1November 2003 saw the creation of an international coalition of NGOs in The Hague to condemn the devastating humanitarian effects caused by cluster munitions. Six years after the ban on anti-personnel mines under the impetus of the International Campaign to Ban Landmines (ICBL), the Cluster Munition Coalition (CMC) brought together the same NGOs: Handicap International, Mines Action, Human Rights Watch, The Norwegian People’s Aid, etc. Working within the continuity of the ICBL campaign, CMC also took up the main arguments of that campaign, emphasising violations of international humanitarian law caused by the use of these weapons, to demand that they be banned. Although this humanitarian interpretation of the problem of cluster munitions  is rarely challenged, the solution of a ban on these weapons as proposed by the NGOs did however crystallise debates within the UN Conference on disarmament. Up to 2006, the question was frequently raised, but never formally included in the Conference agenda. It was the launch of a “parallel process”, outside of the UN framework, which triggered international negotiations on this question. A Norwegian initiative, this diplomatic process was modelled on the Ottawa Process, which led, in 1997, to the signature of the Anti-Personnel Mine Ban Convention. Initially supported by a group of countries, which were neither users nor producers of these weapons, the process was gradually joined by a wider group of countries, which did not however support a ban in the first instance. France was one of these latter countries. Taking this development as a starting point, the aim of this article is to analyse the role played by reference to principles of international humanitarian law: how, and under what conditions, can the recourse to humanitarian arguments trigger such a change in position?
2The case of French engagement in the process of a ban on cluster munitions offers in effect a rich field of enquiry for the study of the mechanisms by which humanitarian rhetoric can compel international action and political decision-making, even when these seem to be guided by strategic and military objectives. Many times during this research, NGO activists, diplomats, and French military officials put forward the view that France, a country where human rights is a founding value, could not avoid committing to a treaty for humanitarian disarmament. This interpretation of the phenomenon serves to support the concept of rhetorical entrapment:  when called upon to honour their existing international engagements and their image of a virtuous country on the international stage, French public authorities had no choice but to conform to the expectations of civil society. Indeed, it is the weight of these international reputational stakes, and, concomitantly, the effectiveness of naming and shaming strategies which will form the focus of this article, with particular attention on the practical issues arising from ethical behaviour, to show what a government can gain, or at least not lose, by attributing a “moral” approach to its actions.
3To understand the mechanisms of this rhetorical coercion, this article studies the way in which institutional actors (re)positioned themselves within international negotiations on cluster munitions, as they reacted to the pressures of international mobilisation. The approach used is similar to that of Joshua William Busby, who analyses the conditions for success of mobilisations with a moral dimension by studying the targets of these mobilisations.  By shifting focus from the advocates to their targets, in particular towards the work of gatekeepers or institutional veto players, he aims to locate the necessary conditions for the success of mobilisations not in framing strategies or in the work of international activists, but in the reaction of the targets of mobilisation, that is, in the political and institutional logic which shapes a state’s decision-making. Focusing on the actors in the process of decision-making, this approach invites us to consider the question of the role of policy gatekeepers, as a determining variable in the sensitivity of states to advocacy and to constraint.  By calling attention to the work of actors in the decision-making process to explain the engagement of France in the process to ban cluster munitions, proceeding in this way opens up approaches, linking international relations literature on the role of norms and values with a political sociology literature of public action, paying attention to the role of civil servants, political leaders, and activists in the development of foreign policy decision processes. Building on a body of international relations research which considers the mechanisms and the conditions within which norms and moral values influence state decision-making on the international stage, this article shows how the rhetorical trap works because the actors within a political-administrative system allow the trap to close. 
4The purpose of the article is therefore to account for the reasons which lead these actors to “allow themselves to be caught in the trap” of their normative engagements. It thus highlights the dynamic tightening and shrinking of the scope for possible actions, by focusing on both the constraints which influence the decision-making process and on the resources which conformity to norms and moral values can offer to those who shoulder this responsibility. Although this approach means that the ambivalences of moral engagements on the international stage can be identified – and, thus, so too can the relative effectiveness of naming and shaming strategies  – the aim here is not so much to weigh these strategies against the advances obtained, but rather to consider the way in which ambivalence of action under rhetorical coercion may represent one of the reasons for the success of mobilisations with a moral dimension.
5This study takes as its starting point research conducted between January 2008 and January 2010 in the context of a doctoral thesis,  which traces the development of the campaign for a ban on cluster munitions via participant observation in the Paris office of Handicap International, and within the French delegation attending meetings of government expert groups of the signatory states to the Convention on Certain Conventional Weapons of the Geneva Disarmament Conference (GGE, CCWC). Data collected at this time were complemented by interviews with the key actors in this process.  Although interviews were the principal method of empirical investigation, careful study of administrative documents produced during this process turned out to be critical in gaining an understanding of the work by the Ministries of Defence and Foreign Affairs in managing these dossiers. Consultation of this “bureaucratic output” (diplomatic instructions, internal notes, minutes of meetings, emails, annotations in the margin of formal and informal documents) effectively provided privileged access to episodes in the decision-making process which the author was unable to witness directly. This approach, with its focus on administrative documents, not only allowed me to reconstruct the evolution of the dossier, by observing modifications and reformulations at each stage of the process: it also, above all, allowed me to reconstruct the progress of power struggles between the two ministries’ administrative agents and policy-makers during the diplomatic negotiations. This focus on the formal and informal exchanges revealed by these documents thus provided insights into the daily routine of negotiations, which is particularly useful in order to understand the underlying logic at play in the evolution of decision-making.
Mechanisms of the rhetorical trap and the effectiveness of naming and shaming strategies: going beyond the alternatives of choice or coercion
6Reflections on the role of ethics in international relations have long been considered the poor relation of a field of research which is essentially focused on the study of relations between states in terms of power struggles, where the protection of national interests and the maximisation of power count for far more than justice or defence of values. However, over the past twenty years, the attention given to the emergence of transnational mobilisations has revived reflection on norms, rules and values  to explain the success of certain transnational campaigns with a moral dimension. Can the moral arguments put forward by the actors in these mobilisations serve as a guide for their behaviour? How? Under what conditions?
