the early days of the revolution.”

"According to Syrian social norms, when you pay somebody a visit you score a point. (...) If you answer an invitation you are honouring your host. It establishes a connection, a situation where it’s ok for you to ask for something, or to be asked; it’s a two-way thing—are you prepared to trade? If the answer’s yes, then you need to establish relationships, while being aware of the social norms. (...) Sure, people ask you for favours, but then you ask for lots too! It worked both ways, and it’s absolutely vital that it does."Interview, 23 June 2015.

This is in very stark contrast to his predecessor:

“I don’t go to eat in people’s homes. You need to be careful about accepting an invitation. Sometimes it’s a two-way thing, and that’s not a trap I want to fall into. So, I thank them for the invitation, no more than that, and we respect each other.

You need to keep your distance; I can drink tea with them in the office, chat at the hospital or in the street without having to visit their homes. And as we didn’t know how the situation would play out. There’s a saying in Arabic: “remember that we have shared salt at my home.” (...) It related to my role as project coordinator. I wanted to maintain that distance that would allow me, if one day the need arose, to say: “I owe you nothing, and you owe me nothing.”Interview with the project coordinator (April-June 2013), 17 June 2015.

So, behind the consensual phrase “to drink tea” were two diametrically opposed practices.

Disagreements about the Situation Analysis

Despite reassurances obtained from local contacts and the Islamic State of Iraq and al-Sham, the project coordinator was still uneasy about the project’s future. He felt that the fact that the coordination team and head office seemed fairly reassured simply meant they were failing to comprehend the extent of the danger. This was the case regarding the management of several human resource issues. Replacing the administrator was the subject of heated debate. As the only Muslim and Arab-speaker among the international staff, he was known in the small world of Qabassin and thus had a vital role in maintaining and gathering information. The coordination team wanted him to end his mission at the end of August as planned, to which the project coordinator answered that he was willing to stay on for longer and that his departure would seriously endanger the mission. Questions about the number and profiles of international staff led to yet more disagreement. While numbers had been increased to nine in order to resume surgical activities, Paris wanted to send in more international staff as well as visitors from head office. And lastly, an American doctor who arrived at the end of August let it slip that he was Jewish. He was asked to maintain absolute discretion about his origins; already in early August, even before ISIS arrived, the project coordinator had been forced to replace a Sri Lankan doctor who had created upset among staff and patients in the emergency room when he revealed that he was a Buddhist.“For some groups in the area, being a Buddhist is really as bad as being a Yezidi (...) it can turn nasty.” Interview with the project coordinator, 23 June 2015.

Amid these preoccupations, on 21 August came the news that the regime had used chemical weapons in Damascus suburb Ghouta, a devastating manifestation of the threat that had been on people’s minds for months. The team on the ground asked the coordination team to send in drugs to treat a possible flood of contaminated patients as well as to respond to requests made by very worried local council health officials. There was outrage when they were told that the drugs they already had—enough to treat a few dozen patients— would suffice for now. They also wanted to prepare for the admittedly quite unlikely possibility of chemical attack against Qabassin or Al-Bab. The protective gear (suit, mask, gloves, boots) was a real organisational headache as it was hard to use, bulky and expensive. So the organisation had made its choice: the team in Qabassin had a little more protective equipment than there were international members of staff. Several team members found this unspoken hierarchy in terms of protecting lives hard to swallow; what were they supposed to do about the national staff, their families and the patients?“I worried about loads of dying people pounding on our door,” remembers the physiotherapist based in Atmah, where a similar situation arose (interview, 11 February 2015).

Adjustments (How to Justify MSF’s Presence in Light of the Risks ?)

“ISIS have asked us to continue to work with an international team in Qabassin. This is both dangerous and of little use,”Document attached to an email from the project coordinator to the head of mission, 23 August 2013.
was the project coordinator’s assessment in relation to what the project was delivering. He felt that the number of births was low (averaging ten per week) and that the hospital saw very few casualties, as most were treated at field hospitals set up by the various military groups (ISIS also had its own facility at a location unknown to MSF). The majority of procedures in the surgical unit were for burns (averaging twenty-four new cases per week), some very serious and which the project coordinator felt would be better treated in Turkey than in the MSF hospital. He therefore suggested “pulling out the international team from ISIS’s clutches before things go really wrong”, relocating the project to Al-Bab (where the opposition forces were more balanced, as reflected in the make-up of the Islamic court) and transferring responsibility for Qabassin to national staff.Email from the project coordinator to the head of mission, 25 August 2013. Two weeks later, heavy fighting broke out in Al-Bab between several political groups including ISIS (which was in the process of establishing a major presence), upsetting the balance of power and thus undermining the appeal of this option. “I didn’t agree!” recalls MSF’s president. As head of the emergency desk team when the Syria mission was being set up, he was still in contact with military commanders he had encountered in Atmah, one a Chechen jihadist who was a member of the Muhajireen (or ‘exiles’), which went on to become a key element of ISIS. The jihadist reiterated that MSF was not in danger in Qabassin.Interview with the president (formerly head of the emergency desk), 20 May 2015.
As for the head of mission, he argued that what had been achieved in Qabassin was not so bad, but authorised the project coordinator to further explore suggestions for adapting activities.

In the meantime, the team set out to redress the balance between the activities and the risks, by further developing activities. The doctor in charge of outreach, who had arrived at the beginning of August, was encouraged by the project coordinator to assess areas where displaced people had settled—they had already been visited by his predecessor but no action had been taken— such as As-Safirah to the south of Qabassin. In this constantly bombed area receiving no assistance whatsoever, small operations were set up to distribute medical supplies, tents and basic items, giving the team the sense of being “where we’re needed.”Interview with the outreach doctor, 28 January 2015; informal discussion with the project coordinator, March 2015.

Procedures for travelling to As-Safirah consisted of making contact with Syrian contacts on site (a doctor and displaced persons manager) the day before, then getting an update on the actual day, notably via Twitter. If the decision was taken to travel, the MSF doctor met up with the two Syrians just outside the bombarded zone. “Stay close to the guy with the walkie-talkie. He gets real-time military information,” the project coordinator instructed him.Informal discussion, 20 January 2015. “I’ve never thought of security as a geographical concept,” he says. “It’s all about finding the right network.”The MSF doctor exposed himself to risks that he says he was “ultimately” not ready to take. He confided: “I set aside time every Friday to ask myself whether it was worth being there” and every week he answered in the affirmative. Paradoxically, by increasing the volume of activities and running operations that were riskier—but that the team felt to be more relevant—identified dangers became more acceptable.

The team’s cohesion during that time undoubtedly contributed to the willingness to take risks: the project coordinator reported daily to his colleagues on the content of his meetings, sharing his interpretations and doubts. According to the medical advisor, this approach differed from that adopted by the previous project coordinator, for whom “nobody had the right to get involved in security issues because that was the project coordinator’s role.” “It felt a bit like he didn’t want to talk about security because he didn’t want to worry people: ‘let me worry about security, and you worry about medical stuff.’” However—and this comes up time and again in interviews—it was by becoming better informed about the significant danger around them that the team found themselves more “at ease”. The risks were appraised and carefully considered, and they were able to witness the efforts that went into minimising them as much as it was possible. Presumably, they also appreciated knowing what they were exposing themselves to so that they could decide for themselves whether the risks were worth taking.

Another factor was that the team’s day-to-day routine had not been impacted by the town falling to ISIS. The fighters were keeping a low profile, leaving Qabassin’s local council and Islamic court to manage day-to-day affairs. Apart from the destruction of a Sufi saint’s tomb, ISIS had not taken any hostile action or introduced any radical measures contrary to local practices. Female international staff already wore long dresses and headscarves, so did not have to alter their clothing. By the end of August, the project coordinator, who had banned going out on foot immediately after the fall of Qabassin, felt it was time to put a stop to the “bunker syndrome” and encouraged international staff to get out and interact with people.Bravo sitrep, August 2013.
Mindful of his previous experience with Islamist groups, he nonetheless took the view that the “honeymoon” between ISIS and MSF would be short-lived. After the group’s initial attempt at winning over local people, which would include respect for MSF, a “deterioration”“MSF in the land of Al-Qaeda”, document attached to an email from the project coordinator to the head of mission, 23 August 2013.
in the relationship would follow. But how could the exact moment when it would become too dangerous be anticipated?

