Introduction

When we at MSFThroughout this article, MSF will refer to MSF-France, because the analysis concerned this section alone. The conclusions expressed here in no way pre-judge or anticipate any areas of agreement or disagreement between the French section and the other sections of the movement.today look at the specific problems and restrictions posed by military occupation, we immediately think of recent situations in which our action has been curtailed, either because we were obliged to terminate a mission, or because the debate behind the choices we had to make in such or such a context led to deep disagreement which - for varying lengths of time - were felt to be paralysing. These examples include our withdrawal from Iraq under American occupation, once the field team concluded in the spring of 2003 that it was unable to work there; our withdrawal from Afghanistan in the summer of 2004 after the murder of 5 members of MSF-Holland, against a backdrop of military/humanitarian confusion and calls for the murder of our members made by a Taliban representative; the dissent of part of the field teams on the Palestine mission in 2001, denouncing the culpable silence of head office and questioning MSF's principle of neutrality, along with the converse and recurring question of the reality of our role in the system of oppression and crushing of the resistance set up by Israel in the occupied territories. Common to these few examples is on the one hand the question of our neutrality, either in the way it is experienced by our teams, or in the way it is perceived by the forces present, thus raising the question of the safety of our personnel particularly acutely. They are also a part of the context of the "war on terror" - or, in the case of Palestine, have been partly reinterpreted in line with this new conceptual framework - and thus oppose a State or coalition of States against opposition groups designated as fundamentalist and terrorist. Are Western humanitarian organisations now faced with an unprecedented situation obliging them to rethink how they act, unless they resign themselves to no longer intervening in situations that are bound to become more commonplace, as those who promote the "war on terror" constantly remind us?For example, see the article by Cheryl Benard, "Afghanistan Without Doctors", Wall Street Journal, 12 August 2004.

If we look at this problem by starting with the notion of occupation, the choice is then to reposition our approach in terms of a longer historical perspective and to refuse to subscribe from the outset to the principle - implicit in the formula "war on terror" - of a new international context that demands new methods. Moving outside the framework promoted by the very partisans of this war is also to stake a claim to independence by looking to adopt a more objective, one could say a more "neutral" viewpoint. However, the decision to choose this notion of occupation then itself becomes problematical. It is a legal notion taken from the law of war and which refers to a de facto situation but it has nonetheless entered everyday vocabulary and has taken on a largely normative meaning. The proof is the virtually systematic refusal on the part of those States whose armed forces enter a territory which is not under their jurisdiction to consider themselves as occupiers. They stress instead their role as "liberator", "pacifier", or "saviour". Conversely many groups or political communities speak of occupation to denounce the illegitimate and oppressive nature of the political and/or military authority exercised over what they claim as their territory. Occupation is therefore clearly a highly politically charged notion.

For the purpose of this study, we decided to examine MSF's experiences in a variety of situations over the last 20 to 25 years, in other words from the time of our earliest missions in Afghanistan and our aid programmes for the Cambodian refugees in Thailand. All the contexts studied have in common the fact of having seen intervention by national or multinational armed forces from outside the territory considered, whether this intervention was the event that triggered a conflict or occurred as part of an already existing conflict. This minimal definition aims to avoid over-reliance on the notion of occupation by one or other of the protagonists as a criterion for inclusion in the study - a criterion which would be both too widespread and too partisan - or on the legal qualification by a competent international body - this criterion then being far too restrictive. In this study we therefore include armed interventions under United Nations mandate, even though the Security Council never imparted them the status of occupation, but only when the UN's mandated forces were in direct armed confrontation with at least one of the local armed forces and thereby themselves became a part of the conflict. These "theatres" include Afghanistan, Cambodia, Western Africa - during the ECOMOG intervention period - Somalia, Iraq, Chechnya, Kosovo and the Palestinian territories. This list however is not necessarily exhaustive.

The fact that the contexts were not in principle chosen with reference to the political standpoints of the parties involved in these situations does not of course mean that MSF's experience in these theatres is envisaged irrespective of these standpoints and the underlying perceptions and representations.For reasons of simplicity, and unless we specify who is speaking and in what context, I will use the terms "occupation", "occupier", "occupying power" and "occupied" during the course of the analysis, but with no value judgement of the parties or situations thereby designated.MSF's approach has for a long time included the political environment in which humanitarian action is deployed. This has been the basis, at least since the Ethiopian famine of 1984-85, of our insistence on vigilance and comprehension to avoid our actions being diverted or misappropriated, thereby making us complicit by default in unacceptable mechanisms of oppression. The purpose of this study is therefore to examine if and how military occupation situations represent a particular variation of this general dilemma that is inseparable from humanitarian action, and to ask questions concerning the specific operational constraints we are likely to face as a result.

Finally, to speak of the MSF viewpoint rather than that of humanitarian action in general, is to postulate a separate collective identity, an accumulated specific experience which includes but transcends the individual experiences of its members and which partly determines the scope of what is possible in terms of taking positions and making operational decisions when faced with a given situation. In other words, the aim will be to identify, if applicable, not only the constraints and pitfalls imposed from outside by the political framework of the occupation situations, but also our own institutional and cultural limits when faced with such situations, limits that could be perceived differently by other humanitarian agencies.

In the light of the above, the nature of the debate, the concerns, the positions of principle and operational decisions in the contexts considered will be assessed on the basis of two questions which, sometimes implicitly but in any case repeatedly, underpin any MSF mission: firstly, that of the validity of our intervention and our presence, both in terms of legitimacy and effectiveness - operational know-how; then, that of the meaning of our action in the light of the behaviour of the forces present and the political dynamics at work, and incorporating the risks that our action may imply for our teams and for the populations we are attempting to help. Finally, to conclude, we will examine the specifics of the problems identified with respect to other missions and contexts and will attempt to identify broader areas for reflection around the notion of occupation.

1. "Do we go in? What do we do there?": legitimacy and effectiveness under occupation

BEFORE THE OCCUPATION: THE INVASION

By definition, occupation is the result of invasion, and therefore an act of war. It is this necessary antecedent of war with respect to an occupation situation that from the outset and almost intuitively legitimises intervention by MSF. The birth of the modern humanitarian movement on the battlefield, with the creation of the International Committee of the Red Cross, and the appearance of the "without-borders" movement in the wake of the Biafra war, meant that war and the goal of minimising its effects on the civilian population have been at the core of MSF's realms of intervention. This principle was reaffirmed on several occasions, both in the light of incidents that occurred in certain fields, and the de facto reduction in the share of the organisation's missions as a whole being devoted to areas of conflict. Thus, after the murder of three expatriates in Afghanistan and Sudan between 1989 and 1990, the question brought before the Board, of whether or not to continue with the war missions led to a consensus, "with each person considering that this is one of the underlying and fundamental aspects of MSF."Rony Brauman, President's Moral Report, 1990This collective priority was once again underlined at the 2003 General Assembly: "We wished to refocus on the victims of conflict, and that is what we have done".Jean-Hervé Bradol, President's Moral Report, 2003 The presence of MSF in the theatres of war is thus clearly the core of the association's identity, but to maintain this operational direction demands a proactive approach.

Therefore, anticipating the fighting and the care to be given to the victims when the question arises of intervening in an imminent theatre of war has always taken precedence over the possible issues involved in a subsequent occupation. This is clearly shown by MSF's experiences in the two "Gulf Wars". In the summer of 1990, after the invasion of Kuwait by the Iraqi army, the nature of the regime of Saddam Hussein and France's involvement in preparations for the first Gulf War had led to serious misgivings about sending out a mission to Iraq. However, as of the month of October, while the debate was going on internally, the first visa applications for sending out an exploratory mission were sent by MSF to the Iraqi embassy. In January 1991, with the beginning of international military operations against the Iraqi army in Kuwait and then in Iraq itself, the arguments rapidly swung in favour of intervention by MSF. This started at the edges of the conflict, in Jordan and Syria, by taking care of foreigners and refugees fleeing Iraq at war. However, the decision to intervene in Iraq had already been taken and now simply depended on authorisation from Baghdad. The fears of Iraqi manipulation and the risk of a political interpretation of MSF's actions gave way before the need to reaffirm the identity of the association: "the humanitarian principle cannot be called into question", "if we don't propose going to Iraq, then why do we exist?".Minutes of the meeting by the MSF Board, 25 January 1991 From the end of 2002 to the beginning of 2003, as preparations were being made for the second Gulf War and when, unlike the first time, overthrow of the Iraqi regime and occupation of its territory by the American armed forces were Washington's clearly stated goals, logistical preparations and visa requests were made in a similar fashion, even though the decision to intervene had not been taken. Indeed, the in-house debate was continuing, more often than not informally, in an extremely polarised context, both at the international level and inside France. Reservations within MSF abated, even if they did not completely disappear, when a few days before the American offensive was launched, the visas were issued by the Iraqi representation.

Even if triggering of hostilities requires intervention by MSF, it was not during periods of invasion that we managed to build up our operational effectiveness. Experience in the fields covered by this study shows on the contrary that routing and deployment of aid at this stage of a war is extremely haphazard.