7To answer these questions, a series of recent studies in international relations sets out to discover what underpins the strength and the prescriptive nature of moral arguments. Seeking to break away from a circular reasoning process in terms of moral persuasion, these studies point to the strength of strategic use of moral arguments in decision-making. In so doing, they invite us to consider the possibility that the normative aspect of these decisions is attributable less to the attractiveness of moral values than to the power of coercion exercised by reference to these values.  There should thus no longer be a strict opposition between coercion and persuasion, power and ideas;  instead this work seeks to show that evoking moral principles, when this succeeds in restricting the choices available, is a powerful tool of coercion.  This article focuses on the conditions for exercising this power of coercion, highlighting in what circumstances, and how, morality imposes constraint in a public action process. It aims primarily to show that rhetoric has a role to play, and demonstrate how it plays this role, without seeking to evaluate the gap between normative acts of engagement and actual practices, nor to point to the possible underlying cynicism involved in the recourse to moral justifications.
8Analysis of the mechanisms of the rhetorical trap most often focus attention on the constraints created by previous normative engagement by states on the international stage. Mechanisms for reduction of the margins for manœuvre are thus one of the levers of rhetorical coercion: from this perspective, the success of an advocacy campaign is explained not so much by its capacity to convince people of the rightness of its struggle, but by the way in which this struggle deprives its opponents of the ability to build a socially acceptable objection. Rhetorical action is thus characterised by its capacity to “tie the hands” of its targets, that is, to restrict their action by imposing a discursive norm that limits the options for expressing opposition. By what means, and under what conditions, can this rhetorical action be effective?
9One of the situations in which rhetorical coercion appears to be most effective is when an actor feels compelled to modify their behaviour in order to maintain coherence between their verbal position-taking and their actions. This is moreover one of the ways in which the practice of naming and shaming works, by publicly exposing contradictions between the behaviour and the normative engagements of their targets. For the advocates, the stakes consist essentially in exerting pressure on their interlocutors, pointing to the gap between their behaviour and their engagement in order to threaten their credibility. A “rhetorical trap” exists when the target of the actions of naming and shaming has no choice but to comply with the demands of the advocates in order to avoid contradicting the principles that the target is seeking to defend.
10It therefore appears that rhetorical coercion works primarily on condition that the target values their reputation. This is the major contribution of research by Ronald Krebs and Patrick Jackson, who suggest that we should take into account the different and unequal results on the behaviour of states. According to these authors, mechanisms for rhetorical coercion must be understood by considering both the framework, which defines the stakes at play in the struggle, and the practical implications of these stakes. From this starting point, Jackson and Krebs identify several situations: the targeted actor accepts the framework and its implications, and therefore modifies their behaviour (policy change); the targeted actor rejects the framework and its implications; they accept the consequences of rhetorical action but reject the framework (framing contest); they accept the framework but not its implications (implication contest). In this article we will study this last situation, “implication contest”: the actor targeted by rhetorical action subscribes to the definition of the humanitarian stakes without necessarily accepting what this implies in practice.  In effect, public authorities never question, at any point, the idea that cluster munitions cause humanitarian effects on civil populations: however, they take up strict positions against the proposal for development of a restraining legal instrument which would ban the use of these weapons.
11Under these conditions, action taken under rhetorical compulsion does not necessarily imply that the actors fully support this;  nor is this seen as an absolute “nonchoice”, contrary to the idea developed by Janice Bially Mattern.  To speak of a “constraint on alternative courses of action” allows us to capture the dynamic of this movement, which consists simultaneously of the constraints which affect decision-making and the opportunities which arise from conforming to these constraints. It is this dynamic that this article seeks to describe.
Is foreign policy trapped by its own principles?
12At first sight, the concept of a rhetorical trap seems a fitting description for the French engagement in the Oslo process in February 2007. This interpretation certainly finds a series of echoes in this field of research.
13Between the launch of the mobilisation of NGOs at the end of 2003 and the start of the Oslo process in February 2007, the question of cluster munitions was addressed within the framework of the United Nations 1980 Convention on Conventional Weapons (CCW).  By virtue of the fact that no decision could be taken without consensus, the member states found themselves with a de facto right of veto. This precluded the option for the NGOs to count on the larger number of “small” states to impose legal constraints on the great military powers such as the US, China, Israel, or Pakistan. In other words, although the support of the “small” states is valuable for NGOs, it does not allow them to overturn power relations in the Convention on Conventional Weapons. In this context, disarmament or arms control depends solely on the will of the member states. In this respect, the Convention on Conventional Weapons “protects” member states, more specifically the great military powers, since it guarantees them control over the outcome of negotiation. France’s interest in keeping discussion on cluster munitions within this negotiation framework is therefore understandable. The blocking mechanisms inherent in UN processes for arms control allowed France to take an intermediary stance, between support for transnational mobilisation and the refusal of legislation to support the strengthening of international humanitarian law. Throughout this period, between the launch of the NGO campaign in November 2003 and the start of the Oslo process, outside of the traditional conclaves of arms control, France’s position was defined by this stance of non-engagement, which sought to demonstrate the futility of a specific treaty on cluster munitions, and recommended instead a reflection, within the CCWC, on the technical and legal measures to be adopted in order to reduce the devastating humanitarian effects of cluster munitions.
14Although this balance of power was maintained until the end of 2006, it came unstuck at the Convention on Conventional Weapons Review Conference. As the moment when the agenda for this negotiating forum was to be defined, the review conference had already been identified by campaign activists as a pivotal stage, either to embed the question of cluster munitions as a priority for the next round of negotiations, or to announce the launch of a parallel process, bringing together member states in favour of a ban on cluster munitions, along the lines of the Ottawa process to ban anti-personnel mines. The war in Lebanon in summer 2006, and the use of cluster munitions during the conflict, provided lobbyists with a new opportunity to warn member states of the devastating humanitarian effects caused by the use of these weapons. By providing images and eye-witness accounts, the war in Lebanon effectively fed into the strategy of condemnation initiated by the NGOs.