From One Guarantee to the Next: Incidents, Negotiations and Breakdown

Red Lines (What Constitutes a Serious Incident?)

On 2 September, a Syrian surgeon from MSF-Spain’s Bab al-Salama project was kidnapped in the middle of the night from the house occupied by Syrian staff. He was tortured and murdered. No group claimed responsibility, but for MSF-Spain there was no doubting that the act was in response to the doctor’s openly atheist views.He had even been the subject of a fatwa issued by an FSA affiliate and MSF-Spain had sent him to Turkey; two months later, wanting to return and claiming the fatwa had been lifted, subject to his good conduct, he returned to Syria, but apparently continued posting his atheist conceptions on Facebook.
The Spanish section concluded that this was a personal affair and did not target the association “as such.”Incident report, MSF-Spain, 7 September 2013.
The kidnap-murder caused significant tension between the various MSF sections. For once in agreement, MSF staff in Qabassin, Reyhanli and Paris were angered that, as in Aleppo a few weeks before when its car had been stopped and its passengers abducted, the Spanish section seemed reluctant to share its information—or at least its suspicions about who was responsible. Furthermore, the Bab al-Salama team had been evacuated without informing MSF’s other projects, indicating in the view of the Qabassin project coordinator “a total lack of consideration for teams on the ground who need information if they are to stay safe.”Email from the project coordinator to the head of mission, 7 September 2013.

On 19 September, ISIS captured Azaz, bringing MSF-Spain’s Bab al-Salama project under its control.This was the fourth MSF project to come under ISIS control, along with Qabassin (MSF-France, Aleppo governorate), Tal Abyad (MSF-Holland, Raqqah governorate) and Bernas (MSF-Belgium, Idlib governorate, which fell under ISIS control on 26 August 2013).
Then, on 25 September, teams on the ground learned that the previous seven days had seen much chatter on social media in response to a tweet on 18 September by an ISIS member showing a photo of an international MSF worker and stating: “#ISIS publishes a map of bases used by missionary spy doctors working for MSF.” The MSF members who had passed on the information advised against responding. “It’s easy to decide not to respond when you’re not the one at risk,” commented the exasperated project coordinator, who proposed to counter-campaign on social networks by publishing letters of support from ISIS to MSF.Email from the project coordinator to the head of mission, 25 September 2013.
His mission ended a few days later.

After a gap of ten days, a new project coordinator arrived on 12 October for just one month. The next day, a serious incident occurred in the hospital’s surgical unit when several casualties were brought to the hospital after an altercation with men from the Al-Bab Islamic court who had come to arrest them. Three of the patients were taken away to be judged, two of whom died a few days later from lack of proper treatment. The third was executed by shooting and, on 18 October, the men from the Islamic court brought his body back to the MSF hospital.

Then began a dialogue of the deaf between the project coordinator and the coordination team on the significance of the incident and how MSF should react. As the project coordinator saw it, the men from the Islamic court, who had not been aggressive, “simply did what they had to do to fulfil their objective.” “None of us are comfortable with it, but we’re aware that we couldn’t have done much about it,” he added. At least “the worst was avoided, meaning shooting inside or around the hospital.” The next step was to meet “the Al-Bab and Qabassin courts to talk about what would happen if suspects were taken to the hospital in the future” and “regularly remind all the groups encountered of the principle of no-weapons in the hospital.”Incident report, 13 October 2013.
The head of missionThat is, the head of mission from mid-September to mid-November 2013 (present since the end of August as deputy). pp. [122-129] felt that the project coordinator’s reaction was unsatisfactory and was worried about his attempts to calm things down.Email from the project coordinator to the head of mission, 21 October 2013, in which he describes the team’s stress levels about the situation as excessive, and emails from the head of mission to the project coordinator, 22 October.
On 22 October, the operations department in Paris got involved, insisting that inflicting physical harm on patients should be considered as a “red line” that called for MSF to express its indignation and ask the authorities to renew their guarantees. The project coordinator was asked to go back to see his contacts and “lay the cards on the table.”Interview with the head of mission (mid-September-mid-November), 25 February 2015.

At the same time, the rest of the team appeared somewhat unconcerned. The now many international staff members (back up to around fifteen), almost all newly arrived, were kept in the dark about the discussions between the project coordinator and the coordination team. Present since August, the outreach doctor was the only one to still feel very uneasy.Interview, 27 January 2015.
They were thus all somewhat bewildered (and the doctor relieved) when, following notification from the French Ministry of Foreign Affairs of a threat to “kidnap two MSF doctors”, the coordination team requested an immediate reduction in staff numbers (the outreach doctor was among those who left). Travel restrictions were imposed and French staff still present were advised to leave soon. It was also decided not to post any more French citizens to Qabassin, as was already the case for US citizens.

Following this sudden revelation of the dangers around them, as of 4 November, the team became gripped by anxiety. That day, three ISIS fighters turned up at MSF and requisitioned an ambulance, promising to return it in four or five days. The medical advisor recalled that it was at that moment she realised she risked being kidnapped. “I said to myself: ‘today they’re taking the vehicle; tonight they might come to the house and take me.’”Interview with the medical advisor present in October and November 2013, 4 June 2015.

(Re)negotiate, But How Far Should We Go ?

Following these incidents, the president of MSF-France travelled to Syria. He and the new head of mission (the fourth since the start of the project who arrived in early November) made a visit to Qabassin to try and meet with representatives of ISIS from Qabassin and Al-Bab. The Al-Bab representative, “a Sudanese man everybody was terrified of”, received them and provided the guarantees they sought regarding MSF’s presence and its international staff— including French citizens. The Qabassin representative agreed to appoint a civilian intermediary who would from then on convey any ISIS demands. On the journey back, the president met with his Chechen contact from Atmah and asked him to confirm the validity of these assurances, “He said: ‘stop worrying! I’m telling you that you can do your work.’”Interview with the president, 20 May 2015.

Buoyed by these reassurances, the new head of mission felt it possible to handle ISIS’ demands more calmly. Making the odd donation of medicines or small items of equipment was, in his view, the price to pay for maintaining a dialogue with the group, which had to be acknowledged as the de facto authority. This attitude was in contrast to that of his predecessor (who considered unacceptable any donations to the “armed group”)Exchange of emails between the head of mission and the project coordinator, 22 October 2013.
as well as of the new project coordinator, who was sceptical about whether it was possible to work with ISIS.Project coordinator present from December 2013 to February 2014 (the mission’s fifth). Having already worked in Qabassin in August as a logistics officer, then with the coordination team in October-November, he was one of the “pessimists”.
These differences of opinion became yet more patent after a second MSF ambulance was commandeered on December 18, as one saw a red line, and the other sought to play it down. “We don’t pack our bags over an ambulance (...); for an ambulance you renegotiate.” Three days later the vehicle was returned, fitted with new tyres, and the debate rolled on: stay, yes, but until when, just how far should we go ? The logistics coordinator summed up MSF’s implicit response to this question:

"I think only under extreme circumstances will the organisation decide to cease activities, for example, if an international employee is kidnapped or attacked. For most other incidents, it’s a question of negotiating with everyone involved to decide HOW we can continue our activities, not IF we continue our activities."52. Email from the logistics coordinator to the project coordinator, 23 December 2013.


On 2 January in Bernas, Idlib province, five international staff from the MSF-Belgium project were kidnapped by ISIS, then under great pressure from other Islamist groups and FSA brigades which went on to launch a major offensive against it the following day. On 4 January, a trusted source told the Qabassin project coordinator that the same armed groups had decided to rid the town of ISIS. He advised MSF to keep out of harm’s way for a while and, the next day, all international personnel were in Turkey.

MSF-Belgium asked all sections to maintain their activities in Syria as long as the negotiations to free the hostages lasted. The Qabassin project continued to operate with Syrian staff alone and remote supervision by a project coordinator based in Turkey. Incidents such as arrests and intimidation involving national staff increased and some left the town. The period of winning over the civilian population was well and truly over.

The hostages were freed after five months of captivity in the spring of 2014. In Qabassin, ISIS representatives asked MSF to return, arguing that they were being penalised for the behaviour of their counterparts in Bernas. But in view of the loss of trust caused by the kidnap itself, the treatment meted out to the hostages and the demands made for their release, MSF called for explanations and further guarantees from the very highest levels of the organisation. Several letters were addressed to the “inner circle”; none received an answer. On 21 August 2014, MSF officially announced the cessation of all its activities in Islamic State-controlled territory.

