Start Submission Become a Reviewer

Reading: Legitimate Targets: What is the Applicable Legal Framework Governing the Use of Force in Rio...


A- A+
Alt. Display

Research Article

Legitimate Targets: What is the Applicable Legal Framework Governing the Use of Force in Rio de Janeiro?


Conor Foley

Pontifícia Universidade Católica do Rio de Janeiro (PUC Rio), BR
X close


The belief that the favelas of Rio de Janeiro constitute a ‘war zone’ is widely held in Brazilian society. The intensity of the violence and the level of organization of some of the armed actors are comparable to those in officially recognized non-international armed conflicts elsewhere in the world. As conflicts are increasingly fought in situations where it is challenging to distinguish civilians from combatants, the binary divisions of international law are failing to address the contemporary experience of violence in what has been termed the ‘new wars’. There is an emerging agreement in international jurisprudence and state practice that the provisions of international human rights law and international humanitarian law must be concurrently applied in officially recognised ‘war zones’. This article argues that this analysis could also be used in other situations of violence, such as Rio de Janeiro, but the ‘armed conflict’ rhetoric risks conferring legitimacy of the use of lethal force between its parties.

How to Cite: Foley, C., 2022. Legitimate Targets: What is the Applicable Legal Framework Governing the Use of Force in Rio de Janeiro?. Stability: International Journal of Security and Development, 10(1), p.5. DOI:
  Published on 16 Nov 2022
 Accepted on 27 Oct 2022            Submitted on 20 Sep 2021


In the early morning of 6 May 2021, about 200 heavily armed civil police officers supported by a helicopter and armoured vehicles raided Jacarezinho, an impoverished favela neighbourhood in Rio de Janeiro, killing 27 people. According to Human Rights Watch (HRW) several witnesses said police executed at least three suspects and officers apparently removed bodies from the crime scene to destroy evidence (HRW 2021). The head of the unit in charge denied that any executions had taken place, with civil police investigators later taking cursory statements from only a small number of the officers involved (Nogueira and Barbon 2021). The deputy chief of the civil police claimed that ‘Rio is much safer with these 27 criminals eliminated’ (HRW 2021).

Such sustained and widespread use of force by the police in the favela communities of Rio de Janeiro is not uncommon. For example, in May 2022, a similar operation in Vila Cruzeiro killed 23 people (HRW 2022), 18 were killed in another operation in July 2022 (Lopes 2022), and in November 2021, a similar police operation in São Gonçalo resulted in at least eight fatalities (Viga Gaer 2021). In June 2020, the Brazilian Supreme Court issued an order prohibiting such operations during the Covid-19 pandemic, except in ‘absolutely exceptional cases’. Between June and September 2020, there was an overall decline in police killings compared to the previous year (Instituto de Segurança Pública data). However, in September that year, Rio de Janeiro’s chief of civil police announced that operations were resuming because the level of violence made the situation in the state ‘exceptional’ (Araújo 2020). He also announced that he intended to request equipment from the army to conduct military-style raids in poor neighbourhoods in order to ensure ‘war superiority’. In October 2020, the number of police operations and killings started going up again. From January through April 2021, civil and military police killed 595 adults and children in Rio de Janeiro (HRW 2021).

At a news conference after the Jacarezinho raid, the chief of police operations said that ‘judicial activism’ had made it more difficult for the police to act in some areas and claimed that drug trafficking organizations had become stronger as a result (Betim 2021). This contrasts with a ruling by the Inter-American Court of Human Rights (IACtHR) that condemned Brazil for failing to carry out proper judicial investigations into similar police operations that had occurred in the 1990s. The Court noted a lack of impartiality in judicial investigations and that ‘before investigating and corroborating the police conduct, in many of the investigations an investigation is carried out regarding the profile of the deceased victim. If it is considered that he was a possible criminal the investigation is closed’ (IACtHR 2017).

The belief that the favelas of Rio de Janeiro constitute a ‘war zone’ is widely held in Brazilian society. This implies that both the intensity of the violence and the level of organization of some of the armed non-state actors (ANSAs) are comparable to those in officially recognized non-international armed conflicts (NIACs) elsewhere in the world. The Brazilian state has also significantly increased the role of its military forces in policing operations through a series of decrees and operations such as Complementary Law number 117, the Garantia da Lei e da Ordem (Guarantee of Law and Order) and the attempted ‘pacification’ of favela neighbourhoods (Azzi forthcoming) The debate about public security has been reframed and militarized in the process and ‘juridically replaced by the term defence’ (Kenkel 2006).

In 2018, Brazil elected as its president Jair Bolsonaro, a former military officer and outspoken supporter of the country’s previous dictatorship (Foley 2019). In one of the first interviews Bolsonaro gave after the election he was asked about policing in Rio de Janeiro to which he responded:

The way the Brazilian Army engaged in Haiti was exactly what we need. They considered anyone armed with a rifle to be a legitimate target. What we see in Rio is too much good will towards people carrying firearms. How should we deal with these kinds of people? It won’t be with bouquets or calls to ‘hand over your weapons’. This is an urban area. The collateral damage caused by a firefight would be disastrous. I maintain that what the police or Armed Forces need in a GLO operation is a legal safeguard. They shouldn’t have to worry about being prosecuted or convicted for carrying out their mission. (Rothenburg 2018)

This view was also espoused by then-Governor of the State of Rio de Janeiro, Wilson Witzel, a former federal judge. Days after his election in October 2018, he gave an interview claiming that it was not necessary for police officers to wait for an imminent threat in order to shoot ‘suspects’: ‘The correct course of action is to kill any criminal who is armed with a rifle. The police will do the right thing: aim at the head and… fire! So no mistakes there’ (Veja 2018). Witzel subsequently posted a video of himself on social media inside a helicopter gunship, hovering over a densely populated area, standing beside a police sniper who was aiming his sights at the favela, saying ‘let’s put an end to these criminals’ (Grellet 2019). In the Brazilian elections of October 2022, one of the candidates, Allan Turnowski, a police chief who had commanded the Jacarezinho operation, used the number 27 to identify himself on the ballot and repeatedly stressed this as an example of a ‘zero tolerance’ of crime approach (Alves and Otavio 2022).

The extremely high levels of violence in such concentrated areas and the fact that the vast majority of victims are poor, black and mulatto favela residents are factors that have led many progressive commentators to also draw on metaphors more usually used in war zones. For example, in 2015 Amnesty International Brazil’s then Director warned of a ‘policy of extermination of young black people’ (Carta Capital 2014), while in 2016 a Brazilian Parliamentary Commission compared the level of violence to a ‘genocide’ and an ‘undeclared civil war’ (CPIADJ, 2015).

But most international lawyers and legal scholars, as well as human rights and humanitarian activists, reject this characterization. For example, the International Committee of the Red Cross (ICRC) has stated: ‘What we’re seeing in Rio de Janeiro is urban armed violence, which doesn’t amount to an armed conflict for the purposes of international humanitarian law despite armed clashes between drug traffickers and the militia, police and armed forces’ (ICRC 2011). In 2019, the Geneva Academy’s annual War Report similarly concluded that: ‘The legal classification of the ongoing violence between drug gangs and police forces in the favelas (slums) of Rio de Janeiro, Brazil, is debatable. However, despite the intensity of the violence in the city…the armed groups involved (gangs and militias) do not meet the requisite organization criteria.’ (Bellal 2018, p. 73).