“It is a horrible thing to say, and very cynical, but lo and behold, we were suddenly able to say, ‘Everything that we have been telling you for years, to which you replied that we were exaggerating, well, here you are, this is an extreme example of everything we have been denouncing for years’. It’s an awful thing to say, but in some respects we have to thank Israel. This has brought about an acceleration of the process, with images and eye witness accounts.’” 
16Despite these events, the question of legislating arms control was not put on the CCWC agenda at the end of 2006. It was this context of blocking negotiations within the UN framework that gave the NGO partners in the mobilisation movement the opportunity to renew the initiative of a parallel process, similar to the one which had given rise, ten years earlier in Ottawa, to the ban on anti-personnel mines.
17If the launch of this “new Ottawa” heralded a major turning point in the history of the ban on cluster munitions, this was because the engagement in diplomatic negotiations outside of the traditional UN framework obliged the French government to take a position in relation to this initiative. Previously French public authorities had been content to point to the absence of responsibility on the part of French military officials for the devastating effects observed by humanitarian actors. From this perspective, their discourse focused on technical and administrative measures put in place to solve humanitarian problems. However, this non-decision-making strategy became more difficult to sustain after the launch of the Ottawa process, as the question of France’s participation in this new process was henceforth framed in terms which excluded the possibility of non-engagement. From that moment on, the question was no longer whether the technical and legal measures put in place by industrial powers and by the French army were sufficient to limit the harmful effects of cluster weapons. The question was rather whether France would join the new diplomatic process or whether her initial position of maintaining the status quo would continue to be “tenable” in this new context. Several options were available. Either the public authorities would engage in the negotiation in order to try to influence the negotiation process, simultaneously running the risk of being caught up in a process which they could not control and finding themselves saddled with a ban that they had been refusing since the beginning. Or the French government would choose to remain outside of the new process, challenging its legitimacy to justify maintaining the status quo, but, in so doing, aligning itself with the USA, China, India, Pakistan, and Israel, countries that were firmly opposed to any control of these weapons.
“You were either in or out. If you were out, you would find yourself in the company of countries such as China, India, Pakistan, the USA…. Is that what we wanted? […] To be isolated within the EU and amongst Western countries? Germany was in, so was the UK, what sense would it have made not to be in as well, what image would that have given of France?” 
19Choosing one or the other option depended on the evaluation of perceived risks for each option. By engaging with the Oslo process, where there was no requirement for majority consensus, France ran the risk of not being able to defend her interests when faced with a majority of countries which were not producers or users of cluster munitions. On this point, it is not surprising that the French public authorities preferred discussion to be held within the traditional framework of the UN Conference on Disarmament. In effect, while negotiations took place in this context, France could count on the USA, China, India, Pakistan and Israel to block negotiations. With the protection offered by the veto of these countries, as cluster munition producers and users, France could easily sustain a position of non-engagement, which reconciled the preservation of the status quo and the political acceptability of her position: in other words, a position in which France could reconcile the defence of the interest of her armed forces with a display of humanitarian conscience. However, if negotiating within the Ottawa Process carried risks, so too the risks of inaction had to be evaluated. Although the public authorities had previously been able to rely on the blocking of negotiations within the UN to allow the continuation of the status quo, without public exposure of their opposition to the demands of the NGOs, this position became more difficult to sustain in the context of the Ottawa Process, since it represented de facto open opposition to the initiatives of the NGOs, who were held up as the defenders of international humanitarian law.
20This was the point at which a public denunciation by the NGOs became effective and credible.  In effect, if negotiations at the UN and the rhetorical strategy of the French delegation had hitherto offered little scope for criticism by the NGOs, the emergence of concrete diplomatic negotiation outside of the UN framework now made the contradictions of the French position sharply evident. By remaining outside of this process, France would openly contradict the humanitarian values which she claimed to uphold on the international stage. Handicap International used these very arguments to point out the incoherence and contradictions of the French position.
“When you attended cabinet meetings at the Quai, how did that go?
We did not really go into technical debates. We spoke more about diplomatic conferences, we said things along the lines of, “France cannot possibly not join this process”.
You said that?
Yes, well, we told them, ‘You cannot make speeches about how we must defend civilian victims and then not take part in this process…’” 
22Under these conditions, the threat of public denunciation by the NGOs seemed more difficult to defuse or contradict publicly. Posing a real risk of public censure, the threat deprived the authorities of the possibility of defending a politically legitimate alternative position.
23Here we can see how recourse to strategies of naming and shaming, in the context of the Oslo process, reduced states’ room for manœuvre. The reference to humanitarian engagements, mobilised at the start of the campaign, was not sufficient to preclude any desire for resistance. But mobilised at the time of the launch of the Oslo process, that is, with a limited range of roles to be adopted, it seemed to make France’s participation in the new negotiation inevitable. The effectiveness of shaming strategies therefore does not only lie in the delegitimisation of practices which do not conform to the principles of human rights. It is true that the threat of deligitimisation represents a constraint, but one with which public authorities managed to live, up until the launch of the Oslo process.  It was only following the launch of this parallel process that this threat became effective; that is to say, reference to human rights gave the NGOs and representatives of the transnational mobilisation the means to put pressure on public authorities. In these conditions, can we speak of moral coercion at this particular point? The question is even more pertinent in that the French position did not change immediately following the announcement of this new process. It was only when France took part in the opening conference in February 2007 that she engaged officially in negotiating a disarmament treaty. What forces were at play at that moment which meant that evoking human rights and humanitarian principles had the power to limit the margins for manœuvre of public authorities, and to influence their decision-making?