Primarily, the frequent use of shelling or bombing constitutes a considerable risk to the physical safety of the teams and also makes it very hard for the victims to reach the health care structures likely to assist them. To this risk can be added the threat of hostage-taking, of use of the teams as "human shields" or direct reprisals against MSF by the regime or the armed forces targeted by these large-scale attacks. These threats are particularly great when the offensive is being led by western forces and we are perceived either as being linked to them and therefore a legitimate enemy, or as a means of pressure or a valuable bargaining counter with the belligerent western governments. Thus, in March 1999, as of the first days of NATO's aerial bombardment of the Serbian forces in Kosovo and Serbia, we were forced to withdraw from Kosovo to Montenegro.Philippe Biberson, President's Moral Report, 1999: "At the end of March, the beginning of NATO air-strikes had been preceded by the withdrawal of OSCE observers. As of the initial strikes on Pristina and Belgrade, all the NGOs withdrew their teams and their representatives. This was also the case with Médecins Sans Frontières who, after holding out for a few days more in Pristina - although with instructions to do nothing - decided to leave. Given that it was impossible to act directly inside Kosovo, Médecins Sans Frontières based their actions in the three neighbouring countries."Similarly, in Autumn 2001, MSF expatriate teams left Afghan territory under Taliban control before the beginning of the air strikes against the regime in Kabul by the United States and its allies. In 2003, in Iraq, the Iraqi regime imprisoned two expatriates from the MSF team, which had remained in Baghdad during the American offensive, thereby obliging us to suspend our activities. It is worth noting - and we will come back to this point in the second part - that in these three cases, it is more the threat posed by the regime or armed groups on the receiving end of a western offensive, than the direct risk from bombing, that we felt to be a determining factor in our inability to work during the air strikes.We should for example mention the presentation by Graziella Godain to the Board Meeting of 26 March 1999 (two days after the NATO air-strikes began): "It is very difficult to move around and the situation is fairly confused. We held many security meetings before sending out exploratory missions, because anti-foreigner feelings are running high." Also see the presentation by Pierre Salignon to the Board Meeting of 28 September 2001: "MSF's foreign volunteers were evacuated from the zones controlled by the Taliban in Afghanistan, not so much for fear of American air-strikes but more because there were no guarantees from the authorities concerning the safety of foreign personnel (linked to the increasing numbers of non-Afghan Muslim militias). Just when the question was being debated of whether to leave a small team in place, for each of the three sections present on the Taliban side, the authorities ordered the departure of all representatives of "western" institutions."

In invasion situations where we have been able to act in the theatre of war itself, either using expatriates or local teams, assessing these actions has usually proved problematical. Thus, during the Russian offensive on Grozny in the summer of 1996, as during the Israeli offensive on Jenin in 2002, the MSF expatriate team was unable to gain access to the injured and all they could do was ferry medical equipment to the area under attack, with there then being no possibility of assessing how it was used.Presentation by François Jean to the Board meeting of 6 September 1996: "During the period 6 to 20 August, Grozny was under attack from Russian forces and there was a very clear desire to obstruct humanitarian aid in the capital. Drugs and equipment were nonetheless carried on foot from Nazran to Grozny. It was hard to assess the needs, how things were working and what was happening to the wounded."Although close to the site of the offensive, this action therefore differed little from the donations of drugs and equipment by the MSF teams in Montenegro to groups linked to the UCK crossing the border with Kosovo in the spring of 1999. During this conflict, as in fact in many other cases, emergency aid was concentrated in the neighbouring countries for the refugee populations. In those situations in which local teams who stayed behind after evacuation of expatriate personnel have managed to provide medical care - which is necessarily limited when the offensive is at its fiercest - this has been the result of individual acts of bravery, accepted and indeed praised, but not recommended by head office. This was the case of the MSF Afghan teams during the air-strike phase of the American "Enduring Freedom" operation in autumn 2001.

Unless a major risk of direct political targeting is identified, we can therefore suppose that MSF will not on principle rule out the possibility of maintaining an effective presence during an invasion, even if this phase of hostilities rarely permits large-scale care being provided for its victims. This refusal, confirmed in several contexts, has not however prevented worries and major disagreements from breaking out at head office over whether or not a team should be maintained in each of these situations.This type of disagreement naturally happens in the field but, in the final analysis, it is how the debate is decided at head office that is determining because a negative decision requires evacuation by the whole team, whereas the decision by certain expatriates individually not to stay is not binding on the rest of the team.Nonetheless, the absence of any position of principle for this type of context suggests institutional confidence in the ability of certain experienced individuals to creating a working space - even symbolic -during the invasionThe authority wielded by François Jean doubtless explains to a large extent why the operation conducted in and around Grozny in the summer of 1996 was ratified by the Board, even after the fact. The cross-border operation in Afghanistan, although it started after the actual invasion by the Red Army, is also a good example, as witnessed by the homage to Dr Gérard Kohout in the 1982 Moral Report: "Without him, this mission would never have been possible. For more than a year now, this tireless voyager has travelled the length and breadth of the country undercover […] stopping here to give medical care, here to discuss the transit of medical teams with the resistance leaders, or there to debate in a tea house and explain in Farsi that not everyone has abandoned the Afghan people." The limits of the mission thus established were not hidden: "Our great regret is that in the conditions in which we are working, we cannot provide medical care that is as effective as we would wish and we sometimes have the impression of simply distributing drugs, because the war prevents us from envisaging the various aspects of a complete health care intervention."but also, and doubtless to a greater extent, to paving the way for deployment of effective action once the offensive is over.

INSTALLATION OF THE OCCUPATION

During a time of war, occupation of a territory in principle indicates a military victory and, at least to start with, a significant reduction if not a cessation of fighting, even temporarily. Given the problems, sometimes the impossibility, of maintaining a working space during the offensive or at the height of the fighting, the return to relative calm offers MSF the opportunity to regain the initiative. In Liberia, the ECOMOG intervention in September 1990 and its pacification of Monrovia reopened access to the Liberian capital to MSF in November.See 1990-91 Activity Report, Liberia mission: "In September 1990, the first intervention troops (ECOMOG) from the Economic Community of West African States (ECOWAS) arrived in Monrovia to pacify the capital and separate the protagonists. As of November, Monrovia was once again accessible and enjoyed a situation of precarious calm."In Kosovo, in 1999, it was the capitulation of the Serbian forces, along with the cessation of air-strikes, that enabled the MSF teams to settle in the hospital in Pec even before the deployment of the KFOR. In Afghanistan, on 13 November 2001, the flight of the Taliban from Kabul and the end of air-strikes on the capital opened the way for a return by MSF teams from Panshir, a few hours after the forces of the Northern Alliance entered the city. The Coalition's land forces were not to enter until several days later. In Iraq, the release of the two expatriates from prison, due to the collapse of the Baath regime, was followed immediately by a mission to assess the hospitals in the city, where the bombing had stopped.

Initially conceived as a regional peacekeeping force between the various armed groups in Liberia, ECOMOG was to play no role in organising the health care system. In Kosovo, and in Afghanistan and Iraq however, the international forces which deployed into the field were from the outset to play a determining role in the political issues involved in stabilisation and reconstruction. In these three cases, MSF's role was to fill the void left by the end of the bombing and move back into an area from which the expatriate teams had been excluded. We also had to find our place in health care structures that had been disorganised by the war, before the occupying forces attempted to establish a form of order in which political and authority considerations would radically complicate negotiation of our working space. In the case of Iraq, we were refused permission to set up independent assistance programmes in the hospitals almost immediately by the occupying power. In Kosovo and Afghanistan, the time immediately following the offensive was the time for MSF to carry out emergency operations: urgent care to be given to the sick and wounded who had received no help at the height of the fighting, urgent need to deploy and organise our actions as independently as possible.

It is therefore clear that the freedom we are given to define how we are to intervene depends to a large extent on the mandate or the political and military objectives of the "occupier", regardless of how one defines it. We will come back to this point later on. However, it is important as of now to stress the fact that this time of "emergency" between the cessation of fighting and deployment of political/military authority varies considerably from one situation to another, precluding any predetermined and systematic approach by MSF. For example, in Somalia, the American military intervention in December 1992 in response to the current famine was regarded with great suspicion. The precarious security situation characterising Mogadishu since 1990 nonetheless led the MSF Board to opt for a position of prudence. This consisted of a press release combining the desire to cooperate with a warning against the possible perverse effects of an international military operation with a relatively vague mandate.See minutes of Board meeting of 4 December 1992. The result of the vote on MSF's position concerning the basic question of intervention by American troops in Somalia shows a lack of consensus on the subject: 7 votes against, 2 votes for, 2 abstentions.This same prudence was advocated by a majority of the members of the Board concerning the situation prevailing in Baghdad in the spring of 2003, in the first few weeks following entry by the American forces into the capital. However, in a political context that was far more heavily polarised than it was surrounding the arrival of the Marines in Mogadishu in 1992, the absence of agreement between the various members of the expatriate team on-site, the head office and the Board, concerning whether or not there was room to work during this period, finally led to withdrawal.From the Board's point of view, see the 2003 Moral Report: "for the Board, it would have seemed reasonable that the problems of the sick in the immediate aftermath of the war would have led to temporary initiation of activities. If this was not possible in the public hospital system, the situation was probably nonetheless sufficiently open for us to work in the private sector, by ourselves renting premises in order to offer consultations to dress a number of wounds and in particular in the popular quarters of Baghdad, to offer medical emergency aid. This was not the option chosen by the teams, and the board regrets the fact."

Finally, there are situations in which, despite the cessation of fighting or an initially very low level of resistance, the conquest of a territory or city is associated with an outpouring of violence against the civilian populations, who are made to pay the price of defeat by the victorious army. This was repeatedly the case with the Russian forces in Chechnya and the Rwandan and Ugandan troops in the east of the Democratic Republic of Congo (DRC) between 1998 and 2003. In both cases, these large-scale atrocities took place behind closed doors as a natural extension of the offensive before the imposition of order. We then had absolutely no margin for negotiation for access to the victims, regardless of whether or not after assessment we would have chosen to pull out as we had done in the Rwandan refugee camps in Zaire in 1995. In this type of situation, in the same way as during periods of invasion, it was only possible to assist the victims around the periphery of the conflict, in refugee or displaced persons camps, where the populations who had managed to escape sought sanctuary.

Hence, the variety of conditions in which an army occupies a territory runs counter to adopting a position of principle, and to any predetermined operational response on the part of MSF. It doubtless explains that the notion of occupation only very rarely forms part of the debate over the response to be provided to each of these situations. The operation carried out by the United States against Iraq in 2003 could be an exception in this respect. However, the prospect of the American occupation and MSF's lack of familiarity with the Iraqi context both contributed to the failure to reach a consensus on this situation.