Underlying such debates is a concern about the consequences that such a definition could have on the international legal framework surrounding the use of force. This must either be found in the jus in bello provisions of International Humanitarian Law (IHL) or the regulations on the use of force contained in International Human Rights Law (IHRL) (Foley 2017, 2022). IHL prohibits attacks on civilians and civilian objects, while permitting combatants in an international armed conflict to directly engage in hostilities without this being considered a criminal act. It permits troops to launch a surprise attack on an enemy military base even if this involves ‘collateral damage’ to civilians and civilian objects, subject to the necessary precautions and proportional to the military benefit. A soldier may also shoot an enemy soldier, so long as the latter is not hors de combat, even if they are unarmed and do not pose an ‘immediate threat’ at that particular point.

IHRL does not distinguish between ‘civilians’ and ‘combatants’ in the same manner as IHL, and the proportionality and precautionary principles apply to the target of the use of force as well as to any other victims. It would, for example, be a violation of IHRL to use lethal force against a criminal suspect if there was a non-lethal means of apprehending them, or to do so in relation to any but the most serious of crimes. Warnings should also be given before lethal force is used to allow the suspect an opportunity to surrender. IHRL requires an effective investigation into the circumstances surrounding the use of lethal force in all circumstances. It also contains extensive details about the treatment of people in custody, including protection against ill-treatment and their right to a fair trial.

It is widely agreed that the traditional paradigm by which IHRL governs relations between states and their own citizens in times of peace, whilst IHL primarily regulates the conduct of international armed conflicts, is outdated. Since IHRL ‘does not cease in times of war’ (International Court of Justice 1996), it is also increasingly recognized that both legal frameworks apply concurrently during situations of conflict. It has also been argued that the existing international legal framework governing the use of lethal force rests on outdated binaries – such as ‘war and peace’ – that were largely constructed in the state-centred conceptions of the nineteenth and twentieth centuries. This fails to address the contemporary experience of what Chinkin and Kaldor (2017) term ‘new wars’. They argue that these could equally be described as ‘criminal enterprises, banditry, terrorism or massive violations of human rights’ and need to be addressed by a new Human Security model. While much of the discussion about the ‘new wars’ has focussed on violence in places that are widely recognized to be conflict zones, a case can be made that this analysis could also apply to other situations of violence, such as Rio de Janeiro, which is largely neglected in the current literature.

This article will address the question of whether the violence in Rio de Janeiro fits the legal definition of a non-intentional armed conflict (NIAC). It will then discuss who the armed non-state actors (ANSAs) would be, and whether their activity meets the threshold for this legal characterization. The article will also briefly consider how IHL and IHRL have been concurrently applied in other conflict zones. We argue that there is an emerging consensus in both international law and practice that, as conflicts are increasingly fought in situations where it is challenging to distinguish civilians from ANSA members, the ‘minimum protections’ of IHL will always need to be supplemented by those of IHRL. Far from legitimizing operations such as those carried out in Jacarezinho, the ‘new wars’ currently being fought around the world show the legal and practical futility of this approach.

What is a non-international armed conflict?

An armed conflict has been defined as existing whenever ‘there is resort to armed force between states or protracted armed violence between governmental authorities and organized armed groups or between such groups within a state’ (ICTY 1995). There is an obvious difference between these two thresholds and IHL will only apply in NIACs when the second one has been reached. The four Geneva Conventions and Additional Protocol I all apply to international armed conflicts. Common Article 3 of these Conventions and Additional Protocol II apply to NIACs. The treaty provisions relating to IHL in NIACs are much less extensive than those relating to international armed conflicts but, have been ‘enriched and upgraded’ by decisions of the UN Security Council and the case law of international criminal tribunals (Kolb 2003). The International Criminal Tribunal for Yugoslavia (ICTY) and the ICRC have also stated that most of the treaty provisions governing IHL in international conflicts can also be considered customary law in NIACs (Henckaerts and Doswald-Beck 2005; ICTY 1995). The case law of the International Criminal Court (ICC) has also provided considerable detail on the application of IHL in NIACs.

The drafting history of the Geneva Conventions shows that the decision to define NIACs negatively rather than positively was a conscious one, based on concerns about national sovereignty. There was no consensus during the negotiations on what factual situations constituted NIACs (Sivakumaran 2012). States also did not want to grant the status of combatants to groups that they considered to be terrorists or criminals (Sassòli 2019). Article One of Additional Protocol II of the Geneva Conventions, which regulates conduct in NIACs, specifies that it ‘shall not apply to situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature, as not being armed conflicts’.

The closest that IHL treaty law comes to defining the conditions in which a NIAC might be said to exist are set out in the same article of Additional Protocol II. This states that its provisions are applicable ‘to all armed conflicts which […] take place in the territory of a High Contracting Party between its armed forces and dissident armed forces or other organized armed groups which, under responsible command, exercise such control over a part of its territory as to enable them to carry out sustained and concerted military operations [emphasis added] and to implement this Protocol’. The ICTY has ruled that the term ‘protracted armed violence’ refers to intensity rather than duration, and that other factors such as ‘the type of weapons and other military equipment used; the number and calibre of munitions fired; the number of persons and type of forces partaking in the fighting; the number of casualties; the extent of material destruction; and the number of civilians fleeing combat zones. The involvement of the UN Security Council may also be a reflection of the intensity of a conflict’ (ICTY 2008a).

The ICRC (2008) has noted that one indicator of the ‘intensity threshold’ is a level of violence that obliges the government to use military forces, ‘instead of mere police forces’. In the La Tablada ruling, the Inter-American Commission on Human Rights listed the involvement of the state’s armed forces as an indicative factor of the existence of a NIAC (IACHR 2008). In Limaj (ICTY 2005), the ICTY also noted that the deployment of heavily equipped armed forces could be seen as indicative, but not the determining factor in such assessment, as a state may also deploy its armed forces in support of the civil authorities in situations of internal tensions or disturbances.

In Lubanga (ICC 2007), the International Criminal Court stated that an ANSA must be able to sustain military operations for a period of time long enough to procure, distribute and use weapons as well as to carry out some recruitment and training. In Katanga (ICC 2008) the Court recognized that for an ANSA to have a ‘responsible command’ it needs to have a degree of organization and discipline sufficient both to carry out sustained and concerted military as well as tactical and strategic coordination and planning to enable it to speak with one voice and participate in negotiations. This does not necessarily mean a rigid hierarchical structure or a single leader, but it does presuppose some degree of organizational command that enables individual commanders to enforce the provisions of IHL. Some argue that, for a NIAC to exist, the ANSAs need to be fighting for a discernible political cause, while others warn that this criterion may be too subjective (Bruderlein 2000; Hauck and Peterke 2010; Vité 2016). Neither the texts of the Conventions and Protocols nor the ICRC’s interpretive studies mention political motivation as being of any relevance to the determination of the applicability of IHL. In the Haradinaj case, ICTY stressed a more objective emphasis on the level of an ANSA’s organizational coherence as indicative or not of the existence of a NIAC:

the existence of a command structure and disciplinary rules and mechanisms within the group; the existence of a headquarters; the fact that the group controls a certain territory; the ability of the group to gain access to weapons, other military equipment, recruits and military training; its ability to plan, coordinate and carry out military operations, including troop movements and logistics; its ability to define a unified military strategy and use military tactics; and its ability to speak with one voice and negotiate and conclude agreements such as cease-fire or peace accords. (ICTY 2008a)

Does the violence in Rio de Janeiro fit the paradigm of a non-international armed conflict?