Constructing the inevitability of change: the double game of constraints
24During the months following the launch of the Oslo process in November 2006, up to the opening of negotiations in February 2007, an internal series of deliberations took place intended to determine whether France would join the Oslo process or whether she would remain on the outside. An inter-ministerial negotiation was set up between military officials of the division Maîtrise des armements de l’état-major des armées, who were keen to establish their initial position securely, and their counterparts at the Quai d’Orsay, who were less and less convinced of the viability of this position in the prevailing diplomatic context of early 2007. This progressive fragmenting of interests highlighted the role played by civil servants at the Quai d’Orsay in the shift in position. Exchanges between the official bureaucracies of the two ministries offer in this respect a rich empirical resource for detailed study of the constraints which influenced decision-making, as well as allowing us to follow the process of reformulating the stakes in the new context: rhetorical coercion worked because civil servants at the Quai d’Orsay recognised the inevitability of a change in position and worked hard to convince their counterparts at the Ministry of Defence. Several arguments were put forward by the diplomats in this respect: they pointed to the prospect of diplomatic isolation if France’s traditional partners announced their involvement in the process, and the anticipation of a window of opportunity to win votes at national level. We will study in more detail this interdependence of constraints arising from both the external context and from domestic politics, and the respective influence of each in the decision-making process, in what follows.
The threat of diplomatic isolation
25The period stretching from the third CCWC Review Conference in November 2006 to the Oslo process in February 2007 was characterised by a strong sense of uncertainty as to the French position in relation to the Norwegian initiative. Having declared, in November 2006, that she would not participate in this process, France nevertheless went back on this position. Up to that point, the diplomats had tended to align with the position adopted by the military. However, the need to take a clear position with regard to the Norwegian initiative changed the stakes to some extent. A set of diplomatic constraints had been added to the operational and strategic constraints described by the military officers. The increasing level of intervention by diplomats in public information and in reflection on the implications of the evolving international context went hand in hand with the shift which reformulated the issues at stake. As the time for the Oslo Conference drew near, France engaged in trilateral consultations in Geneva with the British and German delegations. The view that the engagement of France in the Oslo process was inevitable rapidly emerged, given that the majority of European states, notably Germany and the United Kingdom, would be participating in it. This was the moment when the threat of diplomatic isolation became real for the French delegation.
“Between the moment when the Norwegians launched the Oslo process at the end of November and the first Oslo Conference in February, there were several consultations in Geneva. So we met on an almost daily basis, asking, ‘Are you going to Oslo’? Paris was constantly hanging on our coat tails, saying, ‘Tell us who is finally going to Oslo?’. Paris had no way of knowing who would be going and who would not. So, almost every day we met with the other delegations, everyone was saying, ‘if you go, I will’”. 
27This threat had been anticipated long before the Oslo Conference, including by the military members of the French delegation: however, it played an ambiguous role in the development of the French position. It is difficult to identify the precise moment in the deliberation efforts and preparation of decisions when it became a determining factor.
“Even if they shared our objectives, certain states (Germany and the United Kingdom) could be tempted to follow the two processes in parallel, specifically in order to seek to ‘control’ or to direct from the inside this quite radical process, initiated by Norway. As a domino effect, such rallying actions risked causing a swing by other EU countries in favour of the Norwegian initiative. This development would strike a blow to the credibility of the CCWC and France would find herself in a position of isolation that it would be difficult to sustain for very long”. 
29Following the CCWC Review Conference, and therefore following the announcement of the launch of the Oslo process by the Norwegian government, a change of position by France did not appear an obvious next step. French actors were presiding over a position of uncertainty at this point in the process, without deciding a priori whether they would align with the positions of their traditional allies, Germany and the United Kingdom. By asking Norway to grant them the status of observer at the conference, the leader of the French delegation prolonged this situation of uncertainty, thereby delaying the decision to join the Oslo process until the end of the opening conference. Thus, whilst almost all of the states officially expressed their support by taking part in the new process, the French delegation went to Oslo without any political instructions on the position to adopt. Responsibility for deciding on which position to take with respect to the new process was therefore left to the diplomats present at the Oslo conference.
“In practice, what happened is that we went to Oslo without knowing if we were going to sign up to the process, without any political instructions to sign or not to sign, and we had to evaluate on the spot whether we thought it was in our interests to take part or not.” 
31The account of events gathered through interviews  shows how the diplomats first initiated the change in position, stressing the threat of diplomatic isolation to convince military officials and political leaders that the French government found itself in a position where it was no longer possible not to engage.
“The Quai d’Orsay, in its negotiating role, said – and this is a point on which we did not even need to seek the view of the Ministry of Defence or the President ‘we believe that our interests will be better served if we take part than if we do not. In this situation, the Ministry of Defence allowed themselves’….(silence)
Allowed themselves to be guided?
Allowed themselves to be guided, let us say, they hesitated, they were somewhat apprehensive” 
33In this moment of “conversion” by the diplomats, it was the domino effect argument that was brought to bear, echoing a chain reaction scenario already anticipated within the Ministry of Defence.
“When the Oslo process was launched in 2007, we (Defence) were feeling strong following the result in the Sénat, we had the situation under control and we announced fairly quickly that we would not be going to Oslo. The Germans had to attend, but we, that is, Defence, we agreed with the Quai d’Orsay [Ministry for Foreign Affairs] and with the British. And then, at the last moment, the Quai d’Orsay changed its position, for fear of being isolated. The British sent someone and the Quai sent someone to Oslo in February. At that point our position was: the Ministry of Defence refuses to attend because if we were to do so, we would be supporting the process. It was somewhat twisted reasoning, I agree, but that’s how it went. At the last minute the Quai decided, ‘We’re going’. Defence said, ‘We’re not going’. I myself went on leave at that moment, which goes to show…it was important, because the Quai said, ‘we’re going but we’re not signing anything, we will not align with anyone, we’re not committing ourselves to anything’. And so they went. And then I had a phone call, I remember, around 10 o’clock in the morning, it was Grand who said, ‘I am not happy about this, but the British have said they will sign, I have to sign’. He wanted carte blanche from the Ministry of Defence, he held the knife to our throat. So I called my superior and we found ourselves committed to signing the final declaration.” 