LONG-TERM OCCUPATION

Whether they allow a precarious calm to return, or on the contrary justify unbridled violence, occupation situations and their implications for our working conditions evolve with time. It is clearly this evolution that is at the heart of MSF's operational problems. The paradox in fact lies in that, because they are part of a conflict and directly place a foreign army in contact with a civilian population, occupation situations are very clearly legitimate areas for intervention by ourselves, but at the same time, they create pockets of relative peace or normalisation of the war, in which economic and social life resumes - or more precisely adapts - and in which we struggle to define our programmes and characterise the real meaning of our presence. In other words, in MSF language, even though we consider the occupation to be a conflict situation, we are usually obliged to deploy "post-conflict" type programmes, likely to be applicable to the long-term.

This problem with categorisation is apparent in how the notion of occupation is employed by the MSF representatives in describing a situation, or not. We only rarely give the name of occupier to armed intervention forces under a United Nations mandate, even in situations in which they exercise effective control over all or part of a territory. If we distance ourselves from them, it is because they are "parties to the conflict" or are likely to become so. This distancing can nonetheless lead to opposing attitudes depending on the actions of the forces under UN mandate. In Somalia in 1993, bombing of civilians and of a building shared by MSF and ACF led us to file a complaint with the UN concerning the applicability of the Geneva Conventions to its forces. In Kosovo, however, the control exercised over the territory by the 40,000 men of the KFOR, imposed on Serbia as a result of the NATO air-strikes, helped with the widespread deployment of aid. In this context, we terminated our programmes on seeing that reconstruction was under way. Internally, the expatriate team observed that the KFOR was itself the only rampart against a further upsurge in violence between the Serbian and Kosovan communities. Although these notions do not appear in our debates, these two examples suggest that a distinction between "belligerent occupation" and "peacetime occupation" could be useful.

In situations in which armed opposition to foreign troops is being reorganised, then we would tend to use the term "war" rather than occupation to qualify the situation, and to justify our continued presence or our mobilisation, if such a presence were to prove impossible. In his 2001 Moral Report, the President of MSF described the situation in Palestine as the resumption of a "war that dare not speak its name",Jean-Hervé Bradol, 2001 Moral Report whereas the situation in Chechnya was a "total war",Jean-Hervé Bradol, 2003 Moral Report; in his 1995 Moral Report, Philippe Biberson had already described the conflicts in the Caucasus, in particular in Chechnya as "all-out war situations which make no distinction between combatant and non-combatant populations." a war "referred to as a "counter-terrorist operation" by Moscow"Jean-Hervé Bradol, 2001 Moral Reportwhich had to be brought out into the daylight, despite the occupier's claims. In these two contexts, reference to the notion of "occupation" is far less to characterise a particular politico-military situation than the nature of the suffering experienced by the civilian populations: "foreign occupation is a plague on the Palestinian population: in medical, material, moral and psychological terms", occupation and violence contribute to the same "humiliation";Jean-Hervé Bradol, 2002 Moral Report; we find this same meaning in the article by Pierre Salignon, Fouad Ismael and Elena Sgorbati, "Healing the Mind", a prologue to the MSF Report Palestinian Chronicles: Trapped by war, July 2002: "Nearly the entire Palestinian population suffers from the confinement, occupation, fear and an absence of hope for the future on a daily basis." in Chechnya, "the Russian army is deployed like an army of occupation and multiplies the acts of violence (disappearances, rapes, etc.)".Minutes of Board meeting, 28 January 2000; similarly, in denouncing the "colonising attitude" observed among the international contingents arriving in Somalia, Rony Brauman did not talk of a situation of actual occupation, but rather of "the impunity enjoyed by people who sometimes behave as killers, as all-conquering warriors: they feel that they have the right to take the women and even the lives of anyone who could in any way represent the slightest threat to their own safety." Rony Brauman, 1993 Moral Report Using the notion of occupation would therefore seem inseparable from situations involving humiliation and widespread abuses. Occupation is thus seen as something which can only generate more violence and a fresh intensification of the war. In short, in a context of occupation, MSF perceives its role as not only that of calling a war a war, but also of anticipating and sounding the alarm concerning a radicalisation of the conflict."Only the optimism of an ideologist could let one imagine that armed occupation could bring about anything other than more instability and violence." Pierre Salignon, Rony Brauman, "Irak : la posture du missionnaire" in Fabrice Weissman (ed.), A l'ombre des guerres justes, Flammarion, 2003, p. 291 ; "The military occupation is becoming more firmly entrenched […]. This clear deterioration in the situation, seen in the high toll of civilian dead and injured over a period of a few weeks, does not bode at all well for the future." Jean-Hervé Bradol, "Put to the test", Palestinian Chronicles, July 2002

The purpose here is not to contest the analyses made of these various contexts, but to underline the fact that the occupation is not seen by MSF as a particular situation or moment in a war, but more as a series of abuses, usually arbitrary. The persistent violence perpetrated against civilians, the continuation or the emergence of guerrilla actions against the occupying forces is, as far as we are concerned, a sign that the war is real, a war admittedly that can vary in intensity, but in which an emergency is at all times still possible. This is the central motivation for our presence. However, examining those situations in which we have made the decision - often imposed on us - to suspend or even terminate our missions, clearly shows that occupation contexts confront MSF with the limits of its action.

Between 1992 and the first months of 1993, MSF withdrew from three countries which were home to some of its largest missions: Iraqi Kurdistan, Somalia and Liberia. In these three cases, the rising insecurity, primarily borne out by the deliberate targeting of NGOs, was a determining although not the only factor in this withdrawal. In Iraqi Kurdistan, there was perceived to be "no real emergency situation" and despite human rights violations and political problems, there were in the end "few things on which a humanitarian organisation such as MSF could have any real impact".Rony Brauman, Moral Report 1993The assessment of the situation in Somalia was similar: "The emergency phase is over, and the time has now come for reconstruction."IbidemFinally, in Liberia, our withdrawal was justified "on the one hand by the multitude of humanitarian agencies present in Monrovia, and on the other by the fact that it is impossible to gain effective access to the areas of conflict".Activity report 1992-1993 To be sure, the bombing by ECOMOG aircraft of an MSF convoy heading towards the rebel zones was, for good reasons, in everyone's minds. However, this incident only occurred after our decision to withdraw from Monrovia's and Bomi's public hospitals, hospitals in which we had got things up and running again and in which the local health authorities now aimed to assume full control. In these three countries, it is therefore the absence of "a fair balance between on the one hand the real impact, and on the other the risks we are prepared to run to ensure an effective mission"Rony Brauman, Moral Report 1993. It should be noted that the question of the meaning of the MSF presence in Kurdistan had been posed at head office and to the teams, even before the rise of insecurity, culminating with the murder of a Handicap International expatriate, probably at the initiative of the Iraqi regime: "In the 1992 Moral Report I had already mentioned our role as a relative bystander […] It is in this context that the security problems convinced us to withdraw from Iraq." As with Liberia, the withdrawal from Kurdistan was justified by the fact that the areas in which MSF was needed most, zones which were not in fact occupied, were out of reach: "I would like to underline the fact that even though the spotlight is turned on Iraqi Kurdistan, the greatest difficulties, the greatest human tragedies are in fact occurring in the centre and south of the country, where no humanitarian organisation can penetrate." (Ibidem)which led us to opt for departure.

We have been faced by this dilemma between safety and the added value of our actions in other contexts. In Chechnya, the fall of Grozny in February 1995 and the gradual occupation of the south of the country by the Russian army had created a highly volatile situation, in particular during the first half of the year. Despite the positive results obtained by adapting the size of the programmes to this changing context, the constantly shifting battlefield permanently called into question the pertinence of where the teams were positioned."Overall, since the beginning of the ceasefire, there have been far fewer admissions (down by a factor of 10). Is this due to a falling off in the fighting or to the fact that the wounded are being taken elsewhere?" Minutes of Board meeting of 24 February 1995; "It is impossible to be with the civilian populations. We estimate the number of displaced persons at about 400,000, but many of them left before the conflict and there is a high proportion of Russians in the population caught in Grozny. […] The problem is to know what do we do in Chechnya. Do we stay or do we leave?" Minutes of board meeting of 9 June 1995. As of July 1995, a process of reconstruction was begun in Grozny with return of the Chechen refugees on a large scale. However, this was rapidly followed by a rising number of security incidents affecting the NGOs, leading to withdrawal by MSF in November."Massive amounts of aid reached Chechnya as of July 1995. The security incidents to which we were exposed for 4 months in succession led to withdrawal of the teams." Activity report 1995-1996.Finally, in Afghanistan, we publicly stressed the direct threat to our teams as the reason for our withdrawal in the summer of 2004, following the murder of five members of MSF-H with the crime being claimed by a representative of the Taliban. This very real security issue should not however make us forget that our working space and our working prospects were further complicated daily by the influx of NGOs and the Afghan health system reconstruction plan put together by the World Bank.Concerning the questions raised at MSF by the policy of health care system reconstruction, see Simone Rocha, "Use and Abuse of Humanitarian Aid in post-conflict Afghanistan: the example of health care provision", internal report, July 2004 and Xavier Crombé and Denis Lemasson, "Is Independent Humanitarian Action over in Afghanistan?", The Afghanistan Monitor,September 2003.