A ‘high-intensity conflict’ has been defined as one with ‘well over 1,000 lives per year’ (ECP 2022). In 2021, over 40,000 people lost their lives in Afghanistan, which was the highest death toll, while more than 22,000 died in Yemen. (ibid) According to the Center for Civilians in Conflict (CIVIC), which only measures civilian deaths, the ‘most intense conflicts in the world’ were Afghanistan, Central African Republic, Iraq, Libya, Nigeria, Somalia, South Sudan, Syria, Ukraine, and Yemen, with over 20,000 civilian casualties occurring between them (CIVIC 2019a). The conflict in Afghanistan was the deadliest of these, with around 3,000 civilians killed and 7,000 injured almost every year since 2014, which was when the UN started recording figures (UNAMA 2021). Approximately 3,000 civilians were killed in the terrorist attacks by Al Qaeda in the United States on 11 September 2001, which led to the United States and its allies invoking its ‘inherent right to self-defence’ at the UN Security Council to justify military intervention in Afghanistan.

There have been between 50,000 and 60,000 homicides each year in Brazil over the same period (FBSP 2021, pp.17–45). The state of Rio de Janeiro registers around 4,000 to 5,000 homicides a year, likely an underestimate (FBSP 2019, pp.119–29). Although there are other Brazilian states where the death rates are higher, the city of Rio de Janeiro has a particular symbolism in the debates about violence, policing and militarization. The metropolitan area of the city has an estimated population of 12.7 million inhabitants, of which around two million live in favelas. Most of these are under the physical control of organised criminal groups and much of the violence is concentrated in these communities, making them amongst the most violent places in the world (Gonçalves 2017). If one of the indicators of what constitutes an armed conflict is intensity (ICTY 2008) then the communities in Rio de Janeiro in which the violence is concentrated to appear to have reached this in threshold terms. Rio’s police and public security forces killed almost 7,000 people between 2016 and 2020 (FBSP 2021, pp.57, 67–68) The police routinely claim that the majority of those that they kill are criminals with links to the heavily armed narco-traffic gangs that dominate Rio’s drug trade, although they do accept that stray bullets sometimes claim ‘innocent’ victims.

The violence in Rio involves a multiplicity of actors, here broken down in three categories: state forces, militias, and drug factions. The main state forces are the military and civil police, which act under the authority of the state governor, and the national army, which reports to the Brazilian government’s Ministry of Defence. The military police are technically reserve army officers, acting under military hierarchy and subject to military jurisdiction. They are responsible for conducting street patrols and upholding public order, while the civil police are responsible for investigating specific crimes. The command structures of Brazil’s police and army are largely unchanged since the military dictatorship that governed the country from the mid-1960s to the mid-1980s.

From 2001 onwards the Brazilian state has enacted a number of decrees establishing guidelines for the deployment of the military in public security operations. In 2013, this terminology was systematized by the Ministry of Defence in a Manual for the Guarantee of Law and Order. In 2014, a revised version of this manual classified Law and Order Operations (Op GLOs) as ‘non-war military operations’ (Brazilian Government Ministry of Defence 2012, 2013, 2014). Azzi (forthcoming) has noted that of a total of 23 urban violence-type Op GLOs deployed throughout Brazil between 2010 and 2020, at least nine were in Rio and these have placed parts of the city under military occupation for an equivalent of four years and four months. As will be discussed below, policing operations in Rio are also often carried out in a highly militarized manner.

The militias were initially formed as vigilante groups by serving and former police officers and other uniformed personnel, purportedly to expel the drug traffickers from particular communities (Manso 2020). However, as Cano (2008) has noted, their main motivation seems to have been the realization that they could make more money controlling territories than receiving bribes from traffickers. The militias have evolved into powerful criminal organizations with close links to corrupt politicians and some state institutions. They run protection rackets, ‘tax’ residents for basic services such as water, electricity and gas, and have become significant actors in the city’s real estate market (Dalby and Francisco 2019). According to the report of an official inquiry, up to a third of the members of Rio de Janeiro’s local parliament have connections with the militia (ALERJ 2008). One of President Jair Bolsonaro’s sons has documented links with one militia group, which is suspected of having carried out the 2018 assassination of Marielle Franco, a left-wing city councillor (Estado de Minas 2020).

The drug factions (facções) were initially formed in the state’s prisons, partly to protest for better treatment and against state brutality, but also to physically control the sale and distribution of drugs. The first drug faction, the Comando Vermelho (Red Command) drew some inspiration from political prisoners, under the dictatorship in the late 1960s (Marinho 2019.) By the end of the 1970s, the Comando Vermelho began to organize criminal activity – primarily bank robberies and kidnappings – from inside the prisons, a pattern which has continued to this day (Leeds 1996; 2007). Affiliated members of the Comando Vermelho began organizing their favela territories for drug sales within a loose structure of mutual support. Hierarchically structured quadrilhas (squads) were established to defend sales points and the surrounding communities from police invasion or attack from other gangs. From 1986 onwards, the Comando Vermelho began to fragment internally and by the mid-1990s, three rival factions had emerged within Rio de Janeiro: Comando Vermelho; Terceiro Comando (Third Command); and Amigos dos Amigos (Friends of Friends) (Alves and Evanson 2011).

The militias and rival drug trafficking gangs have fought bitter conflicts for control over the favelas that function as their headquarters and territorial bases. Each favela controlled by the drug traffickers is ‘owned’ by a dono who enforces a strict code of conduct upon their subordinates (Lyra 2020). The donos assign gerentes (managers) to organise drug trade activities within that territory. The vapores are the salespeople, olheiros and fogueteiros are look-outs tasked with raising the alarm of a possible police incursion, while the soldados are responsible for defending the territory against the police or rival factions (Hirata and Grillo 2017). A faction is composed of multiple allied donos, acting largely independently in a horizontal association for mutual protection, with the dono of the most profitable favela being considered the faction’s leader.

Dowdney estimated that there were around 10,000 young people involved in Rio de Janeiro’s drug trade in the 1990s. He noted that the ‘utilization of high-powered weapons and the types of armed violence caused by inter-faction disputes and confrontations between the police and factions’ mean that ‘stark similarities exist between children employed in [the city’s] drug factions and “child soldiers” in almost every functional and definitive aspect’ (Dowdney 2003, p.13). The intensity of violence in Rio de Janeiro has fluctuated over the years, but data compiled by civil society groups for the period 2016–18 show that the police carried out 1,000–1,200 operations per year, with individual confrontations commonly lasting more than 12 hours, and happening multiple times a week, often days in a row (Hirata and Grillo 2019; Olliveira et al 2018). Although the state does not publicize the exact number of munitions fired during confrontations, a local datalab that relies on community reporting received information of 2,202 shootings in 2016; 5,443 in 2017; 9,634 in 2018; and 7,368 in 2019 giving an average of 16 per day (Fogo Cruzado 2020).

