On Thursday November 15th 'M23' - the rebellion movement which was created earlier this year - launched a major attack on the city of Goma which, after a quieter day on Friday, culminated in a very stressful weekend. The question on Sunday evening was: will or will they not take Goma. On Monday they did. These are thoughts compiled throughout the last week, culminating in the fall of the city.
The birth and growth M23
M23 was founded when a part of the CNDP leadership returned to the maquis after the government of Congo tried to arrest their commander Bosco Ntaganda. There had been an International Criminal Court warrant on Bosco for many years and after the 2011 elections there was a lot of pressure on Kabila to deliver him as a sign a good will.
Kinshasa wanted to capitalize on this arrest to replace Bosco with a more loyal commander, thus dismantling (at least partially) the 'army within an army' that the CNDP had remained since it was integrated in to the FARDC in 2009.
Despite this integration, the CNDP maintained its parallel chain of command. Formally they belonged in the government camp, but neither the government nor the army command had a proper grip on the former rebel movement.
Bosco's departure was seen as a new beginning. Kinshasa searched in the circles of Congolese Tutsi for a new leader who had both the confidence of the CNDP leadership and was closer to the government. This would have been an important step towards real integration.
At the same time, a big CNDP taboo was broken - at the end of April the first contingent of CNDP soldiers was sent outside Kivu. For three years, CNDP had refused to operate outside of its home territory, afraid of having their troops dispersed over the country. These two developments were the real start of M23.
A lot has happened since then, but three elements are essential to understanding the movement's development.
First is the evidence provided by the UN Panel of Experts of the support given by Uganda and, to a much greater extent, by Rwanda to M23. This support has been political, technical - for example, providing facilities for communication to the movement's leaders - and, most crucially, military: recruitment, training and weapon delivery, even direct military support from the Rwandese army in certain operations.
Secondly, M23 hasn't received a lot of support in eastern Congo. Their capacity to mobilize and recruit people has generally remained low - considerably lower even than in the days of earlier rebellion movements which had their roots in the Rwandophone communities in Congo. They not only managed to mobilize a big part of the Tutsi community, but many Hutus as well.
This has not been the case this time. Very few Hutus have joined M23 and an important segment of the Tutsi have also refused to come on board. The Banyamulenge (Tutsi from South Kivu) distanced themselves from M23 from the start, this remains the case today.
Thirdly, the mobilization of military force by M23 was diminished by the prompt action of the international community, which reacted faster and sharper than usual. When it became clear that Rwanda was very actively supporting M23, it was heavily criticized by some of its most loyal partners. In Washington, London, The Hague, Berlin and Stockholm immediate measures were taken to cut or suspended parts of their bilateral support.
These measures clearly hit Rwanda where it hurts and also had an important discouraging effect in Congo itself (individuals and groups thought twice before joining M23) and in Uganda, which retained a lower profile and acted with greater discretion than Rwanda in its support for the rebel movement.
The attack on Goma
Initially nobody really believed that what we were dealing with was a rebellion that wanted to start a war. Many, myself included, thought that the first and only reason for M23's existence was to obtain, through negotiations, better positions within the army and the government.
On Thursday November 15th M23 launched a huge attack, firstly on Kibumba and then Goma. It was such a large offensive that it was way beyond its own military capacity. Very soon we received confirmation, through independent sources, of the massive support and even the direct involvement of the Rwandese army in the operation. What followed was days of confusing and often contradictory bits and pieces of information.
The first analysis coming through was that M23 did not intend to take Goma. Nkunda and the CNDP had reached edge of the town in 2008, but under pressure from Kagame and international diplomacy he had not taken the city.
This seemed the most likely scenario once again. M23 had already experienced several difficult months; putting extreme pressure on Goma seemed the best way to force negotiations from a position of relative strength.
Taking the city didn't look like a viable option, such an act would have highly significant consequences and the risk of violence and massacres in and around the town would be high. Taking Goma would change the entire outlook of the conflict, one which had already had heavy consequences for Congo.
The fact that the government has failed to bring the armed group under its control has accelerated the disintegration process within the army and has had a very negative influence on the relationship between the ethnic groups in eastern Congo. Goma's fall could set the peace process back many years.
On Sunday November 18th it became obvious that Goma would not resist the attack. Thousands of people fled the town and the surrounding refugee camps. The army and the political authorities also left. But on Sunday afternoon, as in 2008, the rebels stopped some kilometers short of the town.