The problem of the definition and effectiveness of our programmes is thrown into the spotlight by our withdrawals when the safety of the MSF teams is threatened, but it is also the subject of recurring in-house discussions concerning the occupation contexts in which we remain present. Before the decision was taken to wind up the MSF mission in Kosovo -a decision mentioned earlier - the operational choices made by the field teams and the desk responsible for this mission had to be defended internally against a certain scepticism, owing to their atypical nature: emphasis on psychological care after the initial involvement in disorganised and understaffed health structures and in the absence of any large-scale, pressing medical needs; supply of materials for reconstruction of damaged roofs as winter approached.The presentation by Philippe Biberson to the Board on 27 August 1999 clearly shows the underlying effort of explanation required to have this "1000 roofs" operation accepted: "The question of what is a humanitarian"need" … and what is a luxury is one that will not go away. A roof at a given moment in a specific history may be a humanitarian need in an emergency situation. One of the preconditions determining MSF involvement is also to meet needs that nobody else is meeting."In the Palestinian Territories, the relevance of the psychological care programmes has led to the same soul-searching within MSF, since the first programme of this type opened in 1994.See the question from one of the Board members at the meeting of 27 October 1995: "despite its value, is this mission not missing the point? isn't MSF present because a page of history is being written and we had to be there?"This debate, which is usually expressed informally, or whenever activities and decisions are presented to the operations meeting or to the administrative board, concerns either the value of the psychological care itself in this context, or the fact that it accounts for the majority of MSF actions in the Palestinian Territories. In effect, despite a clear desire on the part of the desk to incorporate a medical side into the programmes through the presence of a doctor alongside the teams of psychologists and psychiatrists, this medical dimension in practice proved relatively unsatisfactory. First of all, the expatriate doctors had trouble in situating themselves with respect to well-equipped care structures and competent Palestinian medical personnel, who were often extremely reluctant to work with them. Furthermore, this medical presence had hitherto proved unable to provide an appropriate response to the localised emergencies caused by the Israeli operations, either because we were refused access to the victims, or through fear or lack of reactivity on the part of the MSF medical teams in situations where there was very little room for intervention and the number of victims was relatively limited.

Although deadly over time, the occupation situations encountered by MSF in effect only cause few victims on a day to day basis. They require no large-scale care operations by our teams and lead us to question the impact of our activities, all the more so as the violence against the civilian population is increasingly targeted (imprisonment, executions, etc.) leaving us little opportunity to provide care for direct victims of the occupying forces. The particular nature of this type of conflict is not overlooked, as seen in the frequent references to the notion of "low-intensity warfare". This is particularly the case in Palestine since 2001."The current situation in Palestine, particularly in Gaza and Hebron, can be defined as a state of "low-intensity warfare"." Christian Lachal, "Gazaouites et Hébronites" Mission to assess the emergency programme in the Palestinian Territories, January 2001; "A permanent order has been created by the violence. It is the product of a war that is cruel, but of low-intensity and spread out over decades." Jean-Hervé Bradol, "Put to the Test", Palestinian Chronicles:Trapped by War, July 2002 ; "Every day one or two civilians are killed, and then another two the following day somewhere else. Usually people going about their daily business but whose movements were felt to be suspect by the soldiers." Jean-Hervé Bradol, Moral Report 2002However, this identification of a particular situation does not seem to have provided answers to the operational difficulties encountered on a constant basis by the MSF teams. Neither has it managed to convince all members of the organisation of the relevance of undertaking atypical programmes, the impact of which it is hard to assess accurately.

The first problem faced by MSF in occupation situations thus arises from the contradiction between, on the one hand, an analysis in which the primary representation is one of war - the theatre of choice for humanitarian action as we conceive of it - and on the other, the solutions we are able to provide in these shifting situations in which violence and even fighting coexist with day to day life for the populations and with possible dynamics of reconstruction. This contradiction triggers a recurring internal debate in these terms: if there is a war, then why do we not respond in a conventional way? If there is no room for programmes tailored to war or if there is not really a war, then why are we going there or why are we staying there? This is for example the debate surrounding the proposed return of the local team to Chechnya that took place during the operations meeting in October 2004. The desk proposed opening a maternity/paediatric health care programme. The terms of the debate surrounding the proposal were precisely focused on this apparent discrepancy between such an approach and the logic of war at work in Chechnya, and the imbalance between the risks involved in returning to Grozny, on the one hand, and the probable impact of such a programme, on the other. The desk's argument of greater safety than before and a certain normalisation of life in Grozny, as manifested by the return en masse of the Chechen refugees from Ingushetia, then raised the question of whether or not there was actually a war and whether our return was relevant in a context of general but low-level insecurity and of reconstruction. The proposition was finally accepted on the principle of an evaluation of our ability to act in such a context through an initial programme in a Grozny maternity clinic. The programme proposed could not therefore constitute the sole operational objective of the mission. As with the psychological care programme in Palestine, the underlying objection - unevenly shared - is that of a "programme by default".

The existence of pockets of normality or dynamics of reconstruction during the course of a conflict is not however specific to occupation situations. Most long-term conflicts go through periods of "neither war nor peace", as we often call them, requiring that the nature of the programmes be adapted to changing needs. As Rony Brauman remarked at the end of the 1980s when talking of wars in general "the transition from open warfare to relative peace, and vice-versa, considerably modifies the needs, particularly owing to the population movements which are one of the primary consequences."Rony Brauman, President's Moral Report, 1988For MSF, then, are not the issues of an occupation situation more political in nature, as this dimension is itself at the heart of the question of safety in the field? Occupation contexts would indeed most often seem to be extremely politicised situations, from which neither the humanitarian organisations nor the media are spared. Behind the objection of the "programme by default" then lies, more or less implicitly, the accusation of the "alibi programme". Are MSF missions in occupation situations more exposed than in others to the risks of politicization and of manipulation and targeting by the occupier or the occupied?

2. Why are we really here? How are we perceived?": MSF, the occupier and the occupied

THE WEIGHT OF HISTORY: OCCUPATION AND IDEOLOGY

Like war in general, the notion of occupation has been forged through the experience of the conflicts of the 19th and 20th centuries. This is reflected both in the evolution of international humanitarian law and in the history of ideas concerning international relations. As a historical community, MSF and its members as individuals - albeit with varying degrees of awareness - are of course part of these currents of legal and philosophical thinking. The definition of the rights and duties of the occupying powers contained in the Hague Peace Conferences of 1899 and 1907 reflected the view of war prevalent at the end of the 19th century, it being an affair of States conducted by professional armies, away from the civilian population. Occupation was seen as a transitional period preceding the signing of a peace treaty, during which the occupier was required to provide a basic level of management of the territory under its control, without affecting the institutions nor the everyday economic and social life of the population. Upon signing of the peace treaty, it was understood that the occupied territory was to be returned to the vanquished State, hence the importance of preserving the status quo, or annexed to the victor State, who was now free to exercise sovereign power over it.

The experience of the First and above all the Second World War, marked by the scale of the crimes committed against the civilian populations by the occupying forces of the Axis, demanded a change in the approach to occupation law, which was reflected in the 4th Geneva Convention of 1949. As clearly shown in its title, occupation law was now seen from the viewpoint of "protection of civilians", binding upon the occupier whatever the changes in the status of the territory under its control and regardless of the political regime that existed prior to the occupation. Unlike the Hague regulations, the occupier was now given considerable responsibilities, including not only the obligation to respect the civilians, but also the obligation to provide for their essential needs in the broadest sense. Insofar as these responsibilities were met, the Convention nonetheless recognised certain rights of the occupying power, both to guarantee its own safety and in the requisition of certain public goods and structures, thereby reflecting the interests of the victorious powers of the Second World War, which occupied and oversaw the reconstruction of the Axis territories.

This view changed again in the 1970s, in the wake of decolonisation. The experience of the independence struggles, the rising weight of the former colonies in international bodies and the influence of third-world ideas contributed to condemning occupation as an unacceptable regime of oppression, in the same way as colonisation or apartheid, in the name of the right of peoples to self-determination. The principle of the rights of people to self-determination had already been included in the 1946 United Nations Charter, but had been placed on the back-burner in deference to the sovereignty of States and the strategic demands of the Cold War. It was now back at the forefront of international concerns, demanding recognition of the legitimacy of the peoples' resistance to occupation, of which the Israeli/Palestinian conflict was already the symbol. The influence of this ideological movement was in particular manifested in the 1977 Additional Protocol I to the Geneva Convention. Its first article added the following to the list of international armed conflicts subject to international humanitarian law: "armed conflicts in which peoples are fighting against colonial domination and alien occupation and against racist regimes in the exercise of their right of self-determination."Article 1 paragraph 4, Protocol of 12 August 1949 relating to the protection of victims of international armed conflicts (Protocol I) (8 June 1977). This historical recapitulation is inspired by the article by Eyal Benvenisti, "The Security Council and The Law on Occupation: Resolution 1483 on Iraq in Historical Perspective", Israel Defense Forces Law Review, n° 23, 2003; also see Françoise Bouchet-Saulnier, Dictionnaire Pratique du Droit Humanitaire, La Découverte, Paris, 2000 and Eyal Benvenisti, The International Law of Occupation, Princeton University Press, 2004

This was the context in which MSF was founded and this moral condemnation of occupation, based on the principle of the self-determination of peoples, was initially explicitly stated by its members and its leaders. Starting with the Biafran war, reference to the right of the "Ibo nation" to exist was a powerful driving force behind the solidarity of the future founders of MSF.See Anne Vallaeys, Médecins Sans Frontières, la biographie, Fayard, 2004This solidarity, seen as only natural, with an occupied people - and its desire to resist - was also at the heart of MSF's engagement in Afghanistan and in the refugee camps on the Thai border with Vietnam-occupied Cambodia during the 1980s.This attitude is particularly clearly expressed in the 1982 Moral Report. Thus, when referring to the Nong Chan refugee camp, it was stated that "several thousand refugees still live there. Several thousand resolute individuals who had decided not to return to their country as long as it was occupied." Similarly, MSF fully assumes the motivations of the volunteers who went out to Afghanistan: "All of them have accepted the danger of clandestine operations, that of isolation in the interior of the country, that of being caught up in the fighting or imprisoned. They were aware of this danger when they left […] but they agreed because they felt that this proof of solidarity with the Afghan people was necessary, I was going to say self-evident, at a time when the world is doing everything it can to forget them and sit back and do nothing."For Rony Brauman, looking back, it was these two missions that forged the attitudes of MSF and both significantly contributed to establishing its reputation and shaping its identity. They were both founded on a clear refusal to adopt a neutral stance"As with the vast majority of humanitarian organisations active in Afghanistan, MSF never sought to take a neutral stance. […] Like our forerunners in Biafra, we had implicitly picked our side. We had made our choice clear by undermining, deliberately and indirectly, the diplomatic position of our side's enemy. We all saw it as our duty to expose the scale of this war to the world, especially since it was mentioned so seldom in the media during its early years." Rony Brauman, Foreword to MSF, World in Crisis: the Politics of Survival at the End of the 20th Century, Routledge, 1997 - a neutrality further made impossible by the belligerents themselves.