35Seen from this angle, it was above all via a mechanism of interdependence of decision-making within diplomatic circles that scope for alternative courses of action was constrained. The legitimacy of the new process depended in effect on the number of states that would sign up to it, and the nature of the states which left the process. If it was supported only by small states, non-producers and non-users of cluster munitions, the negotiation would have little chance of producing a legally enforceable, effective document in humanitarian terms. If, however, the process were supported by producers and users of these weapons, such as Germany, France, and the United Kingdom, the diplomatic process would gain legitimacy, offering the possibility to endow the text of the future treaty with some degree of effectiveness, however relative. The more the treaty gained in legitimacy, the less France could argue against that legitimacy to justify not aligning with the process. Thus, faced with the about-turn of her European partners, Germany and the United Kingdom, France had no other choice than to commit to supporting a ban whose futility she had hitherto denounced. At that moment, the only way to protect France’s image was to avoid diplomatic isolation.
“The thinking was based in part on the fact that almost all EU countries had signed up to the process, that our great European allies, the British and the Germans, were signatories to the process and therefore it was important to join them and then, later, that it was important to sign the Oslo declaration.” 
37That the prospect of non-participation in the negotiation should become in this way inconceivable is highly revealing of the importance that the actors of the government departments, including the Ministry of Defence, attached to image and reputation on the international stage. It was this same concern that had guided the use of humanitarian rhetoric to oppose a proposal for specific regulation of cluster munitions up until this point, and which explained the initial consensus between the Ministry for Foreign Affairs and the Ministry of Defence on this position of non-engagement. However, it is not clear that evoking the stakes relating to image and reputation functioned only as a form of coercion. In fact, several other arguments were put forward by the diplomats to justify the shift in position. These arguments reveal that this shift, and the benefits that the French delegation could expect during the negotiation as a result, had been anticipated. However, what weight can we attach to these chain reaction mechanisms? What was the role played by the constraints of the external context in the deliberations? To what extent were they not simply deployed to persuade the delegation of the need to reposition?
“But perhaps the opposite happened, the United Kingdom signed because France signed…But we can’t know that.” 
39In asking these questions, the point is not so much to reject or to minimise the influence of conjunctural explanations as to seek to contextualise them by showing the place that they occupied in the deliberation mechanism, and the way in which they were constructed as a diplomatic and political stake by and for the diplomats. In this respect, if the shifting alliances between states influenced decision-making, to the point of giving the impression that the decision depended solely on events in the international context, they only offer a partial explanation for the tightening of constraints.  Elements of domestic politics also played a role in creating the inevitability of the change in position. The work of deliberation and preparation for decision-making by senior civil servants enables us to clarify the connections between these two levels of constraints.
Politicisation of the question and anticipation of domestic political constraints
40Amongst the actions initiated by the advocacy unit of Handicap International, practices at the time included email campaigns targeting elected representatives, the French President, and senior ministers concerned with this question. In the context of the 2007 presidential elections, the association called on candidates to feature the ban on cluster munitions in their election programme. Between March and April 2006, the Communist Party, the Greens and the UDF (Union pour la Démocratie française) demonstrated their support for the association. In September 2006, the association repeated this appeal to the 2007 presidential election candidates. In November all candidates, with the exception of Nicolas Sarkozy (UMP Candidate – Union pour le movement populaire), declared their support for a ban on cluster munitions. Echoing the position of the French government as expressed by Michèle Alliot-Marie, Minister for Defence, Sarkozy, as Minister of the Interior, recognised the impact of these weapons on civilian populations, but refused to envisage the possibility of a French initiative supporting a ban. On 20 December, Sarkozy went back on his position and in a letter addressed to Handicap International, indicated his intention, “resolutely to engage France in the international process towards a ban”, should he be elected.
41Ahead of the change of government which would take place in May 2007, the civil servants at the Quai d’Orsay seemed, in February 2007, to take more notice of the mobilisation of presidential candidates than of the positions taken by their cabinet.
“The decision to enter the Oslo process was also motivated by the fact that it was felt that the subject was gaining ground in political terms, both in France and on the international stage, etc. And I could see, at that time, that the Douste-Blazy cabinet was sufficiently aware of the fact, such that they did not want the minister to speak or to reply to members of parliament who wanted to ask questions [….]: they were told, ‘move on, there’s nothing to see’. It was clear that the issue was becoming a sensitive topic. The conditions of the moment played a role in our history: presidential candidates, at least the credible ones, such as Ségolène Royal, Nicolas Sarkozy and François Bayrou, with some nuances of degree, were more or less in favour of the campaign by Handicap”. 
“At that point, I recommended to Paris that they should join the Oslo declaration, on the basis of simple logic: firstly, ‘it is better to be part of the process than remain on the outside’. And secondly, our politicians were likely to change their minds on the issue of cluster munitions at the time of the elections, plenty of things were going to happen.” 
43The initiative thus came not from political decision-makers, but from senior civil servants who were anticipating a change in position by ministerial cabinets in tandem with a change in government. It was the possibility of engagement in the Oslo process by the future government, more so than political instructions from the government at the time, which won the day in the deliberations of senior civil servants. It was therefore the politicisation of the question within the electoral context which allowed senior civil servants to “escalate” the issue,  emphasising the risks involved in joining the negotiation “further down the line”.
“Our view was that we might as well be involved in the process and gain legitimacy for being there from the start, rather than behaving as we did, catastrophically, on the anti-personnel mines and join the party under pressure from public opinion, from NGOs, and from members of parliament, whilst being isolated.” 
45By recommending to political leaders that they sign the Oslo declaration, the diplomats in the French delegation were bowing not only to reasons of conformity which would compel France to align with the position taken by her European partners. They were also anticipating a shift in the French position which would lead to participation in the negotiation, without any opportunity to benefit from the political gain from a “moral” engagement, nor to benefit from the resources that being part of the process from the start could offer in terms of building alliances and securing compromise. In seeking to limit the political and symbolic risks of inaction, they were also responding to a rationale in favour of tactical engagement, which would allow them to limit the risks of that engagement.
Making the inevitable acceptable: the benefits of conversion, or how to avoid the trap whilst appearing to be “caught”
46Engaging in a diplomatic negotiation with a moral dimension in order to protect one’s reputation as a virtuous state on the international stage is undeniably a form of action taken under rhetorical coercion. However, this view does not take into account the resources that actors caught in the trap can derive from the use of moral arguments: if the naming and shaming strategies work, the target is not necessarily reduced thereby to silence or to inaction.  Understanding the way a rhetorical trap works therefore involves consideration of this trap within its temporal context, in order to make a timely evaluation of its capacity to “restrict” options for action.
