Terrorism, counter-terrorism, and mimetic rivalry. Interview with Mathias Delori

28/04/2025

In his recent book La guerre contre le terrorisme comme rivalité mimétique, published by Peter Lang, Mathias Delori, a political scientist at the CERI (Sciences Po, Center for International Studies, CNRS) examines the conceptual categories that encompass the notions of terrorism and counter-terrorism, their relevance for thinking about and understanding these phenomena, and their impact. Based on a socio-historical investigation involving archive research and interviews, Mathias Delori analyses how supporters of armed jihad and the global war on terror have constructed a world that has made an escalation of violence possible. Read our interview.

In an earlier interview you stressed the difficulty of defining terrorism, which you said could not be considered a sociological concept. Could you tell us how you went about defining terrorism and studying its forms in this book?

There are academic definitions of terrorism. For Raymond Aron, any act of violence whose “psychological effects are out of proportion to its purely physical result” is labelled terrorist.1 The sociologist Isabelle Sommier refined this definition by suggesting that armed groups can produce such effects by creating a “disjunction” between the victims, often civilians, and the target, namely the state.2 These definitions are useful for describing and understanding certain phenomena of domestic political violence, especially when liberal states refrain from resorting to exceptional counter-terrorism policies, thus creating a clear demarcation between terrorism and counter-terrorism. Norway provides an interesting example. On 22 July 2011, the neo-Nazi Anders Behring Breivik killed 77 people in a terrorist attack—according to the above definition—and the Norwegian state responded by mobilising security instruments governed by liberal law: the police and the judiciary in relation to Breivik, and the intelligence services to prevent future tragedies of the same kind.

On the other hand, these definitions are not operational for describing and understanding political violence in illiberal contexts and in the context of war, because the methods of both sides tend to be similar. At least, this is what I observed during this research on the “global war on terror”, namely all the military operations carried out by the United States, the United Kingdom, and France since the late 1990s to fight al-Qaeda, then the Islamic State (IS) organisation, and their cronies. Reflexivity leads us to conclude that the above definitions apply as much to certain tactics of “counter-terrorist” warfare as to what we call “terrorism”. For example, the archives of the Bush administration's policy of “enhanced interrogation” in the 2000s leaves no doubt: its instigators knew that the majority of people tortured were innocent, and one interrogation technique—technique number 35 in the repertoire of “standard interrogation techniques” at Abu Ghraib prison—involved the use of dogs to “frighten” detainees.3 The same can be said of drone surveillance and attacks in Yemen, Somalia, Pakistan, and Iraq since the mid-2000s. Our colleague Rebecca Mignot-Mahdavi points out in a recent book4 that these aircraft produce psychological effects that are out of all proportion to the physical result of the action, as in the Aronian definition of terrorism. While drones kill fewer people than fighter-bombers, all other things being equal, their constant presence in the sky, their whirring, and the constant threat they pose create stress and trauma, the extent of which is difficult to measure in our part of the world.

Especially because the terrorists of some are often the freedom fighters of others...

Yes. In addition to the problem of the non-discriminatory nature of these definitions of terrorism in a war situation, there is another: it is difficult to strip the term of its negative normative overtones. If I argue that the Bush administration's policy of “enhanced interrogation” is terrorism, I run the risk of being suspected of morally condemning this policy, or of wanting to relativise al-Qaeda's terrorism, which is not my intention. I simply want to do the work of a social scientist: to understand, not to excuse, but to enlighten.

I get around these epistemological obstacles by using the term “terrorism” in quotation marks. These quotation marks are not intended to reverse the accusatory logic. They simply indicate that I am not using the term as a sociological concept. The “terrorism” referred to in this book includes all acts classified as such by the three main states in the “global war on terror”: the United States, the United Kingdom, and France. This “terrorism” has a material aspect—explosions, deaths, injuries—and a discursive dimension: the fact of describing this violence as “terrorist” rather than as an act of resistance, an asymmetric warfare tactic, or even a common crime. Talking about “terrorism” rather than “political violence perpetrated by non-state groups”, as a number of researchers do, has an advantage: one can question what this categorisation produces. In this case, my argument is that this categorisation, and the warlike practices that accompanied it, produced an escalation of violence between 2000 and 2010.