Monusco conducted negotiations with the M23 leadership. They requested immediate and direct negotiations with the government which started on that same Sunday in Kampala. The governor of North Kivu returned to town, order was maintained by Monusco and the Congolese police. It looked like the worst had been avoided.
The fall of Goma
Early afternoon on Monday, two messages reached me simultaneously. M23 communicated that the negotiations in Kampala had failed because of the government didn't want them to succeed.
At the same moment, I received an SMS from a friend in Goma which stated that heavy shooting had started again, probably a few kilometers north of town. It was the beginning of another cascade of messages and information, unconfirmed, confusing and sometimes contradictory.
Heavy weaponry around the airport - had M23 attacked? Shooting of light weapons rapidly moving north - counter-attack of the regular army? Grenades in certain neighbourhoods of Goma - launched from the Rwandese border town of Gisenyi?
At least one shell was launched from Congolese territory. Many people on the run but most of them shivering under the kitchen table. The Republican Guards stopping people in the streets and stealing money and watches. Rwandese troops reported to have crossed the border. Direct confrontations between the Congolese and Rwandese armies. And on, and on.
The later it got, the more difficult it became to verify or double-check messages. I switched off my mobile, after I got the message "Angolese and Zimbabwean troops on the ground, ready to join the fighting tomorrow morning" - something I hoped and believed was absolute nonsense. But in any case, the events around Goma looked like they were going to become the worst crisis since 1998. On Tuesday morning the city finally fell.
On Monday morning everybody had believed that we were going linea recta towards another unworkable 'negotiated solution'. This would euphemistically be categorized as a 'peace agreement', which would create (through another empty form of power sharing) a bit of space and probably a cease fire.
But this ceasefire would contain no elements that could lead to sustainable stability or a workable political construction capable of building confidence and a common agenda between the groups, parties and communities involved. This did not happen. Twenty four hours later we were able to observe the bankruptcy of a mis-conducted, ill-accompanied peace process.
A legacy of failure
At this very moment we must acknowledge that all the money and work invested in security and democracy in Congo has resulted in a very limited sustainable impact. We happily believed that we had contributed to the rehabilitation of the Congolese state, but neither two elections (2006 and 2011) nor years of army reforms have allowed it to arise from the ashes.
The West was very ambiguous about democracy. On the one hand, we found elections very important, but on the other we went very far in accepting the undemocratic way they were organized.
The 'International community' invested in an army, but after all these years the FARDC has remained much more a part of the problem then a part of the solution. Programs and policies meant to reinforce democracy and security were designed and implemented by people in offices far away from the complex realities on the ground, by people with very limited understanding of them.
A lot of these policies were based on a rather superficial analysis of the problems in the region. We imposed on Congo a standard package of post-conflict measures and their accompaniments, not taking into account the fact that the conflict in Central Africa never really finished. Cruel wars and dreadful, unworkable peace agreements were the result, and now we are reaping what we sow.
In the eyes of the Congolese population, the West has lost all its credibility. Despite lip service being paid to democracy, the people haven't seen real commitment.
The big challenges weren't addressed: bad governance and poverty are endemic, the land issue remains a time bomb, the Congolese state is still very fragile and cannot rehabilitate the instruments it needs to guarantee the rule of law. This will not change whilst support is limited to the technical dimension of these problems.
If we really want to be loyal to the Congolese population, we have to understand the quality of the security and the democracy to which we contribute. This can only be achieved through open dialogue with the Congolese leadership, which will enable us to define with them clear bench marks which allow progress to be measured. We also badly need better insight in to the way local, provincial and national levels of government influence each other.
On top of all this comes the regional context. We already mentioned that Rwanda supports M23 in many ways. We suspected this since the very beginning and we have known it to be true since June, when the UN Expert Panel published its provisional report. Rwanda has made little effort to hide this, even if it has continued to deny it.
If Rwanda gets away with this, we will have to accept that future generations of Congolese will be stuck in a vicious circle of wars and dreadful, ineffective peace agreements.
Kris Berwouts has, over the last 25 years, worked for a number of different Belgian and international NGOs focused on building peace, reconciliation, security and democratic processes. Until recently, he was the Director of EurAc, the network of European NGOs working for advocacy on Central Africa. He now works as an independent expert on Central Africa.