The notion of occupation which prevailed during the 1970s and 1980s was not solely the fruit of ideological streams born of anti-colonialism. The creation of the Freedom Without Borders Foundation in the 1980s was in fact a conscious break with the third-worldist movement. For the founders of MSF, as for their successors, the notion of occupation doubtless evoked images of the occupations of the Second World War perhaps more than attacks on the right of peoples to self-determination. As we have seen, occupation as conceived of by the laws of war up to the end of the 19th century, was nothing other than a state of affairs common to most wars between States. During the First World War, in a context of immense patriotic fervour on the part of the vast majority of the populations of the belligerent powers, the term occupation was not felt to be strong enough to mobilise citizens and combatants to win back the land that had passed into German hands. As the historian Annette Becker writes, "throughout the war, the territories are rarely referred to as 'occupied', a de facto situation, but as 'invaded', a temporary status meant to be cancelled out by the Allies' victory."Annette Becker, Oubliés de la Grande Guerre, humanitaire et culture de guerre 1914-1918, Noesis 1998, p.33For its part, the Second World War was marked by the experience of totalitarian occupations, with the deportation and massacre of entire communities. For the historian Anne Duménil, in the wake of the international brigades who volunteered to fight in the Spanish Civil War, patriotism was no longer the sole motive for joining the resistance: "The notion of war for the rule of law was a strong motivation: engagement in the struggle against tyranny is also one of the springboards to action for the volunteers."Anne Duménil, "La Guerre au XXème Siècle -2. L'expérience des civils", La documentation photographique, n° 8043, March-April 2004One can therefore consider that the political charge of the notion of "occupation" in the eyes of the western humanitarian volunteers themselves, largely sprung at the time - and probably still springs - from this historical reference.

MSF's position on the occupation situations of the 1980s was in fact closely linked to a denunciation of totalitarianism, embodied by the Soviet Union and its Vietnamese ally. The brutality of the occupier towards civilians was seen to be an integral part of totalitarian ideology. As we saw in the first part of this study, occupation is first of all seen as a series of unacceptable abuses. It is by denouncing these practices that MSF aimed to undermine the legitimacy of the communist ideology in Afghanistan and the camps in Thailand."Our aid unapologetically involved publicly denouncing the atrocities committed by the occupier, backing the investigations into its war crimes and speaking out to official bodies […] Resistance strongholds and refugee camps gave other people's trouble a face: that of Soviet imperialism. This explains why the principles which MSF defended in the 1980s owed more to French writers Albert Camus and Raymon Aron than to those who theorised about the Third World. Our battle was to inform public opinion about the brutality of totalitarian regimes to civilians. We sought to prevent these regimes from misappropriating aid and to defend the independence of humanitarian action." Rony Brauman, Foreword to MSF, World in Crisis, the politics of survival at the end of the 20th century, Routledge, 1997Consistent with this goal of denouncing a murderous ideology was MSF's refusal in 1979 to work in the camps controlled by the Khmer Rouge, the main force of resistance to the Vietnamese occupation. Hence, from the outset, there were limits to the association's empathy with occupation resistance movements.

For a time, the collapse of the Soviet bloc seemed to change the picture. The departure of Soviet and Vietnamese occupation forces between 1988 and 1989 was accompanied by increasing disillusionment with the Afghan and Cambodian resistance movements. This led to a reconsidering of the chosen affinities of the previous decade, at a time when denunciation of the oppressive nature of communist ideology was losing its raison d'être, following the fall of the Berlin Wall. In Cambodia, withdrawal of the Vietnamese troops was followed by a resumption in fighting between the resistance factions and the Hun Sen government, leading to fears of a return to power by the Khmer Rouge. This situation led MSF alongside MDM to denounce the siphoning off of humanitarian aid on the Khmer-Thai border - although this is something that had been done out in the open by all the resistance factions for the past ten yearsFiona Terry, Condemned To Repeat : the Paradox of Humanitarian Action, Cornell University Press, 2002, chap. 4- as well as the Khmer Rouge representation at the UN. They also asked for a neutral camp to be set up by the United Nations. Despite the many internal reservations concerning de facto recognition of the pro-Vietnamese government of Hun Sen, the choice was made to open a mission in Cambodia, justified both by the war situation and by the desire to counter the Khmer Rouge.Minutes of board meeting of 1 December 1989This desire was again demonstrated by the refusal to carry out an exploratory mission into their zones in 1992.Minutes of board meeting of 28 February 1992In Afghanistan, the departure of the Soviet troops saw the question resurface of a mission to Kabul, still governed by the communist regime of Najibullah. This question led to fierce debate, with certain members of MSF considering it a betrayal to work with the government. The decision was postponed on the principle of impartiality, with the hospital structures in the government zone being considered able to meet the medical needs. Although initially raised to justify the refusal to work in the areas under Soviet occupation, this principle nonetheless proved increasingly hard to support internally in the light of the end of the occupation"This decision, taken jointly with all MSF sections, along with MDM and the AMI, after frequent and often fierce debates, was unsatisfactory with respect to our principles and was considered by all to be provisional and to be regularly questioned in the light of the information reaching us from the government zones." Rony Brauman, Moral Report 1989and the new face shown by the leaders of the "resistance"."A certain number of people believed that the departure of the Soviets would mean the end of the Najibullah regime. In fact, this withdrawal revealed the fact that this war in Afghanistan is above all a civil war, which had taken the form of a war against the invader but which, since the Soviet troops have gone, have gone back to being a civil war. Six months after the Russians left, the Najibullah regime is getting stronger. It has become the sole pillar of stability, the only person with whom one can talk. The resistance is fragmented and increasingly resembling the Lebanese model, with the resistance being manipulated by Pakistan into a dead-end. Everything is being confused, the resistance, the fundamentalists, the bandits, the Iranians, etc. …" Board meeting of 15 September 1989This refusal by some to "betray" was compounded by the fear of reprisals from certain Mujahideen armed groups. Although mentioned on several occasions, the dispatch of a mission to Kabul was repeatedly postponed. It only finally took place in 1992, after the capture of Kabul by the Afghan parties of the former resistance and the fall of Najibullah, even though MSF had pulled out of the Mujahideen zones in May 1990 after the murder of an expatriate in Badakhshan.

The end of the occupations by Communist regimes at the end of the 1980s broke the link between occupation and totalitarianism. Other forms of oppressive regimes and ideologies with totalitarian intent triggered fresh debate at MSF, at a time when reference to human rights and the democratic ideal, closely tied in with the notion of the rights of people to self-determination, appeared to remain unquestioned. Already in the mid-80s, in Hazarajat, a central Afghanistan area with a Shiite majority, the influence of Iranian fundamentalism on the Hazara chiefs and local political rivalries had closed the door to us in this region. One year later, we had to temporarily suspend our mission in Afghanistan departing from Pakistan, faced with the refusal by the resistance leaders to escort expatriate women. MSF had then refused to give way to a measure which would have meant that the teams could only have access to men, in particular to combatants. This further manifestation of religious fundamentalism, allied with an increasingly widespread attitude of defiance and aggressiveness towards westerners, was already eating away at our ability to assist and show our solidarity with the "Afghan people" as victims of the Soviet invasion."This attitude reflects a state of mind that is particularly worrying for the future. The presence of western doctors and journalists would seem to be interpreted in certain regions as offensive to Islam and its fundamental values and thus increasingly unacceptable. If this state of mind gains ground - and there is every indication that some countries in the region are actively doing everything they can to ensure that it will - then it is to be feared that Afghanistan will close itself off completely from the outside world. One can only imagine what the human rights situation would be in this country, sealed off and in the hands of the fanatics. What does appear almost certain now is that the same type of mission as before will be impossible. We will have to adapt or give up." Rony Brauman, Moral Report 1986This fundamentalism was to reach its culmination in the Taliban regime and the reference to totalitarianism was then explicitly mentioned in MSF's assessment of this regime. This reference was also to contribute to the dilemma in terms of attitude and operations during the continution of the conflict between the American army, officially working to reconstruct the country under the control of the new Afghan government, and the Taliban guerrilla."The nature of the Taliban regime was clearly a totalitarian one, even if they did not have the material resources or the apparatus of State to implement their doctrine to the extent they would have wished. The excesses of this regime had led us to reduce humanitarian activities in this country in recent years and we were at an impasse. How, in these conditions, could we not celebrate its passing?" Jean-Hervé Bradol, Moral Report 2002

THE ADVERSARIES OF HUMANITARIAN ACTION

As of the early 90s, however, the invasion of Kuwait by Iraq and the prospect of the first Gulf War raised in a new light the question of the position to be adopted by MSF. The risk of manipulation by the Baghdad regime, felt to be totalitarian, was counter-balanced by the fear of adhering by default to the invocation of international law and democratic values as the basis for the international intervention against Iraq. Throughout the Gulf crisis, and more so after deployment of the coalition forces into Iraqi Kurdistan in the name of humanitarian action, this dilemma obliged MSF to seek an always precarious middle ground in defining its positions and its operations. This question of the possible perceptions and interpretations of the positions it then adopted was in fact ever-present. Thus, after an initial position of principle opposed to intervention in Iraq,"In the light of the nature of the Iraqi regime, the humanitarian issues at this stage in the crisis and the clear risk of political manipulation of any humanitarian mission on its arrival in Baghdad, the Board declares itself against the principle of intervention in Iraq." Board meeting of 14 September 1990 the beginning of the conflict shifted the emphasis to the humanitarian principle of assistance to the victims of the war. To this was added the need to distance ourselves from France's position in the conflict,"This is the opportunity to tell people that MSF is not France. France's involvement complicates the problem, but we must clarify matters." Presentation by Rony Brauman at the Board meeting of 25 January 1991. leading to a mission being dispatched to Baghdad as soon as Iraqi visas were obtained.