Pictures of the execution of American journalist James Foley had been released two days previously. Four Western hostages, three of them aid workers, would be beheaded before the end of the year. On 6 February 2015, the Islamic State announced the death of Kayla Mueller, a young aid worker who, it transpired, was the woman kidnapped in the MSF-Spain vehicle in August 2013.


It is very difficult to look at the history of the Qabassin mission without being influenced by what we now know about how events played out for MSF and, more importantly, about Islamic State. While its very name today inspires dread, this group was little-known in August 2013, and those best informed at MSF viewed it as a sort of “Al-Qaeda, Mark 2.” Until it captured Qabassin, the teams (rightly) saw clashes between armed groups, and most of all bombing by the regime, as the most serious risks to themselves. As for kidnappings, journalists had been abducted from 2012 on and the threat was viewed with varying degrees of gravity depending on the successive heads of mission—but in Qabassin in 2013 this threat was consistently perceived as remote given how well-accepted the project was. This perception did not alter until the arrival of the project coordinator in charge during August and September, which was closely followed by that of ISIS. These factors underline the difficulty of what is known in humanitarian aid jargon as “context analysis” (forming the basis for risk analysis)—rather like trying to feel the way forward in thick fog by identifying, singling out and analysing scraps of contradictory, and sometimes confused, information. This task is still more arduous when information is lost due to a rapid succession of project coordinators, usually with no handover period, not to mention the gaps during which the post is held by a replacement for a few days or even left vacant.

In this regard, analysis of the Qabassin project appears to show that practices intended to compile as much intelligence as possible in order to obtain exhaustive information—be it via a multiplication of tracking tools, incident reports and statistical data, or compiling politico-military events, press dispatches and articles—do not always help when it comes to analysing a context. While they have their uses, they also take up much time and energy because of their non-selective nature. It may also be assumed that the coordinators would have learnt of the most pertinent facts among all this information sought so vigorously, provided their network of contacts had been working effectively. The story of the Qabassin project shows time and again the importance of quality relationships, something that the expression “to drink tea” does not, as we have demonstrated, necessarily describe with its masking of opposing practices. Several members of the project team maintained such relationships with their Syrian colleagues and contacts external to the organisation, which helped them not only to gain an understanding of the local dynamics but also ensured a constant flow of information. It also enabled a sharing of crucial information that afforded the team at least a few hours’ notice of what was going to happen, such as when the team was evacuated in January 2014. Contrary to the usual recommendations, trust in these instances was not built so much on invoking “MSF principles”, adopting a neutral stance and a “standard” discourse, as on relationship-building, dialogue, careful listening, a kind of openness—necessarily combined with respect for the codes and beliefs held by others—and “dependability”: “do what we said we would do and say what we did.”Interview with the head of the emergency desk since June 2013, 28 January 2015.

Nevertheless, the end of the story also serves to highlight the limitations of the guarantees derived from some of these relationships. We have placed much emphasis on differences in analyses made by, on the one hand, those in the field, and on the other, the coordination team and head office in the immediate aftermath of ISIS’ takeover, as well as on the differing degrees of disquiet of the successive project coordinators and heads of mission (a disquiet felt far more in the field than at head office). Leaving aside the subtleties and nuances in the various points of view, these divergences related to whether assurances from ISIS representatives were to be trusted. The consistently affirmative response provided by head office to this question, in the face of the concerns expressed by those in the field, was directly linked to the assurances MSF’s president obtained through the close relationship he had established with several important contacts who had proved to be trustworthy since the opening of the Atmah project. Conversely, it was because he judged ISIS to be different to all the other armed groups operating in northern Syria that the project coordinator on site during August and September was pessimistic and gave these assurances little credibility. His pessimism was rooted in his experience with jihadi groups, in what he had heard from national staff and other contacts in Qabassin and Al-Bab, and also in his personal research into the group, its statements, writings and publications, its political programme—in short, its ideology.

But what this story also shows is that, despite everything, MSF stayed. Regular challenging of the relevance of the project, the deterioration predicted by some, Twitter accusations of spying, rumours about the kidnap threat from France’s Ministry of Foreign Affairs, the incidents in October and December, the murder of the Syrian doctor and the kidnap of MSF-Belgium’s five international staff—none of these led the team to pull out. Evacuation (supposedly temporary) was finally ordered on the basis of detailed information provided by a reliable source on an imminent outbreak of hostilities. Once the team left, senior managers realised they were “relieved.”Ibid.

 The fact that successive teams were kept in place provides much food for thought about the way in which MSF functions as an institution as well as the reasons that motivate individuals to stay in a situation and accept substantial risks. These reasons may be rooted in an attachment to a place, to their work, a refusal to abandon a civilian population, or to less easily avowed sentiments, such as the thrill felt at certain moments or not wanting to let down MSF by a failure to overcome fear.



The Shadow Theatre of Kidnappings: an account of Arjan Erkel's rescue

Duncan McLean

“In spite of the broad solidarity that individual persons and international organizations demonstrated towards Arjan Erkel, today, it must be underlined that the management of his case remains a failure:

Failure of the Russian authorities that have legal responsibility to solve the case. Failure of the Dutch Government who have maintained a low profile diplomatic approach to the case. Failure of partner governments of the Russian Federation that remained largely passive and indulgent, allowing the escalation of violence towards humanitarians in the region. And failure of MSF for trusting that all of the above were doing their utmost to resolve the case and would do so automatically without requiring mounting public pressure to be applied by MSF.”

Médecins Sans Frontières, “Arjan Erkel, Hostage in the Russian Federation since August 12, 2002: One Year of a Kidnapping”, press pack prepared for the briefing on the occasion of the one-year anniversary of Arjan Erkel’s abduction in August 2003.

Map 4: Russian Caucasus
Map 4: Russian Caucasus

Arjan Erkel, MSF-Switzerland’s head of mission in Dagestan, was kidnapped in the suburbs of regional capital Makhachkala on 12 August 2002 and freed in somewhat obscure circumstances 607 days later. Arjan’s captivity was punctuated by episodes of considerable publicity, many initiated by MSF, with some, like the example above, epitomising the frequent conflicts opposing the Erkel family, the Dutch government and MSF, despite all three parties sharing the same stated objective of securing his release.

The entire affair is unusual, both in the lengthy court case that opposed the Dutch government and MSF and in the many accounts of his abduction that have been published.Dutch journalist Coen van Zwol has written a history of the kidnapping and Erkel himself has published a memoir of his time in captivity, both of which are referred to in this text. See Coen van Zwol, Gijzelaar van de Kaukasus: De ontvoering van Arjan Erkel, Amsterdam/Rotterdam: Prometheus, 2005; and Arjan Erkel, Ontvoerd: 607 dagen tussen leven en dood (My Abduction: 607Days Between Life and Death), Amsterdam: Uitgeverij Balans, 2005.
In conjunction with interviews of the principal actors and internal reports, these documents provide a rare window of analysis into the experience and dilemmas faced by an aid organisation seeking to obtain the release of one of its employees.


MSF in the Caucasus

Right from their first interventions in the Caucasus in the early 1990s, and more particularly in Chechnya, MSF teams were confronted with a context of denial of access, intimidation and extreme violence. The First Chechen War (1994-1996) resulted in the near decimation of its populationVarious estimates put the number of Chechens dead or missing between 50,000 and 100,000 out of an initial population estimated at around 1 million people. See Thornike Gordadze, “Chechnya: Eradication of the Enemy Within”, in Fabrice Weissman (ed.), In the Shadow of “Just Wars”: Violence, Politics and Humanitarian Action, London: Hurst & Co., 2004, pp. 183-208., while the Second (1999-2009) saw the military occupation of the country by 100,000 Russian Federation troops from April 2002 onward. Occurring on a daily basis, insurgent attacks and counter-insurgency operations created a climate of terror. According to human rights organisations, at least 25,000 civilians disappeared or lost their lives between 1999 and 2007.Amnesty International report, “Russian Federation: What justice for Chechnya’s disappeared?”, May 2007.