An entryism strategy and instrumentalisation of the trap
47Emphasis on the diplomatic constraints brought to bear on the French delegation to explain their entry into the Oslo process should not overshadow the delegation’s ability to secure compromise and to influence the conditions of their involvement. In effect, behind the decision to seek an invitation to the Oslo Conference, there was also the idea that the French delegation would contribute to the production of the text of the declaration in order better to control the terms of its engagement. The action of the delegation with regard to the formulation of the Oslo declaration – together with a group of like-minded countries opposed to a total ban  – gives indisputable proof of this. The text of the declaration specified that signatory states to the Oslo declaration were committed to securing a ban on cluster munitions “which cause unacceptable harm”.  The addition of this deliberately vague phrasing, “which cause unacceptable harm”, was not without significance: it offered to like-minded countries some margin for manœuvre in the negotiation, by creating the option of withdrawing from a ban on advanced technological weapons, and leaving open the option of withdrawing support should the negotiations be heading towards a total ban. Therein lies the ambiguity of a step which involved increasing the number of safeguards, in order to commit to the process whilst shifting one’s fundamental position as little as possible.
“Before the Oslo conference, we succeeded in creating a like-minded group, which very fortunately ensured that the Oslo declaration was sufficiently ambiguous to allow us to avoid a total ban, thanks to that famous phrase, ‘ban cluster munitions which cause unacceptable harm’. Without that phrase, we would have been heading for a total ban.” 
49If this phrase in the Oslo declaration was so important for the French delegates, it was because it allowed them to engage in the negotiation whilst reserving the option to modify the outcome, either to secure a partial ban, or to withdraw from the process and break up the negotiations if these were seen to be leading to the adoption of a total ban. This is the underlying principle of “entryism” strategies: participating so as to control the risks of engagement, entering into negotiation in order better to control from the inside and to defend one’s interests.
50Up until February 2007, the question on which the representatives of the international mobilisation and the French delegates disagreed was the question of the advantage of a specific regulation on cluster munitions. With the entry of France into the Oslo process, the focus of the debate was moved to the question of the ban, and set those in favour of a total ban against those in favour of a partial ban. Despite this shifting of the focus of the debate, the fundamental French position remained unchanged, as by helping to define the objectives of any future negotiation France would effect a noticeable modification of the conditions of her engagement. By defending the notion of a partial ban on cluster munitions, France protected the overall stash of advanced technology cluster munitions in her arsenals. And by asserting the existence of a principle of “interoperability”, France kept open the option to participate in military operations with states who were not signatories to the treaty,  thus only slightly modifying her capacity for intervention alongside NATO forces in particular.
51At this point, an explanation in terms of a rhetorical trap certainly allows us to account for a process of reduction of public authorities’ room for manœuvre, by showing how evoking the international reputational stakes can influence the construction of foreign policy. It does however throw a veil over the capacity of public authorities, once “caught in the trap” to regain margins of freedom of action in order to reconcile a public position of humanitarian concerns with the defence of military and strategic interests. The deliberation efforts of the members of the French delegation reveal that the trap works because it recreates margins for those actors who allow themselves to be trapped.  The case of French engagement in the Oslo process thus explains how the power of humanitarian rhetoric rests as much on the legitimacy of its ideals as on the multiple appropriations of this rhetoric and on the diversity of actions which can make claims on this legitimacy.
Normative constraints and institutional legitimisation
52If the position of the French delegation throughout the Oslo negotiation process was intended to protect the weapons in her arsenals, it was nevertheless always in thrall to the constant concern for France’s image. Diplomatic instructions insist, in this respect, on the need to “protect the public diplomacy” of France.
“Given that the Oslo process was a political exercise, it was important to protect our ‘public diplomacy’ and to avoid shouldering any blame.” 
54Whilst they assessed the gaps between the text proposed by the promoters of the Oslo process and the “red lines” of the French position, the French delegation did however refuse to amend the treaty proposal.
“By proposing amendments we might have given the impression that we accepted the text as the basis for negotiation: if we offered clear opposition to certain measures, the NGOs would immediately have pointed the finger at us.” 
56On the other hand, drawing up a “non-paper” offered the opportunity to make proposals in an anonymous and non-official way, and therefore to oppose without being exposed. Most importantly however, because this “non-paper” was only distributed to countries holding cluster munitions, it allowed shared claims to be highlighted and, in this way, enabled the creation of a federation of states holding cluster munitions who are opposed a priori to a total ban. In this tactic we can observe one of the characteristic features of the “non-paper” identified by Jean-Michel Eymeri: “to allow ideas to circulate” or “to structure majorities based on ideas”, without being required to take an official position.  In having recourse to this tactic, public authorities show that they are as keen to control domestic processes, in order to control the outcome of the negotiation, as they are to reap the symbolic benefits of participation. This two-part approach was typical of the position of France, who would seek throughout the process to “salvage what she could” “by saving face”,  in other words, to (re)adjust the room for manœuvre which had been reduced by the government’s declared stance on humanitarian engagement. By the conclusion of the negotiation in May 2009, France seemed to have secured a certain number of guarantees on the need to take into account her interests in the areas of security and defence. The compromises granted by the core group and the NGOs on the question of definitions allowed France to agree to a treaty which was acceptable to the French Defence Ministry. In this respect, this analysis shows the extent to which the internationalisation of law does not necessarily go hand in hand with an erosion of the powers of the state. Building on the analysis by Julien Seroussi on the subject of universal jurisdiction, which shows how “state sovereignty represents a compulsory step in the fight against impunity”,  this research shows how legal gains secured by civil society remain pragmatically determined by the margins for manœuvre that they offer to the agents which they are intended to regulate.