Precisely, you want to understand these two forms of violence—Islamic terrorism and warlike counter-terrorism—and you reject explanations in terms of radicalisation (of the terrorists) or imperialism (of the supporters of the war on terror). Your argument is that “terrorism” and warlike “counter-terrorism” represent two sides of the same violent relationship. Can you elaborate on this?

Yes. These explanations in terms of the radicalisation of the one and the imperialism of the other have one thing in common: they focus on only one of the two belligerents, while at the same time casting a normatively critical eye on their practices. The relational approach to violence that I use in this book is intended to be reflexive in relation to the normative relationship to the object. I am not unaffected by this violence. I have lost an acquaintance in an attack, but I try to find the right distance. This approach is also more enlightening. First of all, it allows us to do justice to the meaning that both sides give to their practices. The supporters of the “global war on terror” and the “terrorists” of al-Qaeda or IS make it very clear: they are at war. They do not use this term metaphorically, as Emmanuel Macron did in the early hours of the COVID-19 pandemic. More than being at war, they are waging it. Specifically, the French bombings in Iraq since August 2014 and commando attacks like the Bataclan in 2015 are acts of war. War is a violent relationship.

The relational approach to violence that I mobilise differs from that of the hard core of strategic studies. The latter postulate that war is a violent interaction structured by rational adaptations to the actions of the opponents. Indeed, such a logic can be found in the global war on terror. For example, the centrality of drone strikes in Iraq, Yemen, Somalia, and Pakistan from the late 2000s onwards can be interpreted as a strategic adaptation to the suicide bombings that caused too many casualties among the counter-‘terrorist’ ground forces in Afghanistan and Iraq. This interactional and rationalist interpretation can also shed light on the Islamic State organisation's use of attacks against civilians from the countries of the international coalition since 2014/2015 onwards. Targeted from the air, this organisation was unable to retaliate against combat troops. It reacted like many organisations in the same situation did: by intensifying its violence against civilians.5 However, this framework for analysing strategic adaptation has a blind spot. It is unable to account for the evolution of identities, representations of the Other, and warlike affects over the course of the violent relationship. This is where the theory of mimetic rivalry comes in.

Can you tell us a little about this theory of mimetic rivalry and how you set about “sociologising” it?

I borrowed this theory from the anthropologist René Girard. He is a unique author in the field of the social sciences and humanities. Indeed, while most researchers endeavour to produce medium-range theories, i.e. theories that shed light on a limited set of phenomena, Girard sought to explain violence in all its forms and in all latitudes from antiquity to the present day. The source material he used—in particular Greek tragedies and novels of the modern era—is also original, especially as he interpreted them not only as fictional narratives but also as commentaries on reality. It should be noted that Girard demonstrated a certain reflexivity on the iconoclastic nature of his approach when he wrote that his theory is based “on facts whose empirical nature is not empirically verifiable”.6 Girard’s theory does not seem to me to be applicable, as it stands, to a social science investigation. I therefore applied myself to sociologising it, and I am not the first to do so. Didier Bigo, Daniel Hermant, and Xavier Crettiez were pioneers in this field. The journal Cultures et Conflits also contributed to this research. Finally, my work owes a great deal to the reflections carried out in the 2010s within the OCTAV group (Observatoire Collaboratif sur le Terrorisme, l'Antiterrorisme et les Violences) led by Philippe Bonditti.