The regime's repression of Shiite and Kurdish insurrection movements and the Kurdish exodus to the mountains on the Turkish border then posed the problem of assistance to and protection of the Kurdish populations. Considering the scale of the crisis, MSF felt that involvement of the coalition forces and the United Nations was necessary. The organisation was however already aware of the risk of "providing an alibi for a deployment of forces which, albeit slightly humanitarian in nature, is also political and strategic." Faced with the deployment of the American army, MSF asked for UN support and supervision both of the assistance programme and the return of the Kurdish refugees. MSF nonetheless refused to place itself under their coordination as their political role in the conflict and the negotiations with Baghdad ran contrary to the declared aim of neutrality.See the minutes of the Board meeting of April 1991. Uncertainty over the position to be adopted given the evolution of the crisis is clearly expressed: "In reality, in the current situation and given the absence of political data, MSF is obliged to improvise. MSF is in Iraq under its own name. The question arises with respect to the various approaches made to it by the allies and by the United Nations. For the time being, the response has been: we are here, but we're aside from them." MSF's problem with the role played by the United Nations in the Iraqi crisis was to grow and be emphasised by Rony Brauman in the 1992 Moral Report: "It is in the light of negative aspects that one must understand MSF's refusal to place itself under the UN banner, particularly in Iraq, where UN involvement both in the conflict and in its aftermath is evidently highly problematical."Even though the mission was rapidly restricted to operations in Kurdistan in the absence of authorisation to work with the Shiite populations, MSF retained its coordination office in Baghdad. It nonetheless justified maintaining its presence in Kurdistan, after the emergency phase, by the risk of reprisals from the Iraqi regime after the announced departure by the international troops. MSF's balancing act was to be finally terminated with the closure of the Baghdad mission in June 1992, as a result of the rising tensions with the Iraqi regime and its deliberate targeting of NGOs in the Kurdish areas in the following months. The increasing defiance of MSF head office with respect to the Kurdish parties and the teams' dependence on protection by the Peshmerga combatants was however not totally unrelated to the decision to pull out.

With the episode in Kurdistan, the question of neutrality - challenged in Cambodia and Afghanistan, once again arised for MSF - or rather was imposed on it. For the first time, MSF had to adopt a stance against a belligerent and then, in Iraqi Kurdistan, an occupier that looked very similar to itself and invoked the very principles MSF had referred to in its denunciation of totalitarian occupiers and their practices. To avoid subscribing to the "right of interference" and distance itself from the military deployment by the coalition forces, MSF made the gesture - which was of necessity political in this context - of recognising the sovereignty of Baghdad in Kurdistan by insisting that its teams working in the Kurdish zone obtain Iraqi visas. This in no way prevented these teams from experiencing a very real empathy toward the Kurdish cause against the oppressive Iraqi regime. From the standpoint of Baghdad, MSF was only granted authorisation because of the pressure of events, particularly its military defeat. The obstacles placed in MSF's way in the south, followed by the killing of NGO members, quickly shattered the illusion of a position of neutrality in this crisis. More than the presence of the coalition forces, whose appropriation of the humanitarian slogan MSF contested, while nonetheless recognising the utility of their intervention, it was the multiple roles of the United Nations which posed the problem. Welcomed by MSF in Cambodia and initially in Somalia, the UN in Iraq confused approval of the armed intervention with coordination of the humanitarian assistance that its agencies were negotiating with Baghdad. Although MSF distanced itself from the UN in the light of the political nature of its role as negotiator in this context, it did not in this particular case assimilate it with the international armed forces. Liberia and Somalia were to be a decisive turning point in MSF's search for independence in a world in which the UN was once again bringing its weight to bear.

In Liberia as in Somalia, the political role of the United Nations convinced MSF of the need to adopt a confrontational attitude. In the name of peacekeeping, the ECOMOG regional interposition force, dominated by the Nigerian army, set itself the priority military goal of eliminating one of the protagonists in the Liberian conflict, the NPFL of Charles Taylor. Routing of humanitarian aid into the zones under the control of this rebel movement was felt to be contrary to this goal, in that the aid could help strengthen Taylor, both in terms of material resources and legitimacy among the civilian populations. ECOMOG therefore attempted to ban NGO access to these zones, and in April 1993, it even attacked an MSF convoy. In Somalia, the international contingents, primarily consisting of American Marines, landed in December 1992 with a humanitarian mandate, which included securing Mogadishu and dispatching food aid in response to the famine which had been raging for several months. This mandate was however very quickly reinterpreted as an objective of stabilising the "failed state" that Somalia had become. Here again, this goal led to the attempted elimination in June 1993 of the leading Somalian warlord, General Aidid, and his forces. This military and political goal led to numerous excesses by the foreign troops, including attacks on civilian demonstrations and the bombing of an identified ACF building, housing an MSF team.

As we have seen, these incidents were not the only, nor the main reasons for MSF's temporary withdrawal from these two countries.In the case of Somalia, the bombing of the ACF building happened several months after the decision to gradually withdraw had been taken.Central, however to the perception of these excesses by decisions-makers in MSF was the UN's refusal - in both cases - to distance themselves from the international forces they had mandated or to express any condemnation of these practices. Quite on the contrary, the Secretary General's Special Envoy to Liberia, Trevor Gordon Summer, fully endorsed ECOMOG's policy by declaring that "certain organisations have a mandate to provide assistance to populations in need. We have a more important mandate: to bring peace. If assistance stands in the way of the peace process, there will be no assistance".Quoted in Fabrice Weissman, "L'aide humanitaire dans la dynamique du conflict libérien", MSF Internal report, May 1996, p.61Similarly, his counterpart in Somalia discounted the complaint filed by MSF denouncing failure of the international forces to abide by the Geneva Conventions. In both cases, a UN diplomat who had expressed criticisms of the postures adopted by his organisation was relieved of his functions by the Secretary General. The desire to challenge the limits placed upon its actions and the political message aiming to justify them is certainly not unrelated to the increasing numbers of missions dispatched by MSF to the NPFL zones - and a few years later, to the zones controlled by the RUF in Sierra Leone. If the lack of aid to the populations trapped in these areas was the primary motivation for these initiatives, then UN and ECOMOG intransigence probably partly explains the determination of the teams to try to work there, despite the violence and the predatory practices of the rebel movement. It was also doubtless part of the reason that MSF undertook new exploratory missions in Somalia, shortly after pulling out. Many in MSF certainly wanted to avoid that this withdrawal be used as proof that humanitarian aid could only be dependent on an offensive military operation. Until one of its expatriates was killed in 1997, MSF thus attempted to redefine its missions in such a way as to dissociate its image from that left behind with the Somalis by the military-humanitarian episode of 1992-93. MSF's return to Kosovo before the deployment of the KFOR must doubtless also be seen in the light of these experiences.

Public condemnation and operational defiance: MSF's adaptation to the new posture adopted by the UN clearly recalls that developed in the face of the Soviet enemy in Afghanistan. Indeed in the same way, the MSF teams had to deal with "a deliberate will to prevent them from providing assistance", were banned from providing the "material and moral support" that aid could represent and, with respect to the criticisms addressed to the UN concerning its forces, they were seen as "troublesome witnesses"."The future of the French doctors is above all hindered by a deliberate will to prevent them providing assistance. Those who invaded Afghanistan cannot tolerate our presence, both because it provides material and moral support to the population and because we are troublesome witnesses." Moral Report 1983This similarity owes nothing to chance, because even if we can say that a humanitarian agency as a third part in a conflict has no enemies, MSF's experience clearly proves that we do have adversaries. They share a common desire to impose a political vision in which humanitarian law and principles are seen at best as a means to an end, which can be discarded when they no longer serve the ends - the values and ideals - heralded by this political vision. Hence, from the viewpoint of humanitarian action as conceived of by MSF, between communist totalitarianism, the UN vision and more recently the war against terror promoted by the United States, there is of course a difference of degree, but not of nature. According to their respective political agendas, independent humanitarian action is a disposable item. To use the terms of Rony Brauman to characterise the stakes for MSF of the communist occupations of the 1980s, the aim was always to denounce the "atrocities of the occupier" - or at least its "war crimes", "abuses" and "excesses" - thereby "indirectly and consciously" contributing to "weakening its diplomatic position", in other words undermining the foundations of these new ideologies.see notes 33 and 34Thus in the mid-90s, MSF was forced to redefine its identity by abandoning references to democratic values and the defence of human rights, which had become the triumphant values of the post-Cold War period and which henceforth provided the UN and the western powers with the justification they needed for occasionally stifling an independent humanitarian space. As with communism, the need once again was to "inform public opinion", to "expose to the world" the reality hidden behind the speeches, this time by denouncing the "western fantasy" contained in the notion of the right of interference"[In Somalia] under cover of the banner of solidarity, human rights, humanitarian aid, we saw military helicopters attack demonstrations […] This mockery of intervention, or perhaps what was enshrined -this is in any case my opinion - in the very notion of military-humanitarian intervention, perfectly illustrates a western fantasy which we saw beginning to take shape at the time of the Gulf War. It is that of embodying divine providence, and we know that the primary attribute of that is the power to give or protect life as easily as taking it away." Rony Brauman, Moral Report 1994and by rejecting the "megalomaniacal vision claiming to wage a universal war for justice and democracy" and the "UN's vision of well-being shared by all"."MSF is not an organisation devoted to emancipation of the people. Its action is somewhere upstream of that. One must beware of the megalomaniacal vision which aims to wage a universal struggle for justice and democracy and of the UN's vision of well-being shared by all. We would then be missing the point because this is not our mission and not what endangered populations expect of us." Philippe Biberson, Moral Report 1996Indeed, the role of the United Nations in what came to be called "complex emergencies" was chosen as the main topic of the 1993 edition of the MSF collective work, Populations en Danger, a proof that this concern was crucial to the leaders of the association.Concerning the internal debates around this choice, see the minutes of the Board meetings of 5 February, 5 March and 26 March 1993.