57* * *
58By observing how the “moral obligations” of states give rise to ambivalent effects, this article does not seek to deny the power that evoking these moral effects possesses, but instead to show how these ambivalences thereby constitute one of the conditions of the prescriptive nature of these obligations. Humanitarian rhetoric is effective, not only because it allows “states to be trapped by their principles”, not only because it forces these states to honour their commitments, but also because it offers public authorities the opportunity to defend their interests while benefiting from the political gain arising from an engagement in the name of morality. That said, we should not conclude that the normative tendency of decision-making corresponds primarily to a symbolic public action,  by means of which the state seeks less the effectiveness of measures and more the promotion of its action on the public stage; instead, this demonstrates that this tendency aligns with a desire to legitimise institutions. Within this logic, moral stances in foreign policy cannot be explained solely by the political advantages that they secure for the actors who shoulder the responsibility of these decisions. They are also dictated by the institutions within which these actors operate.
59Although the focus in this article is on the limitations and resources of action taken under rhetorical coercion, it places the analysis of these logics of moral constraint outside the framework of strategic action and rational choice. In effect, this article offers an explanation of causal processes and the effects produced by the use of moral arguments in terms of the conversion of constraints into resources, and not by their opposition point by point. Ultimately, this attempt to explain decision-making within foreign policy reveals the extent to which state logics and the logics of human rights are closely interwoven. The article emphasises that state sovereignty was not defined only from the angle of resistance to the NGO’s humanitarian demands, but that this was also expressed by evoking the principles of human rights which the state prides itself on defending. As both factors and products of institutionalisation,  the “myths” promoted by institutions serve as much as a constraint on action as they do as a source of legitimacy that guarantees the continued existence of state institutions, by validating the actions of its leaders.  In this sense, the politics of human rights is nothing more than a dimension of political realism. 
Nicola Short, “The role of NGOs in the Ottawa Process to ban landmines”, International Negotiations, 4, 1999, 481-500; Amandine Orsini, Daniel Compagnon, “Les acteurs non étatiques dans les négociations internationales”, in Franck Petiteville, Delphine Placidi-Frot (eds), Négociations internationales (Paris: Presses de Sciences Po, 2013), 105-40.
Frank Schimmelfennig, “The community trap: liberal norms, rhetorical action, and the eastern enlargement of the European Union”, International Organization, 55(1), December 2001, 47-80.
Regarding the impact of transnational mobilisations on their targets, see Joshua W. Busby, Moral Movements and Foreign Policy (Cambridge: Cambridge University Press, 2010); Jonathan Fox, L. David Brown, “Assessing the impact of NGO advocacy campaigns on World Bank projects and policies”, in Jonathan Fox, L. David Brown (eds), The Struggle for Accountability. The World Bank, NGOs and Grassroots Movements (Cambridge, MA: MIT Press, 1998).
Johanna Siméant, “Advocates, gatekeepers et politique étrangère: comment les acteurs des mobilisations transnationales se font (parfois) entendre par les politiques”, Revue française de science politique, 62(4), 2012, 277-96.
J. W. Busby, Moral Movements and Foreign Policy.
Emilie M. Hafner-Burton, “Sticks and stones: naming and shaming the human rights enforcement problem”, International Organization, 62, 2008, 689-716; and Emilie M. Hafner-Burton, James Ron, “Seeing double: human rights through qualitative and quantitative eyes”, World Politics, 61(2), 2009, 360-401.
Hélène Dufournet, “Gouverner sans choisir. Entre contrainte morale et réalisme politique: l’engagement français dans le processus d’interdiction des armes à sous-munitions (2003-2008)”, PhD thesis in sociology, Cachan, École normale supérieure, 2011.
NGO representatives, officials of the Office for Arms Control (Maîtrise des armements à l’état-major des armées) in the Ministry of Defence, members of the Office for Multilateral Affairs and Disarmament (Affaires multilatérales et du désarmement) in the Ministry for Foreign Affairs, and of the Permanent Mission of France at the Geneva Conference on Disarmament, advisors from the cabinets of the Ministries of Defence and Foreign Affairs.
For discussion of the role of ethics in international relations theories, cf. Jean-François Thibault, “Le tournant éthique”, in Alex Macleod, Dan O’Meara (eds), Théories des relations internationales (Outremont: Athena éditions, 2nd edn, 2007), ch. 23, 489-505.
Janice Bially Mattern, “Why soft power isn’t so soft: representational force and the sociolinguistic construction of attraction in world politics”, Millennium, 33, 2005, 583-612; Patrick T. Jackson, Ronald R. Krebs, “Twisting tongues and twisting arms: the power of political rhetoric”, European Journal of International Relations, 13(1), 2007, 35-66.
Richard Price, “Transnational civil society and advocacy in world politics”, World Politics, 55(4), 2003, 579-606 (590); Ann Florini, The Third Force. The Transnational Civil Society (Tokyo/Washington: Japan Center for International Change/Carnegie Endowment for International Peace, 1999); Kathryn Sikkink, “Restructuring world politics: the limits and asymmetries of soft power “, in Sanjeev Khagram, James V. Riker, Kathryn Sikkink (eds), Restructuring World Politics. Transnational Social Movements, Networks, and Norms (Minneapolis: University of Minnesota Press, 2002), 301-18 (303-04).
J. B. Mattern, “Why soft power isn’t so soft…”.
P. T. Jackson, R. R. Krebs, “Twisting tongues and twisting arms…”.
P. T. Jackson, R. R. Krebs, “Twisting tongues and twisting arms…”; Pascal Vennesson, “War under transnational surveillance: framing ambiguity and the politics of shame”, Review of International Studies, 40, 2014, 25-51.
J. B. Mattern, “Why soft power isn’t so soft…”.
Convention on the ban or the restriction of use of certain conventional weapons which can be considered to produce excessively devastating effects or to strike without discrimination. (Note: all translations from French are by the translator of this article, unless a published English source is given.)
Interview at the Advocacy Unit of Handicap International, March 2009.
Interview at the Permanent Mission of France to the Conference on Disarmament, January 2009.
J.B.Matters, “Why soft power isn’t so soft…”.
Handicap International advocates, January 2010.
Hélène Dufournet. “Quand techniciser, c’est faire de la politique sans le dire”, Gouvernement et action publique, 1(1), 2014, 29-49.