René Girard developed this theory over half a century, during which time he formulated several variations. The first, set out in Mensonge romantique et verité romanesque (1961), centres on the notion of “mimetic desire”. It stipulates that human beings believe they desire things, people, or ideas for themselves when in reality they take each other as a model and desire what the other desires. Girard adds that the actors do not recognise the mimetic dimension of their desires. In Stendhal's Le Rouge et le Noir, for example, Julien Sorel believes he desires Madame de Rênal romantically, that is to say for herself and against social conventions, whereas in reality he imitates Monsieur de Rênal, the mayor of the small town of Verrière, his role model. This is the meaning of the title of Girard's book: there is a “romantic truth” behind the “romantic lie” of desire for the loved one or the loved thing. This carries the seeds of violence, because “two desires that converge on the same object are mutually obstructive”.7 In his book Achever Clausewitz (2007), Girard gives the example of the Franco-German wars, which he believes were caused by the mimetic desire to be the leading European power and the sole standard-bearer of “Kultu”’ and “civilisation”. However, this variant of the theory of mimetic rivalry cannot be articulated with the socio-historical approach that I deploy. His theory of action is too restrictive. Desires, mimetic or not, are not the only determinants of social action, at least not in war. The theory is also too deterministic. The actors appear as elements of mimetic mechanisms that totally surpass them.

In Violence and the Sacred (1977), Girard formulates a variant of his theory that is complementary to the first, independently mobilisable, and more compatible with a socio-historical approach. It focuses not on the origin of violence, but on its development. The question is: why do some rivalries escalate into violence, while others are resolved peacefully or in the form of a violent conflict stabilised within a system of limits? Girard replies that human societies that have survived self-destruction have invented stratagems: scapegoat rituals in the case of traditional societies and criminal justice in the case of modern societies. But international criminal justice is too uninternational and too weak to regulate all conflicts and their mimetic tendencies. This stratagem is therefore inoperative.

What about your approach to Girardian theory?

So my theory of mimetic rivalry is based on another idea that runs through all of Girard's work: the question of the actors' recognition (or not) of the relational nature of violence. Violent situations in which the actors recognise the relational nature of violence do not tend to escalate. They can even lead to reconciliation, as Valérie Rosoux has shown with regard to Franco-German relations.8 Conversely, the representation of the other as the aggressor tends to lead to escalation, because it is easy to move from this idea to the idea that the other is a cruel being (an infidel, an Islamic fanatic, etc.) and to the conclusion that, in the face of cruelty, it is legitimate to go beyond the limits of what is considered in normal times to be just violence. Violent practices become perfectly mimetic when counter-cruelty responds to cruelty and vice versa. Girard speaks of a “monstrous double”, Germaine Tillion of “complementary enemies”.9

In the violent configuration studied in The War on Terror as Mimetic Rivalry , this destructive mimicry begins when al-Qaeda believes that the US bombings in Sudan and Afghanistan in August 1998 justify a large-scale attack on civilians (the attack of 11 September 2001), and when the United States declares on 1 January 2002 that al-Qaeda “terrorists” could be considered “unlawful combatants” and therefore not benefit from the protection of the core Geneva Conventions. Subsequently, attacks on civilians were met with torture, and torture was met with attacks on civilians. The similarity lies in the fact that the counter-violence of some appears as cruelty to others, and vice versa.

What is the merit of this theory of mimetic rivalry in the present case?

Apart from being agnostic about the origin of the violence, this interpretation of the theory of mimetic rivalry avoids the pitfall of determinism. While the logic of escalation has been dominant, it is quite true that some actors have chosen to leave the violent relationship after recognising its relational nature. At the level of collective actors, we can cite the Spanish government’s decision to withdraw its troops from Iraq after the Madrid bombings in 2004. In my book, I also document individual trajectories. At the beginning of May 2004, for example, the American businessman Nicholas Berg was captured and then beheaded by militants from a group led by Abu Musab al-Zarqawi. The video, the first of its kind, was seen around the world. Two years later, the US armed forces retaliated by assassinating al-Zarqawi. In response, Nicholas Berg's father expressed his dismay with dignity at this “counter-terrorist” notion of retribution: “My reaction is that I feel sorry when a human being dies. Zarqawi is a human being. He has a family that reacts as my family reacted when Nick was killed, and I feel bad about that”.10 Credit should be given to those agents who have worked against mimetic escalation.