During both Chechen Wars, MSF was outspoken in drawing attention to the tremendous human cost of these conflicts as well as the danger of delivering aid.Reports such as “The Chechen Republic, Far from Peace”, released in 1996, detailed the “systematic bombing and killing ofcivilians by the Russian military in Chechnya” while the November 2000 report “Chechnya: Politics of Terror” denounced the “massive aggressions and the policy of terror towards the civilians of Chechnya”.
This was tragically illustrated on 17 December 1996 with the murder of six delegates of the International Committee of the Red Cross (ICRC) in their field hospital in Novye Atagi. MSF was no stranger to the perils of the region as a total of sixteen of its employees were detained or kidnapped in the Caucasus between 1993 and 2004.This total includes twelve “express kidnapping” (resolved in less than 24 hours) and four extended-period abductions of which Erkel’s was the longest.

A long-standing tradition in the region, hostage-taking took on an entirely new dimension during the First Chechen War. Arbitrary arrests and detentions of Chechens by Russian and pro-Russian forces led to the practice of exchanging or selling live and dead prisoners. This use of prisoners as bargaining chips was the prelude to the wave of kidnappings that was to sweep through the North Caucasus after the first war.FIDH and Centre des droits de l’Homme de Memorial, “Rapport d’enquête sur les crimes de guerre et les crimes contre l’humanité perpétrés en Tchétchénie”, October 2000.Involving all segments of society—gangs, clans, rebels, security forces, politicians, entrepreneurs, etc.—the flourishing human trafficking industry targeted local residents, Russians and foreigners alike, as illustrated (among others) by the abduction on 19 January 1998 of the head of UNHCR’s North Ossetia office, Vincent Cochetel (freed 317 days later),Véronique Soulé, “Tchétchénie: Vincent Cochetel le miraculé. Enlevé en janvier, le Français a été libéré par les Russes samedi”, Libération, 14 December 1998.
and the kidnap of International Red Cross nurse Geraldo Cruz on 16 May 1999 in Kabardino-Balkaria (released in late July).AFP, “Red Cross removes workers from North Caucasus after NZ kidnapping”, 24 May 1999.

Prior to Arjan’s abduction, the latest MSF member to be kidnapped had been Kenny Gluck, head of mission for the Dutch section. Abducted in January 2001, he was held in Chechnya for twenty-six days before being unconditionally released. Unlike most previous cases, which had involved the Chechen mafia and networks with Russian connections, Kenny had been kidnapped by an Islamist Chechen resistance group. He was released with a “letter of apology”, signed by its leader Shamil Basayev himself. Later posted on pro-independent Chechen website Kavkaz.org the letter explained that Kenny had been kidnapped by a group of mujahideen hoping to negotiate his release in exchange for comrades and family members abducted by Russian military forces. It stated that a High Sharia Court had decided to release him unconditionally because of his humanitarian work. “We also inform you that when examining your case, the Assembly of the High Sharia Court decided to forbid the abductions of members of humanitarian agencies,” the letter concluded.See Laurence Binet, “War crimes and politics of terror in Chechnya, 1994-2004”, “MSF Speaking Out” Case Studies, September 2014, p. 161.

The benevolence shown by the Chechen opposition towards Kenny Gluck did not improve the already strained relations between Moscow and MSF, regularly accused in the pro-government press of siding with the separatists.Coen van Zwol, “Identity of the US Attachés was Unknown to Erkel”, NRC, 16 July 2003.Tensions were further heightened by disagreements over the planned tripartite agreement between Russia, Chechnya and neighbouring Russian republic Ingushetia to repatriate some 200,000 internally displaced persons (IDPs) to Chechnya. MSF voiced concerns over the forced return of civilians to a war zone at a time when Chechnya was subject to zatchiski, i.e. “clean-up” operations executed by masked men travelling around in armoured vehicles with unidentifiable plates whose job it was to torture and slaughter civilians suspected of supporting the insurgents.“In Chechnya, the intensity of violence has not diminished. On the contrary, bombardments, cleansings, rackets, death squads and tortures are the norm”, taken from “MSF Concerned Over New Pressures Exerted on Chechen Refugees in Ingushetia to Return to Chechnya”, MSF Press Release, 3 June 2002.

By mid-2002, MSF operations in Chechnya, Ingushetia and Dagestan included running mobile clinics, supporting hospitals and dispensaries (through the provision of medical supplies and rehabilitation of essential services) as well as assisting displaced populations. The fear of kidnap and targeted attacks against aid agencies had led MSF, in the spring of 2002, to significantly reduce its international staff in the Caucasus.

Unlike Chechnya, neighbouring Dagestan had not plunged into civil war. However, at the turn of the new millennium, it was the second poorest republic in the Russian Federation and the fringes of the Chechen War created an opportunity for gangsters, politicians and businessmen looking to make a profit, notably through arms and human trafficking. Shackled to Moscow, which provided 90 per cent of the Republic’s state budget, Dagestan was prey to rampant corruption, as illustrated by the nebulous role played by “police officers [working] for the state, their clan and their crime organisation simultaneously.”Van Zwol, op. cit., p. 18. See also Arkady Babchenko’s collection One Soldier’s War in Chechnya, London: Portobello Books, 2007, which also provides detailed anecdotes of the regional economic impact of the conflict.

MSF was relatively new to Dagestan. In March 2000, the Swiss section had conducted an exploratory mission with the objective of gaining access to Chechnya and setting up a small operation in Dagestan itself. Despite various warnings, particularly from the press,According to the Jamestown Foundation or the BBC for instance, “The kidnapping threat [was] worst in the north Caucasus—in Chechnya and its neighbours, Ingushetia and Dagestan”. See Nabi Abdullaev, “Foreigners beware: Kidnappers are still operating in the North Caucasus”, The Jamestown Foundation, 27 February 2001; Stephen Mulvey, “Analysis: Caucasus kidnap threat”, BBC News Online, 20 June 2002.
the section had concluded that Dagestan was, for the time being, relatively safe. Small-scale activities limited to rehabilitation and vaccination programmes were set up along with distributions of non-food items for displaced Chechens.

The Abduction

His predecessors having alternated every few months, Arjan Erkel was the Swiss section’s first permanent head of mission in Dagestan. He arrived in mid-April 2002 as the region’s humanitarian agenda was dominated by the tripartite plan to repatriate IDPs to Chechnya. Arjan attended internal MSF meetings in Paris during which strategies were outlined to lobby and advocate publicly against the forced return of displaced Chechens.

In mid-July, the Office of the United Nations’ Security Coordinator informed MSF that the FSB (Federal Security Service of the Russian Federation, ex-KGB) had given warning of an increased risk of abduction. It was assumed by the various MSF sections that the warning was a measure of intimidation to reduce the presence of NGOs and curb their campaigning against the repatriation plan. However, after Nina Davydovich, head of Russian NGO Druzhba, was abducted in Chechnya on 23 July, there was general agreement within MSF to evacuate those international personnel still present and freeze national staff travel in the Caucasus.The reduction and evacuation of international staff occurred progressively within the different sections: the French section suspended all expatriate staff travel in the Caucasus following the initial FSB warning; the Belgian section suspended travel after the kidnapping of Davydovich, as did the Dutch section, although they did maintain a presence in Nalchik, Kabardino-Balkar Republic.

The Swiss section still considered Dagestan safe and made an exception for the area, where it continued its operations. During the same period, Arjan was contacted by a US Embassy military attaché requesting security information and logistical assistance with a visit to Dagestan. On his own initiative, on 4 August, Arjan sent a car to Makhachkala airport to pick up the officer and another American attaché and later dined with them. No other MSF staff were present at the dinner. Indeed, all those consulted opposed contact with the Americans due not only to the incompatibility of their and MSF’s missions, but also to the skewed perceptions this could provoke.Internal MSF document.

It was not until a second FSB warning was received on 6 August—this time relayed directly to the Swiss section via their Khasavyurt office—that international staffwere reduced and restricted to Makhachkala. Arjan remained and pursued a number of discussions with Dagestan security agencies, who provided reassurances that there was no increased risk to MSF’s teams. Nevertheless, the analysis of his Dutch counterpart in Moscow, relayed by Arjan to the head office in Geneva, would echo previous FSB warnings: “MSF-Holland thinks it will not be quiet until a big fish is kidnapped.”Ibid.