To be sure, occupations under the aegis of the UN are not the only and certainly not the worst occupation situations MSF has had to deal with since the beginning of the 90s. The "total war" as we called the Chechen conflict and that in the DRC, particularly during the period of occupation by the Rwandan and Ugandan forces, went beyond simple "excesses" and were the scene of massive and systematic abuses, forming an integral part of a policy of terror. Furthermore, in this type of context, humanitarian workers are frequently the direct targets of violence in line with that carried out against the civilians, thus preventing them from gaining access to the victims. In the Palestinian Territories, the policy of the Israeli occupiers, particularly since 2000, clearly shows the aim of having Palestinian society as a whole pay the price for the terrorist attacks committed on Israeli soil. There is thus no shortage of situations in which the forms of oppression and the identity of the oppressor trigger far stronger and sometimes partisan feelings of solidarity and empathy with the occupied populations on the part of MSFSee for example, the first evaluation report by Christian Lachal in the Palestinian territories after the resumption of fighting: "Daily life in the territories is hell. The decision to go and work in hell is the right one." Christian Lachal, "Gazaouites et Hébronites" Emergency programme evaluation mission in the Palestinian Territories, January 2001than in the case of occupation by international forces with a UN mandate. Nonetheless, whether undertaken by authoritarian regimes or western democracies, and whether or not approved by the UN, armed interventions followed by occupation share the three characteristics attributed by Jean-Hervé Bradol to the emerging concept of the war on terror promoted by Washington: "the lack of any clear definition of the enemy, demonisation of this enemy and a massive imbalance between the forces present".Jean-Hervé Bradol, Moral Report 2002

Lack of any clear definition of the enemy: this is the source of the lack of distinction, whether deliberate or incidental, between combatants and civilian populations. Demonisation: whether targeting Palestinian, Chechen or more generally Islamist "terrorists", the members of an ethnic group, or "warmongers" from Taylor to Aidid, it helps build up the picture of the "enemy to be conquered", justifying the use of force and frequently justifying in advance the human cost that will be entailed. Finally, imbalance between the forces present: this is the most commonly accepted image of an occupation situation opposing a regular State or multi-State army against resistance guerrilla groups. MSF's experience since the beginning of the 90s is that these characteristics, which can be applied to the vast majority of occupation situations, are likely to provide legitimacy for a "disproportionate, in other words excessive, use of force in military operations."Ibidem

THE QUESTION OF PERCEPTIONS: ISSUES AND ILLUSIONS

The lack of any clear definition of the enemy and its demonisation that leads to a blurring of the lines between civilians and combatants, is not the sole preserve of the occupier. In the process of radicalisation that develops over time in an occupation situation, the opposition groups generally develop a discourse that is symmetrical to that held by the occupying power to defend the legitimacy and justice of its cause. This symmetry was underlined by the President of MSF when he spoke of the face-to-face situation between "the eternal victim and the victim of the eternal victim" in the Israelo-Palestinian conflict.Jean-Hervé Bradol, "Put to the Test", Palestinian Chronicles, July 2002Superiority, not only military, but also sometimes economic and political, is however the specific characteristic of the occupying power. It is this superiority that pushes MSF to distance itself from the latter -although now, at least at an institutional level, it refrains from approving movements of opposition to the occupier.

First of all, the civilian populations of the occupier are usually remote from the conflict and therefore comprise fewer victims. Furthermore, the medical infrastructures of the occupying power are usually intact and thus able to deal with these victims. In the name of impartiality, in other words on the basis of needs alone, this general state of affairs justifies MSF intervention primarily or even exclusively on behalf of the "occupied" civilians. In addition, the occupier's control of all or part of the territory implies that MSF usually has to negotiate access to the victims with the occupier. Consequently, the teams directly come up against the occupying power's own political and military interests and in many cases are dealt with arbitrarily. Finally, a determining factor in the case of the United States' war on terror, or the interventions by the UN, is that the economic power of Washington and the "international community" through the UN gives these specific occupiers the ability to finance humanitarian aid and reconstruction under their own supervision. This enables them to further legitimise their policies which find their source in an essentially western, dominant vision of peace, stability and democracy. MSF can be easily assimilated with this vision, not only by western public opinion, and therefore its main donors, but also to a greater extent by the assisted populations and armed groups fighting the occupation, whose identity exists through opposition to the discourse and values proclaimed by the occupier. MSF therefore has all the more difficulty in reaffirming its independence when faced with occupiers who resemble it.

The recurrence of security problems in occupied areas bears witness to the extreme difficulty in defining a clear position in this type of context. This difficulty lies as much in the little control we have over the mutual perceptions of the various players in these situations as in the ambivalence contained in our own approach. From the outset, MSF had to deal with a problem of legibility, even within the western world, concerning its criticism of military interventions conducted under the humanitarian banner. In Somalia in 1992, MSF had generally drawn the attention of the media to the victims of the famine and the direct link between this famine and the civil war in progress. However, its message over the military-humanitarian confusion when the international troops landed in December of the same year was given little coverage. This public position was felt to be too complex and too much in contradiction with the calls for mobilisation evoked by the pictures of starving children.Virginie Raisson, Serge Manoncourt, "MSF-France en Somalie : janvier 1991 - mai 1993", internal evaluation report, February 1994Even today, one frequently hears French journalists stating that MSF were the creators of the concept of the right of interference. This confusion is all the more comprehensible among the armed groups as well as the civilian populations of the countries in crisis, who generally witness an exponential rise in the numbers of humanitarian players in the wake of the deployment of international armed forces. Moreover, the rise of anti-western sentiments cannot be explained by the excesses of the international armies alone. There are other reasons, in particular linked to the apparent wealth of the NGOs. This wealth can be perceived as an attribute of the political and military power of the western states, or as the symbol of a traditional urban dominance over rural areas, as revealed by Fabrice Weissman with respect to the attitude of the Liberian fighters to humanitarian aid.Fabrice Weissman, "L'aide humanitaire dans la dynamique du conflict libérien", op. cit., p. 55

MSF is rarely perceived in just one way, reflecting the divisions in the societies subject to occupation, divisions which already existed but which evolve along with it. For example, in Afghanistan, during the Soviet occupation, certain groups saw the expatriate teams as natural allies against the invader, all the more so as they often saw in us vectors of political legitimacy and material support from the western countries, a belief that was strengthened by Pakistan's coordination of the Afghan resistance. However, for other groups such as the Hezb-Islami of Gulbuddin Hekmatyar, the representation of the Soviet enemy as an infidel power implied a sometimes violent rejection of the western organisations perceived as Christian or sinful. Conversely, in 2003, the virulent criticism by the NGOs of deployment of the Provincial Reconstruction TeamsAmerican reserve units usually supervised by special forces members, responsible for contributing to stabilising Afghanistan by carrying out assistance and reconstruction work and acting as a security deterrent. Officially created to shore up the legitimacy of the new Afghan regime among the rural populations, the PRT were also there to improve the image of the American army and provide intelligence in its war against the insurrection movements linked with the Taliban.to Bamyan no doubt appeared incomprehensible to the minority Hazara population, who welcomed with open arms the protection and the manna promised by the American army, whose bombings had chased out the Taliban.

Our insistence on being seen for who we are - or who we think we are - often comes up against cultural or contextual barriers that are hard to overcome. In societies in which allegiances and loyalties are essential preconditions for survival in times of crisis, our principle of independence is not necessarily self-evident. Nor do we insist on this principle quite so strongly when our assumed dependence on the French state becomes the guarantee of a degree of safety, linked for instance to the popularity in the Muslim world of the French president's opposition to the war in Iraq. Nor is the image of neutrality particularly respected in an occupation context where the political identities of the parties to the conflict are defined entirely by attempts to destroy the legitimacy of the enemy. The sometimes fierce internal debates at MSF concerning the reaffirmation of our neutrality in the Israeli-Palestinian conflict is to a large extent meaningless from the viewpoint of the vast majority of the Palestinian population and authorities, for whom there is obviously not a shadow of a doubt concerning our commitment to their cause.According to a field manager in Gaza, many Palestinians equate the MSF expatriates with the pacifist militants, including American ones, who place themselves between the Palestinian civilians and the Israeli army. One of them, killed by an Israeli bulldozer, is revered as a martyr by the Palestinians in Rafah.The aggressiveness of settlers and some of the Israeli soldiers towards us confirms the fact that this is clear to the protagonists in the conflict at field level. In return, the stated claim by the Israeli chiefs of staff liaison department of wishing to coordinate NGO access to the Palestinian territories is designed to achieve a number of political goals: to exercise control over the activities of the volunteers, to show the Palestinians the clear disparity in treatment between themselves and the NGOs and to help weaken the credibility of the Palestinian Authority."The 'coordinations' put in place with the army to enable us to work in the Gaza Strip or the West Bank, are seen by the Israelis as the beginnings of 'collaboration' as witnessed by their charm offensive when we meet them, which are in stark contrast to events in the field, which demonstrate aggressiveness towards the NGOs on the part of the military." Christian Lachal, "Desolation Row", exploratory mission 30/06-07/07/02 in Jenin (ex-Palestinian territories)In Afghanistan, the quasi-confrontational position of MSF - and many other NGOs - adopted with respect to the American army, and based on the military-humanitarian confusion embodied by the PRT, was unable to convince the Taliban and their allies, if not of our neutrality then at least of their political interest in recognising us as being outside the conflict. The criticisms made by the western humanitarian NGOs were in fact only expressed to the American army itself and to the representatives of the new Afghan government, who depended on the Americans for their status and indeed their survival. They carried little weight given that our programmes were limited for security reasons to the zones controlled by the very regime whose legitimacy was disputed by the Taliban.This contradiction was once again tragically evident in the motive for our departure after the murder of five members of MSF, invoking the inability and the unwillingness of the Karzai government to identify and punish the guilty parties. This criticism could only service to bolster the Taliban in their claim of responsibility for the attack, which was part of their strategy of contesting the order imposed by the"American occupier" by maintaining a climate of insecurity.