Conversely, other collective and individual actors have moved from recognising the relational nature of violence—synonymous with maintaining violence within a regime of limits—to denying this relation and, consequently, to mimetic escalation. As far as collective actors are concerned, the most fascinating case is that of France. Throughout the 2000s, France exercised caution with regard to the military approach to Islamic “counter-terrorism” by participating in a secondary capacity in the war in Afghanistan and refraining from participating in the bombing and invasion of Iraq in 2003. This policy was then based on the conviction, expressed by a head of the DGSE before the Senate in 2010, that the war on “terrorism” fuels “terrorism”.11 Unlike the United States, the United Kingdom, and Spain, France was not the target of major Islamist attacks at the time. From 2013-2014, however, France adopted the warlike approach of “counter-terrorism” in the Sahel and, above all, in Iraq against the Islamic State organisation. The attacks of 2015 and beyond are a consequence of this shift. In my book, I show that “experts” presented these attacks as acts of aggression, even though they had declared, a decade earlier, that France's non-engagement in Iraq had spared it from numerous attacks. Some had even predicted, when France began bombing IS in August 2014, that this decision was likely to provoke a response from IS.

Precisely, you show that escalation was not inevitable, but that mimetic logics often prevailed. Why is that?

Girard thought that mimicry was part of human nature. This is one of the reasons why, toward the end of his life, he became enthusiastic about the biological theory of “mirror neurons”. He saw this theory as confirming his intuitions. My socio-historical approach is based on the notion of the social construction of reality. Through their discourse and practices, social actors produce norms and representations that may or may not lead them to respond to violence with violence. Like all social constructions, the one documented in my book, the escalation of “terrorist” and “counter-terrorist” violence, was unnecessary. Moreover, withdrawals from Iraq, Afghanistan, and the Sahel have led to a de-escalation since 2020. Over the previous two decades, a mimetic logic had prevailed because powerful actors (officials, experts, journalists, etc.) had denied the relational nature of the two forms of violence. They did so not in bad faith, but for reasons that made sense in their respective social fields.

How do you deal with this in the book?

In the second part of the book, I document four forms of this negation. I have already mentioned the first: the representation of the other as a radically different being (”infidel” or “fanatic”). The second dimension is the simulation of reality in Baudrillard's terminology.12 Supporters of the “global war on terror” and armed jihad have constructed imaginary worlds in which, to paraphrase Baudrillard, their own war has not taken place.13 This was the case in the media and on social networks when “surgical strikes” responded to “martyrdom attacks” and vice versa. These symmetrical euphemisms of violence sow the seeds of mimetic rivalry, encouraging their audiences to perceive only one form of violence worthy of the name: that of the other.

The third dimension is time. Girard notes that revenge is often "postponed". When it jumps a generation, as in certain vendettas or in Franco-German relations between 1870 and 1939, the children of former perpetrators have good reason to see themselves as innocent: they are. While jihadists have often practised delayed revenge against the descendants of the “crusaders”, the advocates of the war on “terrorism” have adopted an inversely symmetrical attitude: they have practised anticipatory violence in the context of preventive wars and, at the tactical level, in so-called “signature strikes”, which are not directed against persons identified as combatants or “terrorists”, but against persons suspected of being capable of committing combatant or “terrorist” acts in the future.

The French air strike that killed 19 civilians in the village of Bounty in Mali on 3 January 2021 illustrates this trend, which is also found in the public justifications for torture, of using violence to prevent future violence. Since the perpetrators of the future are as innocent in the present as the children of yesterday's perpetrators, these alleged perpetrators feel that they are the victims of an attack and may be tempted to respond to violence with violence. This inverted symmetry of temporal relationships has, incidentally, contributed to the representation of the other as a radically different being, either because they lived in the Middle Ages or because they use ungodly modern technologies.

The fourth dimension of mimetic rivalry is criminal law. Girard would surely say that an international criminal justice system accepted as legitimate by all parties could have prevented this escalation. This would probably have been the case, for example, if the leaders of al-Qaeda and President Bill Clinton had been tried for the attacks on the US embassies in Kenya and Tanzania in August 1998, and for the bombings in Afghanistan and Sudan. In the absence of such a system, criminal®law produced mimicry as both sides portrayed each other as criminals. Their potestas—their legal force—was thus able to respond to the violentia—the criminal violence—of the other.