Shortly after, on 12 August, Arjan was snatched from outside his girlfriend’s home in the suburbs of Makhachkala. Although slightly injured during the abduction, by his own account he was subsequently treated relatively well. He was held for a week by his kidnappers and then passed on to “keepers” operating on behalf of unseen “owners”. Over the course of the next twenty months, Arjan would be detained at various locations. He was never permitted to see the faces of his captors but nonetheless built a rapport of sorts with them. Bits and pieces of information would filter through to him, partly reflecting possible avenues for his liberation but more often than not false hopes.From a French translation of Erkel, op. cit.

The Wait (August-December 2002)

The Initial Reactions of MSF, Arjan’s Family and the Dutch Government

Launching a crisis plan based on MSF-Holland’s standard protocol, the morning after the kidnapping MSF-Switzerland set up a crisis cell in Geneva under the responsibility of the Dagestan programme manager. The cell was later reinforced with internal specialists in human resources, communications, context analysis and administration. During the following month, field crisis cells were sent to Moscow and Makhachkala with support from other sections, primarily MSF-France. Overall supervision fell to a steering committee made up of MSF- Switzerland’s president, director general and director of operations.Internal MSF document.

While day-to-day operations and decisions remained with the Swiss section, the Dutch section was involved from the start. Initially, their participation consisted of the Dutch head of mission (MSF’s only head of mission in the country at the time) assuming temporary responsibility for the crisis cell in Moscow until reinforcements could be sent in from Geneva. In addition, in view of Arjan’s nationality, a crisis cell was also established in Amsterdam. The primary role of this cell was to provide support with dealing with the Dutch Foreign Ministry and managing relations with the Erkel family “with a bit more cultural sensitivity.”Interview with Thomas Nierle, former director of operations of MSF-Switzerland, 5 May 2015. Setting up an additional crisis cell in the abducted person’s country of origin is standard practice within MSF.

Informed on the day of the abduction, the family were far from self-effacing. To draw the attention of Dutch government circles to Arjan’s plight, his father Dick Erkel immediately started lobbying Dutch Foreign Affairs officials. A member of the ruling CDAChristian Democratic Appeal.
party, Erkel’s political connections enabled him to meet with the Dutch Foreign Minister in person within the first few weeks of the kidnapping and, more generally, to access government officials.Internal MSF document.

To begin with, the Dutch Foreign Ministry treated the kidnapping as a simple consular affair, but the case came to be afforded more importance with the involvement of senior officials within the administration. This included the setting up of a crisis cell headed by the director for consular affairs and frequent interventions by his superiors, but did not alter a policy of “silent diplomacy” whereby the Dutch government would officially limit its role to that of a facilitator and nothing more.“Positioning on the Arjan Erkel affair in the public domain”, MSF internal notes, 2004.
The Hague was formally committed to a policy of non-payment of ransoms and non-negotiation with hostage-takers, but did not prevent families, employers or other benefactors from doing so in their stead. The Dutch government seemed all the more reluctant to become directly involved as Russia was a key strategic and economic partner. Indeed, in 2003, Moscow had become the main supplier of crude oil to the Netherlands, and the country ranked as the Russian Federation’s third trading and investment partner.See “Russia main oil supplier to the Netherlands”, Statistics Netherlands Web magazine, 15 March 2004.

The Silence of the Kidnappers

Shortly after the abduction, rumours circulated in the local press as to who was responsible—the FSB, bandits or Chechen rebels. Claiming to know who the culprits were, Dagestani leaders blamed “Wahhabis, radical Muslims, the enemies of Russia”, while insinuating that the abduction was intended to demonstrate that Dagestan was still “dangerous and unstable.”The more conspiratorial made the link to oil politics in the Caucasus: it was in the interests of Western corporations to fuel chaos in Dagestan and foster stability in Georgia in order to promote Georgia as the main transport hub for Caspian oil and gas. Van Zwol, Gijzelaar van de Kaukasus, op. cit., pp. 11-12, 66.
Meanwhile, a number of humanitarian agencies made the connection between the kidnappings of Arjan Erkel and Nina Davydovich and the plan to repatriate Chechen refugees from Ingushetia. As one journalist reported in August 2002: “Under cover of anonymity, the heads of humanitarian organisations in Moscow stated yesterday that the purpose of these kidnappings was to make NGOs leave Chechnya and the surrounding republics as the refugees begin returning home.”“Résurgence des activités des rebelles tchétchènes”, Le Figaro, 21 August 2002.

In accordance with the usual recommendations drawn up by specialists in kidnap management, the MSF-Switzerland crisis cell initially adopted a patient, low-key approach.The need to exercise patience and discretion in order to facilitate the resolution of an abduction through a secret commercial transaction is one of the main recommendations delivered during trainings on kidnap management, cf. MSF-France Internal Training, “Abduction and kidnapping”, run by a former head of Scotland Yard’s Hostage And Crisis Negotiation Unit, Paris, 12-13 June 2013. This recommendation corresponds to the good practices endorsed by kidnapping specialists. See interview with Alain Juillet, former intelligence director at the DGSE and senior advisor for Orrick Rambaud Martel, in Ministry of the Interior, “Business en milieu hostile. La protection des entreprises à l’international”, Défis, no. 2, 2014, p. 11; Dorothée Moisan, Rançons. Enquête sur le business des otages, Paris: Fayard, 2013, pp. 91-95; Brynna Leslie, “In Harm’s Way”, Canadian Insurance Risk Manager, summer 2011.
Hoping that the Russian or rebel leaders would resolve the case quickly, the cell decided to “minimis[e] interferences and wait for the phone call” from the kidnappers, an intermediary or the authorities. During meetings behind closed doors, MSF appealed to Russian and Dagestani officials to provide “any information or any contact that may lead to a solution of the problem” while promising to “keep strict confidentiality on any support we obtain.”For instance the letter from Thomas Linde, general director of MSF-Switzerland, and Morten Rostrup, president of the International Council of MSF, to the ambassador of Russia in Switzerland, 23 August 2002.
Discretion was used in other initiatives, such as activating local networks and establishing contacts—including among Chechen opposition groups. The Dutch, French and Belgian sections were solicited for help with expanding these networks and attempts were made to consult with journalists, scholars and regional experts, particularly by the MSF team in Moscow. Publically, the demand was for Arjan Erkel’s “unconditional release”, with MSF restricting its press releases to highlighting the “risks civilians and relief workers” were facing in the Caucasus.“MSF Condemns Kidnapping of Relief Worker—Operations in Chechnya, Dagestan and Ingushetia Suspended”, MSF Press Release, 14 August 2002.
Approved by Arjan’s family, this approach fitted neatly with the Dutch government’s preference for quiet diplomacy.

The October 2002 Nord-Ost theatre hostage crisis in Moscow was emblematic of the discretion maintained by MSF during the first months following Arjan’s abduction. Despite being asked by the Chechen separatist hostage-takers to mediate, MSF kept a low operational and media profile, eventually providing medical supplies to local hospitals that had admitted survivors of the Russian Special Forces’ assault on the theatre. The crisis ended in the death of 129 hostages out of 850, all but one killed by the combat gas used by the special forces during their assault. The FSB refused to disclose the chemical composition of the gas, which triggered outrage among Muscovite health workers who had been exposed to it while trying to resuscitate unconscious patients.Medical staff practiced mouth-to-mouth resuscitation due to the large numbers of patients with respiratory distress and the lack of resuscitation bags.
MSF made a conscious effort to stay discreetly in the background so as not to antagonise the Russian authorities and complicate attempts to obtain their help to get Arjan freed. Up to this point, MSF’s communication on the kidnapping can be summarised as “saying next to nothing.”Interview (2009) with Anne Fouchard, former deputy communications director of MSF-France, in Binet, op. cit., p. 231.

A Change in Strategy (November 2002-August 2003)

Public Advocacy Makes a First Appearance (November 2002-February 2003)

For the remainder of 2002, there were no attempts by those responsible for Arjan’s abduction to contact MSF, the Dutch government or his family.

The absence of a ransom demand or proof of life fuelled suspicions within MSF of broader political machinations behind the abduction.MSF internal document.
As the then director of operations of MSF-Switzerland later recalled, “there was still this idea in the back of my mind that this type of exaction had to be authorized somewhere, that in this game, it’s the people with the power who hold the cards.”Interview (2009) with Thomas Nierle, in Binet, op. cit., p. 216.