Between 2008 and 2014 an attempt was made to ‘pacify’ the favelas using tactics based heavily on counter-insurgency operations elsewhere in the world (Foley 2014). This process culminated in November 2010 with the ‘retaking’ of the Complexo do Alemão, the largest favela conglomerate in the city. A combined force of 2,700 soldiers and police, aided by air force attack helicopters, navy marines, armoured cars, tanks, high velocity weapons, and elite special forces launched an all-out assault to capture the territory, where they found over 40 tons of drugs and a vast arsenal of weapons. Over the previous week the factions had attacked police stations with grenades and automatic fire, burned buses and cars, blocked streets with barricades and engaged the police in gun battles. They had used similar tactics several times before. In 2009, for example, they shot down a police helicopter, while in 2006 they mounted a series of attacks in the city, killing 18 people, to enforce a city-wide shutdown that included the closure of shops, schools and hospitals.

The ‘pacification’ strategy was eventually abandoned, but the Brazilian military are still regularly deployed in favelas. The violence has continued, and the drug trafficking factions have shown their ability to plan, coordinate and sustain campaigns of violence, recruit and train members as well as supplying them with military weapons and equipment. In 2016, for example, the police apprehended 9,017 firearms, including 371 rifles, 81 machine and submachine guns, 3,835 pistols and 3,613 revolvers, in addition to explosives such as 603 grenades (FIRJAN 2017). In 2017, videos widely circulated on social media showed drug trafficking groups armed with heavy machine guns and rocket-propelled grenades (Veja 2017). A police helicopter was allegedly shot down by a drug faction during confrontations in Cidade de Deus in 2016 (BBC 2016). State forces have also long relied on the use of large, armoured cars (known as ‘caveirões’ -- big skulls), when conducting operations in favelas controlled by the drug traffic. In 2018 there was also a 200 per cent increase in the use of ‘caveirão voador’ (flying skulls), armoured helicopters from which the police officers fire with automatic weapons into favela communities (Neves 2019).

It has been estimated that around 850 of Rio’s 1,025 favelas are under the control of either drug trafficking or militia groups (Observatório Legislativo). In some cases this physical control is visible because their entrances are blocked with barricades or patrolled by soldados, olheiros and fogueteiros. In other cases, the control is more subtle, although it is reinforced by the ever-present violence against anyone perceived as a threat by the dominant armed group. Members of a rival group or faction who tried to enter such territory without permission would almost certainly be killed. According to the Geneva Academy: ‘It is not uncommon for armed groups to hire former police or military officers as military advisors, thus meeting to some extent the level of organization required to coordinate military offences and defence’ (Bellal 2019, p.78). Peterke (2012) also argues even if all of the characteristics laid out by case law are not fulfilled, a ‘convincing combination of these indicators’ may be sufficient to trigger IHL, although such a claim cannot be made definitively at present.

From the above overview, a case can clearly be made that the level of violence in the favelas of Rio de Janeiro has at times reached the intensity required of a NIAC and that at least some of the ANSAs operating in these communities have a sufficient degree of organization and responsible command structure to apply IHL. They also have physical control of certain territory, the ability to gain access to weapons, recruits and carry out training. Some have also shown their ability to plan, coordinate and carry out ‘military’ operations, use ‘military’ tactics, and formulate ‘political’ demands. However, in international legal terms, the Brazilian state continues to exert both de jure and de facto control over its entire territory. Even if the violence in Rio de Janeiro’s favelas were to be determined as an NIAC, IHRL would continue to fully apply, at least concurrently with IHL. The next section of this article will briefly consider the concurrent applicability of these two international legal frameworks in recognised NIACs.

The concurrent applicability of IHL and IHRL

Recent decades have seen rhetoric about a ‘war on terror’ and a ‘war on drugs’ used to justify a more militarized response to what had previously been considered essentially law enforcement issues. This contrasts with a previous reticence by some states to publicly admit that the level of violence that they were suffering may have reached the threshold of what constituted a NIAC (Reidy, Hampson and Boyle 1997; Reidy 1998). For example, the European Court of Human Rights (ECtHR) has acknowledged its ‘reluctance’ to label situations as armed conflicts in IHL terms, in order ‘to avoid unnecessary controversy – especially where the States parties do not themselves qualify the situation as an armed conflict’ (Chernishova 2013, p.90). This has enabled a number of international courts to establish considerable jurisprudence on alleged violations of IHRL in conflict-related situations.

The first of these cases was McCann and Others v UK (ECtHR 1995), where the Court narrowly ruled that the overall planning of an anti-terrorist operation that killed three members of the Irish Republican Army (IRA) in Gibraltar had resulted in a violation of their right to life, while absolving the soldiers of direct responsibility. The United Kingdom (UK) had twice previously derogated from some provisions of the European Convention owing to the state of emergency which it claims existed in Northern Ireland and both derogations had been upheld by the Court -- see the cases Ireland v UK (ECtHR 1978), Brogan v UK (ECtHR 1988) and Brannigan and MacBride v UK (ECtHR 1993).

Hampson (2008) has argued that the level of violence in Northern Ireland may have reached the threshold level of a NIAC in the early 1970s and one of the three IRA members killed in Gibraltar had previously led protests against the withdrawal of ‘prisoner of war’ status for imprisoned IRA members. Gearty (1995) reasons that the Court found a violation, without explicitly accepting that the British government operated an undeclared shoot-to-kill policy based on a de facto acceptance of an armed conflict between the IRA and its security forces, by blaming a ‘failure of overall planning’, which breached the ‘positive obligations’ in IHRL.

However, in more recent cases international courts have been prepared to apply IHRL in situations of armed conflict where IHL rules were also potentially applicable. For example, in Kaya v. Turkey (ECtHR 1998a) the ECtHR stated that ‘[n]either the prevalence of violent armed clashes nor the high incidence of fatalities can displace the obligation under Article 2 to ensure that an effective, independent investigation is conducted into deaths arising out of clashes involving the security forces’. In Isayeva, Yusupova and Bazayeva v. Russia (ECtHR 2005a), the Court found a violation of IHRL while accepting that ‘the situation that existed in Chechnya at the relevant time called for exceptional measures … to suppress the illegal armed insurgency … includ[ing] employment of military aviation equipped with heavy combat weapons’.

The United Nations Human Rights Committee first ruled that disproportionate force had been used against unarmed guerrillas in Colombia, in the Guerrero case, although it is widely accepted that there was a NIAC in the country at the time (UNHRC 1982). It has reiterated its concerns under IHRL in relation to Israel/Palestine, the Philippines, Uganda, the Democratic Republic of Congo and Sudan. Larsen (2012, p.288) has noted that, as well as holding these states responsible for their obligations under IHRL, even though they had no authority in parts of their own national territory due to armed conflicts, the HRC ‘takes IHL into consideration when making observations about state compliance with the Covenant only to a limited extent, if at all’.

Hampson and Lubell (2013) have noted that international human rights monitoring bodies only have the competence to find a violation within the IHRL legal framework, but their jurisdiction is not displaced during an armed conflict. Where an alleged violation clearly falls within the applicability of IHL rules, they argue that ‘a human rights body has two choices’. It must either apply IHRL through the lens of IHL or it must blend the two bodies of law together, given that IHL contains guidance on issues such as necessary precautions when carrying out attacks on military targets or the rules governing aerial bombardment, which IHRL is not equipped to provide. They conclude that while there is ‘no general, top-down principle which can be applied to establish if an issue should be handled one way or another’, the issues involving the conduct of hostilities appear to be more appropriate for determination through IHL, while issues involving the protection of victims are more likely to involve ‘a blend’ of the two bodies of law.