What was your methodology?

I studied the social construction of mimetic rivalry in four social fields: politics, journalism, the military, and the judiciary. I show that they functioned autonomously, in line with Bourdieu's theory of social fields, while being structured by very similar mimetic logics. I use the example ofLe Monde's coverage of the relationship between France and the Islamic State organisation. Before the 2015 attacks, many articles in the daily newspaper emphasised the relational nature of “terrorist” and “counter-terrorist” violence. For example, Le Monde gave column space to Dominique de Villepin, who warned the French authorities of the risk of engaging with IS in Iraq: “To go to war against this devastated region, torn by identity crises [...], is to run the risk of crystallising and uniting a number of forces [...] against us”.14 After 2015, the discourse changes and we can read in the articles published by Le Monde that France has been attacked and has virtually no choice but to retaliate.

The denial of the relational nature of violence culminated in the assassination of Father Hamel on 26 July 2016. A columnist for Le Monde wrote: "We are not targeted arbitrarily, but because of who we are. We are not being attacked because we are part of the coalition fighting IS in Iraq and Syria: France joined only after being attacked”.15 The last sentence of this extract—“France joined [the coalition fighting IS] only after being attacked”—was factually incorrect. IS carried out its first attack on a French citizen, the mountaineer Hervé Gourdel, on 23 September 2014, four days after the start of the French air war. It is interesting to note that François Hollande also got bogged down in dates during his testimony at the trial of the accomplices of the November 2015 attacks.16 I do not interpret the Le Monde editorial and François Hollande's statement as propaganda, but as signs of the power of imagination they have generated in their respective social fields. In the symbolic worlds of the “global war on terror”, as well as in that of armed jihad, the community (whether liberal or Islamic) has been attacked. This premise takes precedence over all chronologies.

I have proceeded in the same way for each social field: I analysed the significant discourses and practices. On the jihadist side, for example, I look at the practice of executing a prisoner dressed in an orange jumpsuit. This macabre ritual expresses the idea that jihadist violence is a response to a primary violence: that of the sadistic “infidels” who tortured and humiliated “good” Muslims in the prisons of Guantanamo and Abu Ghraib. The political and military leaders of NATO countries also perform rituals that tell such stories. Every year, for example, they commemorate the 11 September 2001 attacks in New York. The ceremony takes place at NATO's Integrated Command Headquarters in Brussels in front of a memorial made from the rubble of the Twin Towers. The memorial is called "Article 5". It refers to the famous article of NATO's founding treaty which states that “an armed attack against one or more of them in Europe or North America shall be considered an attack against all of them”. This article has only been invoked once: after the attacks of 11 September 2001. In NATO, these attacks are interpreted as the starting point of the global war on “terrorism”, because in the imaginary constructed by this organisation, the air strikes on Afghanistan and Sudan in 1998 do not constitute acts of war.

Although I study both forms of violence (“counter-terrorist” and “terrorist”), I devote many more pages to the former than to the latter. This reflects the fact that counter-war “terrorism”, particularly in the United States, claims many more innocent lives than “terrorism” itself. For the Iraq war of 2003-2005, for example, estimates range from 40,000 to 600,000 civilian casualties, depending on whether one counts direct or indirect deaths.17 “Terrorism” can be very violent, as in New York in 2001 or Paris in 2015, but not on the same scale. This asymmetry in the treatment of these two forms of violence in my book also reflects a practical constraint: I was able to conduct research based on archives and interviews in the US and French militaries, but not with jihadists. For them, I relied on the work of specialists such as Jean-Pierre Filiu and on the research of journalists embedded in these organisations.

In the conclusion of the book, you write that mimetic rivalry has often prevailed over strategic thinking. What do you mean by this?