There was general agreement with the Dutch government that the Dagestani police and Russian authorities were doing little to resolve the case. Then, following the Nord-Ost crisis, the first high-level meeting attended by MSF representatives and FSB officials was held. According to the MSF delegation, the FSB passed on a message that could not have been more clear. Arjan’s abduction “was the kind of thing that happened to people who pissed them off, people like us”.Interview (2009) with Jean-Hervé Bradol, former president of MSF-France, in Binet, op. cit., p. 229. The message that “Arjan got what he deserved” was repeated in May 2003 by an assistant director of the FSB during a meeting with the Dutch ambassador and MSF-Switzerland director of operations. Interview (2009) with Thomas Nierle, in ibid., p. 255.

Within MSF, the low-key approach adopted by the Geneva crisis cell was starting to be called into question. Opinions diverged on how best to push the Russian authorities into finding a resolution. Continue with silent diplomacy or publicly embarrass the Kremlin? An earlier move by MSF-Holland to contract a private security consultancy firm, Control Risks Group, had also triggered internal controversy.CRG had already provided advice to the Dutch section on basic training and crisis protocols prior to Arjan’s abduction, as well as during Kenny Gluck’s abduction.
Intended to reassure the family that “we would not put our principles above the utility of saving a life,” the move was poorly received by not only the crisis cell but also the French section. They criticised the lack of added value that such a group could provide and the damaging impact on MSF’s image of contracting a private security company known for its ambiguous links with Western secret services while Arjan’s meeting with US military attachés had already raised suspicions about MSF’s links to such services.“CIA without Borders” and “Doctors without Medicines” were already common epithets for MSF within the Russian army and secret services. See van Zwol, “Identity of the US Attachés was Unknown to Erkel”, op. cit.

By the end of 2002, the lack of progress in securing Arjan’s release convinced MSF-Switzerland’s director of operations to set up a brainstorming group to look into options not currently explored by the crisis cell and enable differences of opinion to be aired.Interview (2009) with Thomas Nierle, in Binet, op. cit., p. 232.
This “advisory group” comprised of key figures from the MSF International Office and the French, Dutch, US and Swiss sections, was not intended to replace the decision-making Geneva crisis cell and steering committee. It was during its first meeting in mid-November that the group suggested using the media to apply political pressure, first on the Russian authorities, and then the Dutch.

As the subsequent Geneva crisis cell communication plan would confirm, in an attempt to draw attention to the case (in private and then publicly), it was now time to step up the pressure on the Russian authorities to get them to secure Arjan’s release by underlining their legal responsibility to do so.“Implementing Phase 2—Rough draft”, MSF-Switzerland crisis cell, 13 January 2003. Initially, this would involve targeted lobbying of key international actors such as the UN, the US, Russia and the Chechen opposition, and, depending on the results of this lobbying, a public communication campaign with a potentially “aggressive or denunciatory message.”MSF internal document. It was later decided to tie in Arjan’s kidnapping with the broader political context, that is, a policy designed to deprive the Chechen population of humanitarian aid, with the passive consent of the international community.

The first obvious manifestation of the shift in MSF’s tactic became clear in February 2003 as the crisis cell in Geneva announced its intention to hold a press conference in the presence of Dick Erkel and the Dutch ambassador in Moscow. The launch of a petition was to follow in March. Both the Dutch Foreign Ministry and the Erkel family voiced their concerns when they found out about the content of the new communication plan. The Ministry considered “direct accusations against the Russian authorities” problematic in securing their cooperation while the family felt an accusatory tone “could work against the case”Letter from Jaap de Hoop Scheffer, minister of foreign affairs of the Kingdom of the Netherlands, to Morten Rostrup, president of the International Council of MSF, 29 January 2003; minutes of Arjan Erkel meeting, Ministry of Foreign Affairs, The Hague, 29 January 2003.—a criticism they would renew several times in the months to come. Exposed to pressure from Dutch public opinion and “dominated by the family and the government,”Interview (2009) with Rafa Vila San Juan, former MSF International Secretary General, in Binet, op. cit., p. 245.
MSF-Holland would use the same arguments to question the relevance of the advocacy campaign.

The press conference went ahead on 12 February 2003, although MSF did tone down its rhetoric. In calling on the Russian and Dagestani authorities to secure Arjan’s release, it was simply noted that the “non-resolution of this kidnapping may have a political dimension” and journalists were left to deduce the implications.“Arjan Erkel, Abducted Six Months Ago in Dagestan, Is Still Missing: MSF Calls on the Russian and Dagestani Governments to Give High Political Priority to the Case”, MSF Press Release, 12 February 2003.
Less than a month later, an international petition demanding Arjan’s release was launched, as the authorities “are not showing the slightest willingness to resolve this matter.”“Birthday of Arjan, an MSF Volunteer who has been Held Hostage for Seven Months in the Caucasus, Marked by an International Appeal for his Release: Russian Authorities not Facing up to their Responsibilities”, MSF Press Release,
In the Dutch press, MSF-Holland’s head of mission went even further, openly mooting a scenario of official Russian involvement: “It is possible that Arjan has been kidnapped by the security services or by someone else and that the authorities are happy to let it happen.”Fenneken Veldkamp and Coen van Zwol, “A Young Fair-haired Man is Worth Millions in Dagestan”, NRC, 5 April 2003.
Yet meanwhile, during private meetings with Russian representatives, MSF was passing on the message that it was ready to drop its campaign, pay a “service fee” to facilitate Arjan’s release, and offer public thanks to the Russian authorities for their efforts as soon as Arjan was freed.

In the meantime, information continued to emerge that pointed to the involvement of the Russian government in Arjan’s kidnapping and/or its failure to be resolved. In March 2003, Arjan’s mobile phone bill showed some sixty-one calls made after his abduction to various numbers in Moscow, the Caucasus and Rostov, including to policemen, members of the FSB, and known kidnapping intermediaries. Eventually dismissed by the Russian authorities as not relevant to the investigation, the number was disconnected shortly afterwards.van Zwol, Gijzelaar van de Kaukasus, op. cit., p. 14.
Then, in April 2003, the Moscow crisis team was informed by the Russian Federation’s deputy prosecutor that the FSB had witnessed Arjan’s abduction as they had been tailing him. The FSB justified their lack of assistance by claiming their agents were unarmed.van Zwol, “Identity of the US Attachés was Unknown to Erkel”, op. cit. According to Erkel’s memoir, the kidnappers also claimed to have noticed the FSB tail.
There was little doubt among MSF’s decision-makers: if the kidnappers’ identity and motives could not be established, at the very least, active obstruction on the part of the Russians was a fact.

First Proofs of Life (March-May 2003)

On 30 March 2003, while Arjan’s father and brother were in Moscow for the handing over of the international petition to the Russian authorities, the Dutch government asked them to return immediately to The Hague. As soon as they got back, they and members of the MSF crisis cell were shown the first proof of life, allegedly secured by the “services of a third country”. These were letters dating from late January, one addressed to each party. Accompanying the letters were photographs of Arjan. In the first he seemed relatively healthy, but in the second, taken on 27 February, he was unshaven and looked exhausted.

From MSF’s perspective, the timing of the first proof of life, unquestionably a positive advance after seven and a half months of silence from the kidnappers, could be interpreted as confirming the relevance of the shift in communication strategy, epitomised by the press conference and press releases denouncing the Russian and Dagestani authorities’ lack of willingness to resolve the case. More disturbingly, this proof of life was received right before more press conferences which the Russian and Dutch governments were particularly unhappy about.Internal MSF document.

After the proof of life, there was a pauseAn exception during this relatively quiet period was the EU summit meeting during which Erkel’s fate was raised publicly by President Putin and Dutch Prime Minister Jan Peter Balkenende. “EU Pledges Support for Russia’s Peace Plan in Chechnya”, AFP, 31 May 2003.
in the media campaign as potential channels of negotiation with the kidnappers opened and then closed. MSF did not pursue a Dutch government “secret contact” in Baku (Azerbaijan) very seriously because of concerns about the security of the MSF envoy. Then, a South Caucasian intermediary in contact with the crisis cell had one of his own men kidnapped while he was trying to liaise with the kidnappers in Dagestan.

First Cycle of Negotiations Ends in Failure (May-September 2003)

In May 2003, MSF learned that the investigation into the kidnapping had been suspended by the Dagestani police as early as November 2002 and had only recently been reopened.Internal MSF document.
One month later, the FSB handed over a video of Arjan to the Dutch Embassy. It included a threat: Arjan would be killed if a ransom were not paid.