In Neira Alegria v. Peru (IACtHR 1995), the Inter-American Court of Human Rights ruled that the authorities had acted disproportionately in demolishing a prison during the course of a riot, basing its decision solely on IHRL, even though most of the detainees who were killed were members of a rebel group, involved in a NIAC. On the other hand, in La Tablada (IACHR 2008), the Inter-American Commission on Human Rights observed that:

the American Convention contains no rules that either define or distinguish civilians from combatants and other military targets, much less specify when a civilian can be lawfully attacked or when civilian casualties are a lawful consequence of military operations. Therefore, the Commission must necessarily look to and apply definitional standards and relevant rules of humanitarian law as sources of authoritative guidance in its resolution of this and other kinds of claims alleging violations of the American Convention in combat situations.

In Bámaca Velásquez v. Guatemala (IACtHR 2000) the Court stated that, although it lacked competence to declare that a state was responsible for the violation of a treaty over which it had no jurisdiction, it could ‘observe that certain acts or omissions that violate human rights […] also violate other international instruments for the protection of the individual, such as the 1949 Geneva Conventions and, in particular, common Article 3’. In Franklin Guillermo Aisalla Molina (IACHR 2010), the Commission noted that IHRL and IHL ‘share a common core of non-derogable rights and the mutual goal of protecting the physical integrity and dignity inherent in the human being’ and that they ‘may influence and reinforce each other’.

Overlapping paradigms and effective control

Many of the recent cases at the ECtHR have related to the responsibilities of European states participating in military operations in Afghanistan and Iraq. In Jaloud v. The Netherlands (ECtHR 2014b), the Court found a violation of the right to life due to a failure to conduct an adequate investigation after Dutch soldiers killed a man at a checkpoint in Iraq in 2004. In Al Skeini v UK (2011a), it found violations after five civilians were shot dead by British soldiers and one beaten to death in British custody in 2003. In Al-Jedda v UK (2011b), it found a violation of the right to liberty after the applicant’s detention without trial for several years. In Hassan v. UK (ECtHR 2014a), it found that, although the applicant’s detention by the British army in Iraq brought him within the UK’s extra-territorial jurisdiction, the otherwise unauthorized detention of suspected combatants was in compliance with IHL provisions in the context of international armed conflict. In Hanan v Germany (ECtHR 2021), which concerned a 2009 NATO airstrike in Afghanistan that killed 100 civilians, the Court did not find a violation on the merits but did accept that it had jurisdiction to hear the case.

In the above cases, the Court was called upon to determine the degree of ‘effective control’ that the UK, Germany and the Netherlands exerted for the purpose exercising extra-territorial jurisdiction over particular people and places. As the ‘occupying power’ in Iraq, the UK was clearly bound by IHL, but the Court found IHRL to be applicable as well. As Gross (2017, p.245) has noted:

Under a law enforcement model of occupation, which assumes the occupying power’s relatively secure hold on the territory, lethal force can only be used under very strict circumstances in order to protect life. Under a belligerency model, however, occupying forces are permitted to attack enemy combatants as well as civilians directly participating in the fighting. Civilian losses are not prohibited either so long as they are not targeted and as long as they fall within the restrictions of jus in bello, particularly that of proportionality.

The favelas of Rio de Janeiro are clearly within the jurisdiction, national territory, and effective control of the Brazilian state. While the Brazilian security forces only ever enter intermittently in heavily militarised operations that feel like those of an occupying army, this does not absolve the state of its responsibility to secure for everyone the full freedoms and protections guaranteed by IHRL. Even if the violence has reached a level of intensity of a NIAC and some of its various ANSAs do have sufficiently ‘responsible command’ structures to enable them to be considered capable of becoming parties to a conflict, this would not mean that their individual members were legitimate targets under IHL. As Lubell (2010) observes, ANSAs participating in a NIAC will almost certainly be considered criminals under their own domestic law and so subject to its legal provisions, assuming that these are functioning.

According to the ICRC, ‘practice is ambiguous’ as to whether members of an ANSA ‘are considered members of armed forces or civilians’ (ICRC Customary IHL Rule 3, Definition of Combatants). It notes that persons taking ‘direct part in hostilities’ in NIACs ‘are sometimes labelled “combatants” […] However, this designation is only used in its generic meaning and indicates that these persons do not enjoy the protection against attack accorded to civilians, but this does not imply a right to combatant or prisoner-of-war status, as applicable in international armed conflicts’. The ICRC has proposed a third category of persons who are ‘members of organised groups belonging to a non-state party to the conflict’ and who ‘cease to be civilians for as long as they remain members by virtue of their continuous combat function’.

The IHL notion of ‘direct participation in hostilities’ is, however, problematic from an operational standpoint, as conflicts are increasingly fought in urban areas and other civilian population centres. In Iraq and Afghanistan, for example, North Atlantic Treaty Organization (NATO) forces have fought counter-insurgency campaigns against well-armed and organised ANSAs whose members do not wear uniforms and are often indistinguishable from the civilian population. There are many comparable situations in other places where the two paradigms may overlap. The ICRC (2013 p.1) has noted: ‘For example, in a non-international armed conflict, when a State is using force against fighters, it may be considered as simultaneously conducting hostilities and maintaining law and order (since fighters are also frequently criminals under domestic law)’. There may also be situations in which civilians are present alongside fighters, for example during a riot, or where civilian unrest escalates into an armed conflict in which the rules regarding the use of lethal force will be different depending on which legal framework is considered applicable.

Partly in response to the cases discussed above at the ECtHR, NATO forces have adopted Civilian Harm Mitigation (CHM) policies to better track and investigate civilian casualties and compensate victims (CIVIC 2014; 2019b). As Keenan and Holt (2018) note, the provisions of CHM go significantly beyond the ‘minimum requirements’ of IHL and read far more consistently with those of IHRL. The adoption of CHM was also a self-conscious part of NATO’s attempts to develop new counter-insurgency strategies as part of their ‘surges’ in both countries and it is increasingly used in military operations elsewhere (Kilcullen 2009).

In October 2012, a group of states adopted a set of principles and guidelines under the ‘Copenhagen Process on the handling of detainees in international military operations’. The principles were drawn up for ‘military operations, such as those conducted by coalition forces in Iraq and Afghanistan, as well as UN peacekeeping operations where the use of force is authorized’ (Hartman 2012). They are intended to address the ‘legal uncertainties’ surrounding detentions in such situations, particularly ‘when the situation in which the military operation takes place changes from an international to a non-international armed conflict or to a situation of no conflict’ (Winkler 2016). The principles, which are not legally binding, were ‘welcomed’ by seventeen of the participating states, although two expressed concerns about whether they fully reflected the provisions contained in IHRL. However, Amnesty International warned that they ‘pander to existing poor practices’ and were ‘ripe for exploitation by those seeking to evade their obligations under IHL and international human rights law’ (AI 2012, p. 3).