Strategic thinking, which should not be mistaken for the unreflective field of strategic studies, consists of thinking about the best way to achieve one's objective. That means understanding your opponent's motives. One then automatically realises, to use a phrase of René Girard, that "aggression does not exist. [...] The aggressor has always already been aggressed”.18 Strategic reflection is therefore a powerful antidote to mimetic rivalry. It can help to break the vicious circle of revenge, as in the case of Spain in 2004. It can also lead to the continuation of the war, while keeping it within limits. Every year, the President of the United States receives a document called the National Intelligence Estimate, which summarises the annual reports of the various intelligence services. The 2006 edition stated that the bombing and invasion of Afghanistan and Iraq had led to an increase in terrorist activity. The authors of this document called for greater discretion in the use of force. This allowed for a brief de-escalation until the intensification of drone strikes at the turn of the 2000s and 2010s. At a time when the sound of marching boots is becoming deafening, this book can also be read as a call to reflect on the often counterproductive nature of using force to stop violence.

Interview by Miriam Périer, CERI.

illustration : Artwork by Felix Del Marle. Public Domain.

• 1. Raymond Aron, Peace and War. A Theory of International Relations, London & New York, Routledge, 2017, p. 196.
• 2. Isabelle Sommier, Le Terrorisme, Paris, Flammarion, 2000.
• 3. Anonymous, “Working Group Report on Detainee Interrogations in the Global War on Terrorism, Torture Documents (The Rendition Project)”, 4 April 2003, p. 65.
• 4. Rebecca Mihnot-Mahdavi, Drones and International Law. A Techno-Legal Machinery, Cambridge, Cambridge University Press, 2023.
• 5. Simon Collard-Wexler, Constantino Pischedda, and Michael G. Smith, “Do Foreign Occupations Cause Suicide Attacks?”, Journal of Conflict Resolution, 58(4), 2013, pp. 625-657.
• 6. René Girard, Violence and the Sacred. Translated by Patrick Gregory. Baltimore: Johns Hopkins University Press, 1977 (La violence et le sacré, p. 463.)
• 7. René Girard, Violence and the Sacred. Translated by Patrick Gregory. Baltimore: Johns Hopkins University Press, 1977 (La violence et le sacré, op. cit., p. 216).
• 8. Valérie Rosoux, Les usages de la mémoire dans les relations internationales : le recours au passé dans la politique étrangère de la France à l'égard de l'Allemagne et de l'Algérie de 1962 à nos jours, Brussels, Edition Bruylant, 2002.
• 9. Germaine Tillon, Ennemis complémentaires. Guerre d'Algérie, Editions Tiresias, 2005.
• 10. “Beheaded Man's Father: Revenge Breeds Revenge”, cnn.com, 9 June 2006, see https://edition.cnn.com/2006/WORLD/meast/06/08/berg.interview/ . Accessed 17 October 2023.
• 11. Alain Chouet, “Al Qaida, Senate Committee on Foreign Affairs, Defence and the Armed Forces”, Conference of 29 January 2010, http://videos.senat.fr/video.22174_57b70c17f0a5f.le-moyen-orient-a-l-heure-nucleaire?timecode=4809000 (accessed on 12 March 2020). The sequence can be found here: 19'50‘’.
• 12. Jean Baudrillard, Simulacre et simulation, Paris, Editions Galilée, 1981.
• 13. Jean Baudrillard, La guerre du Golfe n'a pas eu lieu, Paris, Editions Galilée, 1991. Translated into English as The Gulf War did not take place, Indiana University Press, 1995.
• 14. “Villepin”, LeMonde.fr, 12 September 2014.
• 15. Jérôme Fenoglio, “Editorial: Attack on Saint-Etienne-du-Rouvray: Resisting the Strategy of Hate”, Le Monde, 28 July 2016. It should be noted that Le Monde corrected the electronic version of this editorial after the Arrêt sur image website pointed out the inversion of the chronology.
• 16. Emmanuel Carrère, V13. Chronique judiciaire, Paris, P.O.L, 2022, p. 127-128.
• 17. Gilbert Burnham, Riyadh Lafta, Shannon Doocy, and Les Roberts, "Mortality after the 2003 Invasion of Iraq: A Cross-Sectional Cluster Sample Survey", Lancet, 368, 2006, pp. 1421-28.
• 18. René Girard, Achever Clausewitz. (Entretiens avec Benoît Chantre), ed. by Carnets Nord. Paris. p. 53.