Interview at the Permanent Mission of France to the Conference on Disarmament, January 2009.
Extract from the minutes of the Third Review Conference of the Geneva Convention (1980) on Certain Conventional Weapons, November 2006, office of Arms Control, chiefs of staff, Ministry of Defence.
Interview at the Office for Multilateral Affairs and Disarmament, Ministry for Foreign Affairs, December 2008.
Although I was able to consult a large number of instructions, telegrams, and minutes of negotiations produced by the administrative services of the two ministries during the whole process, those from the Oslo Conference of February 2007 were not accessible. I have reconstructed this sequence from interviews with military officials and diplomats of the French delegation.
Interview at the Office for Multilateral Affairs and Disarmament, Ministry for Foreign Affairs, December 2009.
Interview at the Office of Arms Control, Ministry of Defence, April 2009.
Interview at the Office for Multilateral Affairs and Disarmament, Ministry for Foreign Affairs, November 2009.
Interview at the Office of Arms Control, Ministry of Defence, April 2009.
Robert D. Putnam, “Diplomacy and domestic politics: the logic of two-level games”, International Organization, 42(3), summer 1988, 427-60.
Interview at the Office for Multilateral Relations and Disarmament, Ministry for Foreign Affairs, November 2009.
Interview at the Office for Multilateral Relations and Disarmament, Ministry for Foreign Affairs, December 2008.
Jacques Lagroye, “Les processus de politisation”, in Jacques Lagroye (ed.), La politisation (Paris: Belin, 2003), 359-72; Jean-Michel Eymeri, “Définir la ‘position française’ dans l’Union européenne: la médiation interministérielle des généralistes du SGCI”, in Olivier Nay, Andy Smith (eds), Le gouvernement de compromis. Courtiers et généralistes dans l’action publique (Paris: Economica, 2002), 149-75; and “Frontières ou marches? De la contribution de la haute administration à la production du politique”, in J. Lagroye (ed.), La politisation, 47-77.
Interview at the Office for Multilateral Relations and Disarmament, Ministry for Foreign Affairs, December 2008.
P. Vennesson, “War under transnational surveillance…”.
This term “like-minded” describes the group of countries who, whilst participating in the Oslo process, were opposed throughout to a total ban on cluster munitions, Up until the Dublin conference in 2009, France presided over this group of like-minded countries.
“Recognising the grave consequences caused by the use of cluster munitions and the need for immediate action, states commit themselves to conclude by 2008 a legally binding international instrument that will prohibit the use, production, transfer and stockpiling of cluster munitions that cause unacceptable harm to civilians” (extract from the Oslo declaration, February 2007).
Interview at the Permanent Mission of France at the Conference on Disarmament, January 2009.
Intervention by Ambassador Jean-François Dobelle, permanent representative of France to the Convention on Disarmament, Vienna, 6 December 2007. This intervention by the head of the French delegation to the Vienna Conference echoes point by point the diplomatic instructions given by the Quai d’Orsay to the delegates. It thus provides an excellent illustration of the French position at this point in the process.
P. Vennesson, “War under transnational surveillance…”.
TD DMST GENEVE 328, from the Ministry of Justice, 30 November 2007. Subject: cluster munitions. Preparation for the Vienna meeting 5-7 December. Brainstorming meeting with like-minded countries within the Oslo process on the Vienna document.
Preparatory documents for the Vienna Conference, analysis of the text proposed by the Austrian presidency, permanent Mission of France to the Convention on Disarmament.
“Commission civil servants, and sometimes the rotating presidency of the Council of Europe, tend before putting out an appropriately constructed EU proposal, to produce some form of anonymous non-paper, with a single, very Kafkaesque or Orwellian heading, a document whose main striking feature is its role in ‘circulating ideas’ and structuring ‘majorities based on ideas’ without allowing one of the major member-states to kill the initiative in its infancy, or at least to take an official position” (J-M Eymeri, “Définir la ‘position française’”).
Hélène Dufournet, “‘Sauver les meubles’ en ‘sauvant la face’: faire la norme ‘à l’ombre’ de la communication politique”, Droit & Société, forthcoming, 2016. This article shows that although the two goals, “saving the furniture” without “losing face” may appear to be contradictory, the actions deployed to achieve these goals are quite subtly expressed in reality. Negotiating the rule was done in the shadow of the media coverage sought by the Quai d’Orsay. This silent work of rewriting norms is what allowed for an adjustment of the scope for alternative action, that the necessarily rowdy staging of a more proactive strategy threatened to constrain. The wholly ambivalent nature of French foreign policy in the diplomatic negotiations rested on the gap between the staging of public action, as maintained by voluntarist discourses of political leaders, and the rationale for such action aimed at reshaping the room for manœuvre, for which civil servants took responsibility during the work of producing the norm.
Julien Seroussi, “Les acteurs nationaux du droit pénal international: le cas Pinochet”, L’Année sociologique, 59(2), 2009, 403-15.
Murray Edelman, Politics as Symbolic Action (Chicago: University of Illinois Press, 1976 [1st edn 1964]).
Brigitte Gaïti, “Entre les faits et les choses, la double face de la sociologie politique des institutions”, in Antonin Cohen, Bernard Lacroix, Philippe Riutort (eds), Les formes de l’activité politique (Paris: PUF, 2006), 39-64; Jacques Lagroye, Michel Offerlé, Sociologie des institutions (Paris: Belin, 2011).
On how institutions maintain the loyalty of those who run them, see Jacques Lagroye, Johanna Siméant, “Gouvernement des humains et légitimation des institutions”, in Pierre Favre, Jack Hayward, Yves Schemeil (eds), Être gouverné. Études en l’honneur de Jean Leca (Paris: Presses de Science Po, 2003), 54-5; John W. Meyer, Brian Rowan, “Institutional organizations: formal structure as myth and ceremony”, American Journal of Sociology, 83, 1997, 340-63.
I would like to thank Vincent Simoulin, Thomas Lindemann and Hélène Baillot, as well as anonymous readers of the Revue française de science politique, for their feedback and suggestions on previous versions of this article.