This reinforced the view within MSF that the Russian authorities could, should they want to, provide a resolution. This was seemingly confirmed when, in July, the FSB identified an intermediary of the kidnappers who was prepared to meet with MSF. During a meeting in Makhachkala, this intermediary passed on a demand for a ransom of US$5 million to the Dutch government and MSF’s representatives (accompanied by an FSB-appointed “coach”), who requested proof of life before allowing the discussions to go any further.

A proof of life was delivered on 30 July, and Arjan later confirmed that around that time he was told to prepare for his release. Yet there was considerable frustration on the part of MSF that “if the FSB [was] capable of pulling strings behind the curtains”, why could they not conduct the negotiations themselves?Ibid.
On 12 August 2003, MSF launched its most forceful communication campaign to date to mark the one-year anniversary of Arjan’s abduction. It denounced not only the Russian government, which must be “reminded of their legal responsibility”, but also the “reserved attitude” of their Dutch counterparts.“One year after Arjan Erkel’s Kidnapping, MSF Considers the Investigation to be a Failure and Calls for more Action by the Russian Authorities to Resolve the Case”, MSF Press Release, 12 August 2003.

At this point, the communication channel opened up in Makhachkala via the intermediary identified by the FSB failed. Within the Swiss section, this failure was interpreted in very different ways. While the Geneva crisis cell concluded that this channel had already been blown, the head of the crisis cell in Moscow blamed the anniversary communication campaign. According to him, rather than a greater involvement of the Russian authorities in the negotiations, it had resulted in the FSB being unreceptive to the planned follow-up meeting. The Dutch Ambassador was furious for the same reason.Internal MSF document.

Less than a month later, in September 2003, a potential deal set up through the auspices of a Dutch lawyer and intermediaries in Ingushetia (who had connections with organised crime) appeared sufficiently promising for MSF to authorise for safekeeping the transfer of €250,000 via the Dutch Foreign Ministry to its embassy in Moscow. According to the lawyer’s intermediaries, the deal collapsed because of a police crackdown following the bombing on 15 September of the FSB headquarters in Ingushetia capital Magas. Further compounding the frustration was the Dagestani Ministry of the Interior’s announcement—then retraction—that Arjan was alive.

The Path to Release (September 2003-April 2004)

The Press and FSB Veterans Join the Investigation (September-December 2003)

By the end of 2003, the abduction and many of the details were being covered more extensively in the local and international media, irrespective of MSF’s communication initiatives. Local and foreign journalists were also conducting their own investigations. Writing in The New York Times about the shutting down of the investigation in November 2002, Bob Herbert noted that it would have remained so, had MSF “not succeeded in turning the case into an embarrassment for the Putin government.”Bob Herbert, “Kindness’s Cruel Reward”, editorial, The New York Times, 26 September 2003. A number of articles by Coen van Zwol and by Vyacheslav Ismailov (the latter a Novaya Gazeta journalist and former Russian army officer) published various names of people who they alleged either had direct links to Arjan’s kidnapping or had acted as intermediaries. Ismailov’s investigations placed responsibility with a Dagestan Duma politician who was allegedly implicated in the April 2002 FSB assassination of a Saudi militant fighting alongside Chechen rebels. According to the journalist, the kidnapper was a “gangster boss that poses as a strict Muslim, but is a double agent of the FSB”. As the kidnapper was also being sought by units of the Dagestani police, “Arjan Erkel [was] like a life insurance to him. ‘If you will kill me, then you will kill Erkel and that means a loss of face for the FSB.’ He plays around with this”, stated Ismailov.See Coen van Zwol, “Identity of Arjan Erkel’s Kidnapper is Known”, NRC, 5 November 2003, and Coen van Zwol, “Death for the Living”, NRC, 16 November 2003.

By autumn 2003, another FSB-supported intermediary, a veterans’ association of ex-security service personnel calling itself “Truth and Honour”, provided a new opening that looked promising. Referred by Arjan’s father, MSF contracted the veterans in September with the agreement of The Hague and support of the FSB. At the beginning of December, these veterans were confident in their ability to obtain Arjan Erkel’s release in exchange for €180,000 (Arjan later confirmed that around that time he was told yet again by his “keepers” to prepare for imminent release). But, as with the other avenues, the initiative failed due to the political turmoil caused by the arrest in Dagestan of a high-ranking police officer charged with complicity in kidnapping transactions.Internal MSF document.
Describing the sentiment within MSF-Switzerland and the advisory group at the time, the president and the general director of the French section wrote in October 2004:

"After the ultimate failure, in December 2003, a year and a half after the kidnapping of our colleague, we were deeply concerned for his life and completely demoralized by the disappearance of all concrete leads that would advance his liberation. In this period, Mr. Van Wulfften Palthe [a senior official of the Dutch Ministry of Foreign Affairs] recommended that we remain patient, prepare the Erkel family for the worst, and remain silent."Jean-Hervé Bradol and Pierre Salignon, “Arjan Erkel: enlèvement politique et mensonge d’Etat”, Revue Humanitaire, no. 11, 2004.

MSF Points to the Direct Responsibility of Russian and Dagestani State Deputies (January-March 2004)

On 14 January 2004, contrary to the advice of the Dutch Ministry of Foreign Affairs the advisory group proposed, in the event no channels worth pursuing emerged by 1 February, to “launch a strong diplomatic/communication strategy in late February/early March”. Phase one of the strategy did not differ in that it would continue the objective of maintaining Arjan’s international media profile to push for a “safe resolution of the case”. Phase two would highlight both the Dutch and Russian governments’ “total lack of political will to solve the case”, while the third, “j’accuse”, would be an attack on the Netherlands’ lack of commitment and Russia’s incompetence or complicity.“Crisis Cell Communication Plan”, February-May 2004. Of note, earlier lobbying had focused on the Russian Minister of Foreign Affairs Ivanov, Italy via Berlusconi, the Mexican government as chair of the UN Security Council, Kofi

While the Geneva crisis cell and steering committee decided to proceed with the plan, the Dutch section remained unconvinced that confronting the Russians was “the only way to do this.” Even if the Russian authorities were complicit, “pissing them off is not necessarily going to make them interested in releasing the hostage—it might make them interested in executing the hostage.”Interview (2009) with Kenny Gluck, former director of operations of MSF- Holland, in Binet, op. cit., p. 286.
The Erkel family were even more blunt, holding MSF responsible for Arjan’s prolonged detention and threatening legal action if MSF persisted with its communication campaign.“Embargo on Communications Campaign”, MSF correspondence, 26 February 2004.

Lukewarm attempts to reassure the Erkel family did not prevent the release of a press pack on 1 March denouncing “the scandal of international complacency” and the “local and regional equilibrium of power and profit” that apparently weighed more than Arjan’s life.“Arjan Erkel, Hostage in the Russian Federation since August 12, 2002”, MSF Press Pack, 1 March 2004.
A sense of urgency was fuelled by the news, received in February, that Arjan was “gravely ill with a chest infection and might well be executed before the upcoming Russian presidential elections, scheduled for March.”From a “source” in early February 2004, internal MSF document. It was later concluded by the team in charge of evaluating the management of Arjan’s kidnap, that this news was “a form of deliberate disinformation” by the kidnappers (of course, this was not known at the time) that provoked panic within MSF-Switzerland and the advisory group.
The press pack was followed up by interviews, notably with the president of the French section acting on behalf of the Geneva crisis cell. Alluding to Vyacheslav Ismailov’s investigations, he accused deputies from both Dagestan and the Russian Federation of direct involvement in Arjan’s kidnapping.See “MSF: Officials Implicated in the Kidnapping of the MSF Representative in Dagestan”, AFP, 9 March 2004; “MSF Accuses Russian Officials of Keeping one of Their Volunteers Hostage”, Le Monde, 10 March 2004; “Médecins Sans Frontières Accuses Russian Authorities of Complicity in Abduction of Aid Worker”, Associated Press, 11 March 2004.
Official reaction was limited to noting that MSF’s accusations were “unfounded and far-fetched.”“Russia Officially Accused over Kidnappings—MSF Makes Unpleasant Statements”, Nezavisimaya Gazeta, 11 March 2004.