Images of domination, allegiances, partisanship: our interventions can be politically interpreted in a multitude of ways that our public stance can only rarely reverse, or even modify, however slightly. Therefore, the problem we have in defining our programmes in a time of occupation owing to the chronic and low-intensity nature of these conflict situations, as shown in the first part of this study, is further aggravated by the problem of ensuring that our action of assistance and solidarity can be clearly understood by the parties to the conflict and indeed ourselves, in a context where there is little place for neutrality. As shown by the experience of decolonisation, wars of resistance against an occupation are won or lost on the political rather than the military battlefield. Humanitarian aid must therefore find a place for itself in areas of disputed legitimacy. This explains why the position of the occupier usually consists either in barring access to the humanitarian organisations, those "troublesome witnesses", and imposing its order behind closed doors, or in appropriating the humanitarian project as its own, legitimising its occupation by aid that is planned and supervised by itself. These two trends can also be found alternately in the same context, as shown by the Israeli occupation or the American occupations in Afghanistan and above all in Iraq. Faced with this dual risk, MSF's "know-how", in other words its ability to urgently deploy large amounts of logistical and medical resources to the largest possible number of victims proves to be ill-adapted. Philippe Biberson, in his 1995 Moral Report, thus partly ascribed MSF's failure to intervene during the first year of the Chechen conflict to the fact that our resources were not suited to the context. He stressed that "what is an advantage in many situations becomes a straitjacket paralysing us and making us vulnerable because we are too visible and have too much to lose.""In Chechnya we were faced with major difficulties linked to the little room available for humanitarian action. However it would seem that our approach also suffered from the somewhat mechanical replication of operational stereotypes. Full Charters, white Toyotas covered with stickers and fluttering flags could do nothing in the face of the impenetrable cynicism of the Russian forces, and the VHF radios did nothing to reduce the level of insecurity or even the feeling of insecurity linked to this total war situation.[…] What is an advantage in many situations becomes a straitjacket paralysing us and making us vulnerable because we are too visible and have too much to lose." Philippe Biberson, Moral Report 1995It was for these same reasons that in 1999, pending a return to Kosovo under the auspices of a new military-humanitarian operation, he called for "highly symbolic and peripheral actions, in other words where the others don't go, to help in that which others have left behind or neglected.""Regardless of the developments in this crisis, the humanitarians will have to accompany the process. We know that humanitarian aid on this scale, international humanitarian aid and inevitably military-humanitarian aid will leave little room for a fine discrimination between the needs of the victims and even less so between the actors of the conflict. We will have to take care to ensure that humanitarian aid does not in turn become a source of injustice and discrimination, an instrument of domination and control. For Médecins Sans Frontières, my preference, as we defined it a few years ago, lies with highly symbolic and peripheral actions, in other words where the others don't go, to help in that which others have left behind or neglected." Philippe Biberson Moral Report 1999

In these areas where legitimacy is disputed, the position of the political representatives of the opposition to the occupation is often the reverse of that of the occupying power: even if they ask for or agree to witnesses and their testimony, they generally contest any aid that can legitimise the position of the occupier. From this viewpoint also, large-scale programmes such as taking over public health care structures are often perceived negatively. Since the Oslo process began in 1993, the issue for the Palestinian Authority has been to stake a claim for itself as the legitimate government of a viable State, a position that Israeli policy since the Netanyahu government, has striven to undermine. From 1994 to 2000, MSF had refused to take the place of the Palestinian health ministry and had simply conducted mental health programmes in partnership with local associations. The resumption of the conflict, marked by a radicalisation of the move to undermine the legitimacy of the Palestinian Authority on the part of Israel, once again forced MSF to ask itself questions about medical or even surgical care in this war context, but any proposals along these lines were rejected by the Palestinian Authorities. The evaluation reports by Christian Lachal highlighted this dual requirement from the Palestinians, which was the presence of witnesses on the one hand,"People feel a very strong need for us to support them and be with them in their ordeal, as well as care for them and finally bear witness to what they are going through." Christian Lachal, "Gazaouites et Hébronites" Emergency programme evaluation mission in the Palestinian Territories, January 2001but preservation of the legitimacy of their Public Health system on the other."We must understand this very real fear on the part of the Palestinians, that everything they have built, particularly since 1993, is escaping from them and will be crushed, bulldozed or handed over to the international NGOs." Christian Lachal, "Desolation Row", Exploratory mission 30/06-07/07/02 in Jenin (ex-Palestinian territories)

It is within this same framework that we ask ourselves the question of the legibility of our actions: implementing atypical programmes, out of step with our usual conception of care, raises all the more questions within the association when it appears justified by the political demands of the "occupied". The promoters of these programmes are thus called on to defend them in terms of both relevance and neutrality. This seems paradoxical as the relevance of these programmes lies precisely in the renunciation of an illusory neutrality, while positioning aid on the margins of the most fiercely contested areas of legitimacy. Clearly however, the identification and more still the success of such programmes is anything but self-evident. First of all, because the symbolic value must be matched by the reality of the needs. Then, because although the media generally report on occupation situation in fixed terms, the changing positions of the parties involved, the fluctuating security conditions and needs, demands considerable reactivity in these changing working spaces. Nonetheless, the necessary gamble involved in setting up these operational innovations has all the more chance of failing when it comes up against internal resistance. This was stated by Rony Brauman as early as 1994, when the MSF mission opened in the Occupied Territories."It must however be admitted that -particularly here in Paris - MSF is both mentally and materially poorly equipped to support these missions. Our centralisation, our tradition of logistical resources, crisis and emergency interventions, makes us perhaps somewhat obtuse and in any case relatively impervious to the highly specific problems involved in this type of mission." Rony Brauman, Moral Report 1994The lack of any consensus on the appropriateness of such missions leads in most cases to a corresponding discomfort in our public stance, with our denunciatory postures revealing the same disagreement over the analysis of the situation and our ability to convert it into actions. Thus in occupation situations, it is not only our independence and our "humanitarian space" that are put to the test, but also our ability to analyse, adapt and invent.

Conclusion

This study attempted to identify the problems posed for humanitarian action in occupation situations. Focused on the experience of MSF, it shows how our collective identity, in particular with respect to MSF's relationship with conflict, the conception of its independence and the definition of its programmes, influence our way of looking at these situations and our attempts to find our position within them. Clearly many of these problems are not specific to occupation contexts. Violence against civilian populations, evidently, but also demonisation of the enemy or the existence of periods of relative normalisation during long-term conflicts, periods during which our programmes evolve, sometimes imperceptibly, from an emergency to a long term approach, characterised a number of the areas in which we have worked, despite the absence of any external armed intervention. Similarly, the humanitarian blockade sanctioned by the UN around the zones controlled by the NPFL in Liberia and by the RUF in Sierra Leone also affected the zones controlled by UNITA in Angola, as of 1998, without any international military intervention. For its part, the Darfur crisis has involved a guerrilla war waged by a rebel movement disputing the legitimacy of the Sudanese state and the disproportionate, indiscriminate and asymmetrical response by the state - even though political independence was not, at least initially, the main issue in the war. Finally, misunderstanding regarding the goal of our actions or negative perceptions of MSF or simply of its expatriate personnel are not limited to us being associated with an occupier or an occupied, including in situations of occupation.

This observation leads us to make several remarks. The first is that our identity, our organisational mechanisms and our internal limits help create similar dilemmas for us in different situations. The questions raised by occupation situations have their place in more general internal debates within MSF, particularly those concerning the validity of the operational distinction between programmes referred to as "conflict"/"post-conflict" or "exclusion"/"social violence" and those concerning the management of emergencies in the operations department. Secondly, the notion of occupation, as indeed the idea of a clear distinction between war and peace, makes reference to a conception of war as conflict between States or between a State and a people aiming to create its own State. This concept is the dominant one in international law as it is in the approach to crises by the UN and the western powers. According to Anthony Lang, it is the very reason for the drift in and often the failure of state-sponsored "humanitarian" operations: "The lesson to be drawn from humanitarian intervention [conducted by states] is that humanitarian concerns do not exclusively, or even primarily, focus on individual persons, but, in an intervention, on the creation or protection of state agents. This means that while an intervening state may be able to provide some food for starving peoples, it will soon become more concerned with creating a state entity, usually in its own image."

It is because the community or the groups who undergo this intervention reject this image and create their own political identity in reaction to it, that they also end up rejecting the aid brought to them.Anthony Lang, Agency and Ethics: The Politics of Military Intervention, State University of New York Press, 2002, p.199Although MSF has been involved in a rising number of civil wars since the early 90s, we are culturally heavily influenced by the role of the State in war, in the face of which we maintain an ambivalent attitude. As a non-governmental player, it is first of all by opposition to the State that we affirm our independence and denounce the hijacking of the humanitarian cause. We also call into question the State's legitimacy, intentionally or not, by taking its place in providing care to the civilian populations. However, we do call on States to assume their responsibilities, either in protecting their own populations and the humanitarian personnel, or in exerting pressure to halt a conflict, or intervene politically and materially in the case of a large-scale crisis. In an occupation situation as in a civil war, we tend to lean towards the rebel groups rather than the State -be it occupier or sovereign, but are often faced with the problem, raised by Rony Brauman when talking of Southern Sudan, that it is "sometimes harder to deal with a guerrilla movement than a government. It is harder to exert pressure, to invoke principles with an authority that cares little for its relations with the outside world, whereas a government generally has more concerns of a diplomatic nature."Rony Brauman, Moral Report 1993These problems are probably further compounded by the influence of our conception of war as state-centred. Does this perspective not lead us to evaluate in too simple terms the perceptions that warring parties and civilian populations hold of us : are we seen as neutral or not? Are we seen as independent or not? The position of MSF with regard to the State, therefore makes the question of occupation part of a broader question which it would no doubt be worth examining in greater detail.