Similar concerns surrounded attempts by the Israeli Defence Force (IDF) to create a new legal category of ‘armed conflict short of war’ (Jones 2021). The IDF reportedly developed a six-point test governing the use of lethal force in the Occupied Palestinian Territories, which wove together IHL provisions – such as the principles of distinction and proportionality – with the specification under IHRL that if arrest rather than killing of a suspect is possible, then this must be attempted, so long as the suspect is under ‘Israeli security control’. In 2012, a former UK Foreign Office official published another set of principles governing self-defence against non-state actors, based on informal discussions with senior officials in likeminded states. These set out when states may use force extra-territorially against ANSAs (Bethlehem 2012). But they have also been criticized for implicitly ‘stretching’ the circumstances in which states can invoke their right to self-defence to carry out targeted killings (Rona and Wala 2013).


The provisions of IHL are only relevant to situations of armed conflict – whether of an international or non-international character. In all other situations the rules of IHRL will exclusively govern the conduct of the security forces in law enforcement operations. This article has argued, however, that there are many situations in which the line between a NIAC and violent internal disturbances, which do not reach this threshold cannot be so easily drawn. Chinkin and Kaldor argue that the binary divisions within the existing international legal framework are outdated and do not reflect the realities of violent situations in the twenty-first century. The case-studies drawn on in their study of the ‘new wars’ – such as Syria, Iraq, Afghanistan, Mali, South Sudan and the Democratic Republic of Congo – are widely recognised to be situations of armed conflict, but even here they maintain that ‘the cardinal principles of IHL do not fit contemporary forms of war. The threshold requirement and the principles of distinction, necessity and proportionality are very difficult to apply’ and so its provisions will always need to be complemented by IHRL.

We reject the ‘armed conflict’ rhetoric deployed by President Bolsonaro and his supporters, including those in the police and military. But even if it is accepted that the violence in the favelas of Rio de Janeiro has reached the threshold level comparable to recognized NIACs elsewhere in the world, this still does not give the Brazilian security forces carte blanche for shoot-to-kill operations, since the state retains effective control over these territories and so all the protections of IHRL remain fully applicable. There is, moreover, little clarity under IHL about how to define the status of some of the various state and non-state actors involved.

The Brazilian military police, while heavily militarized and with a reputation for using excessive force, are still recognizably a police force, responsible for law enforcement and acting, at least in theory, within an IHRL paradigm. The Brazilian army is sometimes deployed in support of the civil authorities but operating within rules of engagement based on the same legal framework. The militias are clandestine groups, with links to both corrupt police officers and politicians in Rio de Janeiro. Clearly, though, they are also criminals who extort from and terrorize the communities that they claim to be protecting and are increasingly indistinguishable from the drug factions that they purport to oppose.

To elevate any of Rio’s criminal groups or ANSAs to the status of a party to a NIAC under IHL is both legally and politically problematic. Since the basic cornerstone of IHL is the principle of distinction, it would not be a crime under international law for one group of combatants to engage in hostilities with another in an NIAC. President Bolsonaro and his supporters appear to believe that defining violence in Rio de Janeiro as an ‘armed conflict’ gives greater licence to the police in the use of lethal force whereas, in fact, it turns them into legitimate targets under IHL.

Competing Interests

The author has no competing interests to declare.


  1. ALERJ. 2008. Relatório Final da Comissão Parlamentar de Inquérito Destinada a Investigar a Ação de Milícias no Estado do Rio de Janeiro. Rio de Janeiro: Assembleia Legislativa do Estado do Rio de Janeiro. 

  2. Alves, F and Otavio, C. 2022. ‘Ex-chefe de Policia Civil preso usava mortes no Jacarezinho como bandeiro de campanha’. O Globo, 9 September 2022., accessed 3 October 2022. 

  3. Alves, MHM and Evanson, P. 2011. Living in the Crossfire, Favela residents, drug dealers and police violence in Rio de Janeiro. Philadelphia: Temple University Press. 

  4. Amnesty International (AI). 2012. Outcome of Copenhagen Process on detainees in international military operations undermines respect for human rights. Available at: [Last accessed 19 September 2022]. 

  5. Araújo, V. 2020. Novo secretário de Polícia Civil quer tanques em favelas e promete avançar no caso Marielle. O Globo, 27 September 2020. Available at:,a%20Secretaria%20de%20Pol%C3%ADcia%20Civil.&text=Mas%20agora%20Turnowski%20%E2%80%94%20que%20foi,planos%20para%20enfrentar%20a%20viol%C3%AAncia [Last accessed 18 September 2022]. 

  6. Azzi, VF. forthcoming. ‘The war on criminality in Rio de Janeiro: Pacifying unruly territories?’ In: Solar, C and Pérez Ricart, C (eds.), Crime, Violence and Justice in Latin America. Abingdon: Routledge. 

  7. BBC. 2016. Four dead after police helicopter ‘shot down by gang’. BBC News Brazil, 20 November 2016. Available at: [Last accessed 18 September 2022]. 

  8. Bellal, A. 2019. The War Report: Armed conflicts in 2018. The Geneva Academy of International Humanitarian Law and Human Rights. 

  9. Bethlehem, D. 2012. Self-defense against an imminent or actual armed attack by nonstate actors. American Journal of International Law, 106(4): 770–777. DOI: 

  10. Betim, F. 2021. Em afronta ao STF, polícia do Rio impõe sigilo a operação do Jacarezinho e outras ações na pandemia por cinco anos. El País, 26 May 2021. Available at:,letal%20da%20hist%C3%B3ria%20do%20Estado.&text=A%20corpora%C3%A7%C3%A3o%20estendeu%20o%20sigilo,5%20de%20junho%20de%202020 [Last accessed 18 September 2022]. 

  11. Brazilian Government Ministry of Defence. 2012. Livro Branco de Defesa Nacional. At: [accessed on 21 September 2022]. 

  12. Brazilian Government Ministry of Defence [online]. 2013. ‘Garantia da Lei e da Ordem’. 20 December. [Last accessed on 21 September 2022]. 

  13. Brazilian Government Ministry of Defence [online]. 2014. ‘Garantia da Lei e da Ordem’. 3 February. [Last accessed on 21 September 2022]. 

  14. Bruderlein, C. 2000. The Role of Non-State Actors in Building Human Security: The case of armed groups in intra-state wars. Geneva: Centre for Humanitarian Dialogue. 

  15. Cano, I. 2008. Seis por meia dúzia? Um estudo exploratório do fenômeno das chamadas ‘milícias’ no Rio de Janeiro. In: Global, J (ed.), Segurança, milícias e tráfico no Rio de Janeiro. Rio de Janeiro: Justiça Global and Fundação Heinrich Böll. pp. 48–83. 

  16. Carta Capital. 2014. [Video] Átila Roque fala do “extermínio da população jovem negra”. 30 December 2014. Available at: [Last accessed 18 September 2022]. 

  17. Center for Civilians in Conflict (CIVIC). 2019a. Reverse the Trend: 2021–25 Strategic plan. [Last accessed 22 September 2022]. 

  18. Chernishova, O. 2013. ‘Recent developments in the case-law of the European Court of Human Rights related to the issue of the use of force’. Presentation in ICRC Expert Meeting. Appendix 6, accessed 3 October 2022. 