The Veterans—Again (March-April 2004)

Arjan Erkel was finally released on 11 April 2004, one month after the press campaign and MSF’s public accusation against Dagestani and Russian officials. The exact circumstances are unclear. Officially, his liberation was described in the Russian press as the result of a “joint operation” carried out by Dagestan’s Ministry of the Interior and the local FSB and no further details were provided.“Arjan Erkel, Coordinator of the Médecins Sans Frontières has been Released in Dagestan,” RIA Novosti, 11 April 2004.
According to Arjan’s own account, he was transported in the boot of a car from his place of detention to a room at FSB headquarters in Makhachkala where officials told him he was free. He was then shortly debriefed before being flown back to the Netherlands via Moscow. The Dutch minister of foreign affairs claimed credit for the liberation, announcing his government had given the go-ahead for Arjan’s release to the same group of FSB veterans it had ended up contracting to negotiate with the kidnappers.Van Zwol, Gijzelaar van de Kaukasus, op. cit., p. 114.

A year and a half on, in contradiction of its own official policy of no negotiating, the Dutch government had finally become actively involved—the Dutch Foreign Ministry said as much during a meeting in Geneva on 25 March 2004 with MSF-Switzerland. Furious once again at MSF’s most recent communication campaign, it stated that the “Dutch government has no other choice but to enter into negotiations with Russian authorities”. He added, “the Dutch government will send MSF the bill for negotiations in the Arjan issue, whatever the cost will be.” And, a few months later, the Dutch government went on to sue MSF in court, asking for reimbursement of the €1 million “service fee” it said it had given in cash to the veterans’ association to facilitate Arjan’s release.See the minutes of the MSF-France Board of Directors meeting, 26 March 2004. As MSF rejected its claim for reimbursement, the Dutch State decided to sue the organisation’s Swiss section. After four years of legal proceedings and two rulings in favour of MSF in the court of first instance and then in the court of appeal, Switzerland’s Federal Tribunal ruled partially in favour of the Dutch State by pronouncing that the financial burden of the ransom should be shared between the two parties. “It is unacceptable to ask a humanitarian organisation to share the cost of a ransom negotiated and paid by a government. By agreeing to downgrade the consequences of the abduction of a humanitarian worker to a mere commercial dispute, as requested by the Dutch government, the Federal Tribunal’s ruling is contributing to making unpunished crimes against humanitarian workers— which have become more frequent in recent years—part of everyday life”, commented MSF. MSF Press Release, “Appalling Ruling in Court Case Between Dutch Government and MSF”, 14 July 2008.

Dealing with Radical Uncertainty

Twelve years after Arjan Erkel’s release, we still do not know the identities of his “takers” (the initial abductors), his “keepers” (those who kept him in detention for 607 days) or his “owners” (those who took the decision to abduct and then release him).

In his memoir, Arjan Erkel describes his “keepers” as Chechen Islamist militants who made repeated reference to the amount of his ransom and to the fact that MSF was not being targeted for its work or past outspokenness.From a French translation of Erkel, op. cit.
The FSB continued to refrain from commenting on the details of Arjan’s release or those responsible, and the head of the veterans referred only vaguely to “criminals, a mixed group of nationalities, including, I think, Chechens.” Meanwhile, despite taking credit for Arjan’s release, the Dutch government continued to maintain its stance of public discretion, notably with regard to the Russian authorities, and was unable or unwilling to provide details of those responsible.Simon Ostrovsky, “Light is Shed on Erkel’s Release”, The Moscow Times, 15 April 2004.

From MSF’s point of view, the prolonged tailing of Arjan, lack of assistance from the FSB agents present during his abduction, shutting down of the investigation, disconnecting of Arjan’s mobile phone from which calls had been made after his abduction to FSB and Russian military contacts and the initial lack of a ransom demand despite the proof of life, all contributed to an understanding that, if the Russian authorities had not actually ordered the kidnapping, at the very least they had not put up any opposition. “FSB fingerprints were everywhere—we ended up hiring them”, as one member of the advisory group summarised in 2015.Interview with Jean-Hervé Bradol, former president of MSF-France, 26 June 2015.

According to most diplomats and specialists on Russia consulted by MSF at the time, intense political pressure was required to secure the involvement of the Russian secret services—an analysis shared across MSF, despite diverging opinions on the public advocacy initiatives.
Despite its official demand for an unconditional release, MSF proved it was ready to obtain Arjan Erkel’s freedom at all times and whatever it took. In practice, the crisis cell consistently adopted an approach whereby political pressure was combined with informal proposals to settle the case with a confidential, financial transaction.

However, denouncing the Russian and then the Dutch governments was a particularly difficult decision in view of the opposition of Arjan’s family (and, of course, The Hague) to such a strategy. Relations between MSF and the Erkel family were tense right from the start, well before any disagreements surfaced about how to manage the kidnapping. Contributing factors may have included Arjan’s father’s disapproval of his son working with MSF, but criticism eventually focused on the strategy adopted by MSF-Switzerland, and more especially on the decision to engage in public advocacy, in contradiction of the recommendations of the Dutch government and the security specialists.Letter, Linde and Rostrup, to the ambassador of Russia in Switzerland, op. cit.Apparently convinced that Arjan’s captors were simply a gang of criminals and that the Russian authorities were genuinely trying to help, his father interpreted MSF’s strategy as a sign of the organisation’s lack of professionalism. MSF-Switzerland failed to share with the family its conviction that its strategy was in Arjan’s best interests.Interview with Michiel Hofman, former MSF-Holland head of mission in Russia, 25 June 2015.
Its decision to deal with them through an intermediary crisis cell based in Amsterdam, rather than the people directly involved in the negotiations, was a case in point.

So, was Arjan’s release facilitated or hindered by the media campaigns? While it may well be impossible to answer this question, there are certain observations based on the chronology of events to be made. The first proof of life and his eventual release occurred within a month of media campaigns. This tends to support the assumptions of the majority of the crisis cell and the advisory group: that the Russian authorities were implicated to the hilt and that they and the Dutch government (with the higher stakes it had with Moscow) would only step in under political pressure.

But, in the view of the head of the Moscow-based crisis cell, the 12 August 2003 campaign was actually counterproductive. The one-year anniversary of the kidnapping represented an important, emblematic selling point for the press and the campaign had acquired a momentum of its own. But the head of the crisis cell considered that, during the potential advances being made in August, it was tactically inopportune to “‘poke and prod the beast’ while they are trying to help”.Interview with Steve Cornish, former head of the Moscow-based crisis cell, MSF- Switzerland, 29 May 2015.
Given the direct involvement of the FSB in identifying an intermediary—with whom the negotiations appeared to be well underway— this was not the time to upset Russian secret services. Yet, the Geneva crisis cell and the steering committee viewed this lead as insufficiently promising to modify the communication campaign.

Public advocacy was certainly fraught with as many risks as opportunities. That the Russian authorities would react defensively to accusations of “mishandling the investigation and a lack of commitment” goes without saying.“Arjan Erkel is Still Being Held Hostage After One Year”, MSF Press Release, op. cit. More dramatically, pointing the finger at Dagestani and Russian officials could well have endangered Arjan’s life. The crisis cell was mindful of this possibility but decided to go ahead. While they were receiving alarming news about his health and threats of execution, and advice from the Dutch government to prepare his family for the worst, they considered that Arjan’s life was at greater risk if they just waited.

In the end, this account of Arjan Erkel’s abduction and release shows the limits of the technical guidelines and training described in previous chapters of this publication when it comes to dealing with a kidnapping situation. The veil shrouding the identity and motives of the “takers”, “keepers”, and “owners”, and thus the difficulty of assessing the consequences of decisions, generates an uncertainty that cannot be transcended by standardised algorithms and procedures. Ending a hostage situation is less about complying with protocols and more about feeling the way forward, regularly discussing and revising assumptions and decisions while seizing any opportunity to free a colleague as it arises (or for the hostage to escape).Among the three other MSF employees who were detained for a long period in the Caucasus, one escaped, one was released without conditions, and the third was exchanged for a relatively small sum of money.

After reading the first draft of this chapter in 2015, the director of operations responsible for taking decisions regarding the affair concluded: “The one thing which is positive is that MSF-Switzerland has always been quite transparent about the dilemmas faced and choices made, which allows us today to discuss the subject openly. In the end, we navigated in very thick fog and in stormy waters until the end and still today we don’t know what worked and what didn’t. ... We may have to ask the Russians one day!”