  19. Chinkin, C and Kaldor, M. 2017. International Law and New Wars. Cambridge University Press. DOI: 

  20. CIVIC. 2014. Civilian Harm Tracking: Analysis of ISAF efforts in Afghanistan. New York: CIVIC. 

  21. CIVIC. 2019b. The Sum of All Parts: Reducing civilian harm in multinational coalition operations. New York: CIVIC. 

  22. Comissão Parlamentar de Inquérito do Assassinato de Jovens (CPIADJ). 2015. Relatório Final. Comissão Parlamentar de Inquérito do Assassinato de Jovens., accessed 3 October 2022. 

  23. Dalby, C and Francisco, A. 2019. Militias become luxury real estate barons in Rio de Janeiro. InSight Crime, 23 April 2019. Available at: [Last accessed 18 September 2022]. 

  24. Dowdney, L. 2003. Children of the Drug Trade: A case study of children in organized armed violence in Rio de Janeiro. Rio de Janeiro: Viva Rio/ISER (Instituto de Estudos da Religião). 

  25. ECtHR. 1978. Ireland v UK, Appl. No. 5310/71, Judgment 18 January 1978. DOI: 

  26. ECtHR. 1988. Brogan and others v. UK. Appl. No. 11209/84, Judgment 29 November 1988. 

  27. ECtHR. 1993. Brannigan and MacBride v. UK. Appl. No.14553-4/89, Judgment 24 May 1993. 

  28. ECtHR. 1995. McCann and others v. UK. Appl. No. 18984/91, 5 September 1995. 

  29. ECtHR. 1998a. Kaya v. Turkey. Appl. No. 22729/93, Judgment 19 February 1998, para 9. 

  30. ECtHR. 2005a. Isayeva, Yusupova and Bazayeva v. Russia. Appl. No. 57947-49/00, Judgment 24 February 2005, para 179. 

  31. ECtHR. 2011a. Al-Skeini and Others v. UK. Appl. No. 55721/07, Judgment (Grand Chamber) 7 July 2011. 

  32. ECtHR. 2011b. Al-Jedda v. UK. Appl. No. 27021/08, Judgment (Grand Chamber) 7 July 2011. 

  33. ECtHR. 2014a. Hassan v. United Kingdom. Appl. No. 29750/09, Judgment (Grand Chamber), 16 September 2014. 

  34. ECtHR. 2014b. Jaloud v. The Netherlands. Appl. No. 47708/08, Judgment (Grand Chamber), 20 November 2014. 

  35. Escola de Cultura de Pau (ECP). 2022. Alert 2022! Report on Conflicts, Human Rights and Peacebuilding. Barcelona: Icaria. 

  36. Estado de Minas, Chefe do’ Escritório do crime’, ‘capitão Adriano’ morre em tiroteio com a polícia, 9 February 2020. Available at:,1120569/chefe-do-escritorio-do-crime-capitao-adriano-morre-em-tiroteio-policia.shtml [Last accessed 18 September 2022]. 

  37. European Court of Human Rights (ECtHR). 2021. Hanan v. Germany. Appl No. 4871/16, (Grand Chamber), Judgment, 16 February 2021. 

  38. FBSP. 2019. Anuário Brasileiro de Segurança Pública. Available at: [Last accessed 19 September 2022]. 

  39. FBSP [Fórum Brasileiro de Segurança Pública]. 2021. Anuário Brasileiro de Segurança Pública. Available at: [Last accessed 19 September 2022]. 

  40. FIRJAN. 2017. Avanço da criminalidade no Estado do Rio de Janeiro: Retrato e propostas para a segurança pública. Rio de Janeiro: FIRJAN. 

  41. Fogo Cruzado. 2020. Database Available at: [Last accessed 19 September 2022]. 

  42. Foley, C. 2014. Pelo telefone: Rumors, truths and myths in the ‘pacification’ of the favelas of Rio de Janeiro. HASOW (Humanitarian Action in Situations Other than War) Discussion Paper 8. Rio de Janeiro. 

  43. Foley, C. 2017. UN Peacekeeping and the Protection of Civilians: Saving succeeding generations. Cambridge University Press. DOI: 

  44. Foley, C. (ed.). 2019. In Spite of You: Bolsonaro and the new Brazilian resistance. New York: OR Books. DOI: 

  45. Foley, C. 2022. Human Rights and Protection by UN Peace Operations. New York, International Peace Institute. 

  46. Gearty, C. 1995. After Gibraltar. London Review of Books, 17(22), 16 November 1995. 

  47. Gonçalves, LL. 2017. Letalidade violenta e controle ilegal do território no Rio de Janeiro. Cadernos de Segurança Pública, 9(8): 1–19. 

  48. Grellet, F. 2019. Witzel sobe em helicóptero com sniper e causa polêmica. Terra, 6 May 2019. Available at:,0b107d79aa61ac85cb70ef9a691d69f7gnmzuvqr.html [Last accessed 18 September 2022]. 

  49. Gross, A. 2017. The Writing on the Wall: Rethinking the international law of occupation. Cambridge University Press. DOI: 

  50. Hampson, F. 2008. The relationship between international humanitarian law and human rights law from the perspective of a human rights treaty body. International Review of the Red Cross, 90(871): 549–72. DOI: 

  51. Hampson, F and Lubell, N. 2013. Amicus Curiae brief submitted by Professors Francoise Hampson and Noam Lubell in the case of Hassan v. United Kingdom, Appl. No. 29750/09. Human Rights Centre, University of Essex. 

  52. Hartman, J. 2012. The Copenhagen Process: Principles and Guidelines. Blog of the European Journal of International Law, 3 November 2012. Available at [Last accessed 18 September 2022]. 

  53. Hauck, P and Peterke, P. 2010. Organized crime and gang violence in national and international Law. International Review of the Red Cross, 92(878): 407–436. DOI: 

  54. Henckaerts, JM and Doswald-Beck, L. 2005. Customary International Humanitarian Law. Cambridge University Press. DOI: 

  55. Hirata, DV and Grillo, CC. 2017. Sintonia e amizade entre patrões e donos de morro: Perspectivas comparativas entre o comércio varejista de drogas em São Paulo e no Rio de Janeiro. Tempo Social, 29(2): 75–98. DOI: 

  56. Hirata, DV and Grillo, CC. 2019. Sumário executivo: Operações policiais no Rio de Janeiro. Rio de Janeiro: Fundação Heinrich Böll Brasil. 

  57. Human Rights Watch (HRW). 2021. Brazil: Investigate Rio Police Command for Deadly Raid. Available at: [Last accessed 19 September 2022]. 

  58. Human Rights Watch (HRW). 2022. Another police raid in Rio leaves 23 dead. Available at, Accessed 3 October 2022. 

  59. IACHR. 2008. Juan Carlos Abella v Argentina. Case 11.137, 13 April 2008 (‘La Tablada’). 

  60. IACtHR. 1995. Neira Alegria v. Peru, 19 January 1995, para. 74. DOI: 

  61. IACtHR. 2017. Favela Nova Brasília v Brazil. Judgment (Preliminary objections, merits, reparations and costs), 16 February 2017. 

  62. ICC. 2007. The Prosecutor v. Thomas Lubanga Dyilo (ICC-01/04-01/06). Decision on the Confirmation of Charges, Pre-Trial Chamber I, 29 January 2007. 

  63. ICRC. 2008. How is the Term “Armed Conflict” Defined in International Humanitarian Law? Opinion Paper, March 2008. Geneva. International Committee of the Red Cross. 

  64. ICTY. 2008a. Prosecutor v. Ramush Haradinaj, Idriz Balaj, Lahi Brahimaj (IT-04-84). Judgment, Trial Chamber I, 3 April 2008, para. 49. 

  65. Instituto de Segurança Pública. Available at:, [Last accessed 19 September 2022]. 

  66. Inter-American Commission on Human Rights (IACHR). 2010. Franklin Guillermo Aisalla Molina (Ecuador – Colombia). Report No. 112/10 (Admissibility), Inter-state Petition IP-02, October 21, 2010, paras 117–121. 

  67. International Committee of the Red Cross (ICRC). 2011. Brazil: mitigating the effects of armed violence, Interview, 16 September 2011. [Last accessed 26 September 2022]. 

  68. Inter-American Court of Human Rights (IACtHR). 2000. Bámaca-Velásquez v Guatemala. Series C No. 70 [Merits]. 

  69. International Court of Justice. 1996. Legality or Threat of Use of Nuclear Weapons. Advisory Opinion of 8 July 1996, ICJ Reports 1996. 

  70. International Criminal Court (ICC). 2008. The Prosecutor v. Germain Katanga (ICC-01/04-01/07). Decision on the Confirmation of Charges, Pre-Trial Chamber I, 30 September 2008, para. 239. 

  71. International Criminal Tribunal for Yugoslavia (ICTY). 1995. Prosecutor v. Dusko Tadic. Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, Case No. IT-94-1-AR72, 2 October 1995. 

  72. International Criminal Tribunal for Yugoslavia (ICTY). 2005. Prosecutor v Fatmir Limaj, Haradin Bala, Isak Musliu (IT-03-66-T). Judgment, TC II, 30 November 2005 (‘Limaj’). 

  73. Jones, C. 2021. The War Lawyers: The United States, Israel, and Judicial Warfare. Oxford University Press. DOI: 

  74. Keenan, M and Holt, V. 2018. Preparing to Protect: Advice on Implementing NATO’s Protection of Civilians Policy. NATO ACT OPEN Publications. 

  75. Kenkel, KM. 2006. ‘Language matters: security discourse and civil-military relations in post-ditadura Brazil’. Paper presented at the 47th Annual Convention of the International Studies Association, San Diego. 

  76. Kilcullen, D. 2009. The Accidental Guerrilla: Fighting small wars in the midst of a big one. London: Hurst & Co. 

  77. Kolb, R. 2003. Applicability of international humanitarian law to forces under the command of an international organization. In Faite, A and Grenier, JL (eds.), Report: Expert Meeting on Multinational Peace Operations: Applicability of International Humanitarian Law and International Human Rights Law to UN mandated forces. Geneva: International Committee of the Red Cross. 

  78. Larsen, KM. 2012. The Human Rights Treaty Obligations of Peacekeepers. Cambridge University Press. DOI: 

  79. Leeds, E. 1996. Cocaine and parallel polities in the Brazilian urban periphery: Constraints on local- level democratization. Latin American Research Review, 31(3): 47–84. DOI: 

  80. Leeds, E. 2007. Rio de Janeiro. In Koonings, K and Kruijt, D (eds.), Fractured Cities: Social Exclusion, Urban Violence and Contested Spaces in Latin America. London: Zed Book. pp. 23–35. DOI: 

  81. Lopes, M. 2022. ‘Police raid kills 18 in Rio, sparking outcry and accusations of abuse’. Washington Post, 22 July 2022,, accessed 3 October 2022. 

  82. Lubell, N. 2010. Extraterritorial Use of Force Against Non-state Actors. Oxford University Press. DOI: 

  83. Lyra, D. 2020. Operários da firma: Mundo do trabalho no mundo do crime. Revista Antropolítica, 50(90): 90–115. DOI: 

  84. Neves, Y. 2019. Death from above: The use of police helicopters in Rio de Janeiro. InSight Crime, 15 July 2019. Available at: [Last accessed 18 September 2022]. 

  85. Manso, BP. 2020. A República das Milícias: Dos esquadrões da morte à era do Bolsonaro. São Paulo: Todavia. 

  86. Nogueira, I and Barbon, J. 2021. Antes de apuração, diretor de Departamento de Homicídios diz que ‘não houve execução’ em operação. Folha de São Paulo, 7 May 2021. Available at: [Last accessed 18 September 2022]. 

  87. Olliveira, C, Couto, MI, Hirota, R and Spagnuolo, S. 2018. Tá tudo dominado: Exclusivo: as milícias assumiram o controle do Rio de Janeiro. The Intercept, 5 April 2018. Available at: [Last accessed 18 September 2022]. 

  88. Peterke, S. 2012. Regulating ‘drug wars’ and other gray zone conflicts: Formal and functional approaches. HASOW (Humanitarian Action in Situations Other than War) Discussion Paper 2. Rio de Janeiro. 

  89. Reidy, A. 1998. The approach of the European Commission and Court of Human Rights to International Humanitarian Law. International Review of the Red Cross, 324(30): 513–530. DOI: 

  90. Reidy, A, Hampson, F and Boyle, K. 1997. Gross violations of human Rights: Invoking the European Convention on Human Rights in the Case of Turkey. Netherlands Human Rights Quarterly, 15(2): 161–173. DOI: 

  91. Rona, G and Wala, R. 2013. No thank you to a radical rewrite of the jus ad bellum. American Journal of International Law, 107(2): 386–390. DOI: 

  92. Rothenburg, D. 2018. “Se eu errar, o PT volta”, diz Bolsonaro em entrevista exclusiva. Correio Braziliense, 2 November 2018. Available at:,717119/em-entrevista-bolsonaro-reafirma-papel-dos-militares-no-ministerio.shtml [Last accessed 18 September 2022]. 

  93. Sassòli, M. 2019. International Humanitarian Law: Rules, controversies, and solutions to problems arising in warfare. Geneva: Edward Elgar Publishing. 

  94. Sivakumaran, S. 2012. The Law of Non-International Armed Conflict. Oxford University Press. DOI: 

  95. UNAMA/OHCHR. 2021. Afghanistan: Protection of civilians in armed conflict: Annual Report 2020. Kabul: United Nations Assistance Mission in Afghanistan/Office of the High Commissioner for Human Rights. 

  96. United Nations Human Rights Committee (UNHRC). 1982. Camargo and Suarez de Guerrero v. Colombia, 31 March 1982, CCPR/C/15/D/45/1979. 

  97. Veja. 2017. Homens ostentam armas em vídeos gravados em favelas no Rio, 1 October 2017. Available at: [Last accessed 18 September 2022]. 

  98. Veja. 2018. Wilson Witzel: ‘A polícia vai mirar na cabecinha e… fogo’. 1 November 2018. Available at: [Last accessed 18 September 2022]. 

  99. Viga Gaer, R. 2021. ‘At least 8 dead in mangrove after gun battle with Rio police’. 22 November 2021, Reuters. 

  100. Winkler, T. 2016. ‘The Copenhagen Process Principles and Guidelines for Detention: Legal and political challenges”. In Rose, G and Oswald, B (eds.), Detention of Non-State Actors Engaged in Hostilities. Leiden, The Netherlands: Brill/Nijhoff. DOI: