Neutralizing Threat : Reevaluating the Scope of Acceptable Targets in Unconventional Conflicts

Andrew T. Strong

I.      Two types of international law........................................................................................ 3

A.    Jus ad bellum............................................................................................................. 3

B.     Jus in bello................................................................................................................ 5

II.     ICTY............................................................................................................................ 6

III.       Existence of an Armed Conflict.................................................................................. 7

IV.       Jus in Bello’s two duties........................................................................................... 11

A.    Who is a combatant?............................................................................................... 16

1.     International conflicts........................................................................................... 16

2.     Internal Conflicts................................................................................................. 20

B.     Duty to Discriminate................................................................................................ 24

C.    Duty of Proportionality............................................................................................ 25

V.    Threat provides the context for defining the duties......................................................... 30

A.    Threat to insurgents varies with regime brutality and openness................................... 31

A. Application to specific conflicts................................................................................... 33

1. Private Constriction of Civil Society......................................................................... 39

2. Private acts of violence............................................................................................ 39



The second half of the 20th Century has seen significant leaps in international criminal law.  This field of law has changed dramatically since the Nuremburg Trials sentenced Axis war criminals for their crimes in World War II.  In 1993, the International Criminal Tribunal for Former Yugoslavia (ICTY) became the first international criminal court empowered by the U.N. Security Council to hold individuals responsible for their actions during the war.  In adjudicating the means through which war can legally be waged, the ICTY relies upon instruments such as the 1949 Geneva Conventions, the 1977 Additional Protocols, the 1907 Hague Regulations, and the Universal Declaration of Human Rights.  Together these instruments provide a strong foundation for the ICTY.

The shape of conflict has also dramatically changed in the second half of the 20th Century.  Traditional wars between nation-states have given way to sporadic explosions of violence and sustained insurgency which often occurs not between states but within one state.  The reality of current conflict tests the applicability of international criminal law which, for the most part, was developed with a traditional model of conflict in mind. 

Significantly, the relative power of opposing parties is often dramatically skewed during a modern conflict.  Insurgencies, for instance, rely on secrecy and timing to compensate for being drastically under-manned, outgunned, and under-resourced.  As a result, both governments and insurgents are faced with a different set of threats presented by different actors to the conflict.         

Defining the actors of a conflict, and clearly articulating who is an acceptable target has been one of the most challenging issues the court has had to address.  Clear lines between combatants and civilians are blurred in an insurgency and a set of ambiguous actors in the conflict make it difficult to determine who can be legally targeted. 

The farmer-by-day fighter-by-night insurgent is the most obvious deviation from the traditional notion of a combatant.  A more difficult example to categorize during an insurgency, however, is a regime informant.  Such an individual is undoubtedly a civilian by traditional standards, and yet within the context of an internal conflict an informant often presents a greater threat to an insurgency than an enemy soldier. 

Clearly categorizing such an individual is difficult.  The distinction between combatant and civilian has been a chronic point of contention which international criminal law has yet to adequately address.  This is largely a result of the law not responding to the reality of unconventional conflicts and varying threat similar individuals create in different conflict. 

Part I of this analysis examines the law as it currently exists. It describes where international criminal law is applicable, to whom it applies, and examines the current distinctions between civilians and combatants during both international and internal conflicts.

Part II analyzes the current approach and describes problems with its application to an insurgency and in particular, examines the threat and legal difficulties regime informants present to a guerrilla army.  It then suggests a new approach.                    

I.       Two types of international law

International laws which govern the activity of participants in an armed conflict can be broadly broken down into two categories: jus ad bellum and jus in bello.   

A.   Jus ad bellum

Jus ad bellum defines conditions that render a conflict “just.” This doctrine traditionally requires a nation engaging in an armed conflict to have a legitimate reason to go to war.  It is satisfied when the initial decision to raise arms is justifiable.[1]  The German invasion of Poland during World War II presents a clear example of a nation in violation of jus ad bellum.[2] Not all conflicts are as straightforward, however, and determining what qualifies as a “just” war presents obvious semantic problems.  Ambiguity over which conflicts count as just is addressed by the United Nations in its charter, which specifically defines two types of “just” conflicts.  The first category of conflicts which satisfy jus ad bellum are conflicts fought in self defense;[3] the second category includes conflicts where the UN Security Council authorizes the use of armed force.[4]

The doctrine of Jus ad bellum derives its authority from treaties made between nations.[5]  This includes the UN Charter, which applies to all contracting parties to the UN, and thus the Charter’s limited categories of jus ad bellum extends to all such parties.[6]

It is important to note that Jus ad bellum regulates only the legitimacy of the conflict and does not subsequently restrict the behavior of combatants.  The doctrine cannot be used to regulate the means through which the war is fought.[7]  If jus ad bellum were the lone source of international laws governing armed conflict, a party fighting a ‘just’ war could employ any means it sees fit to dispose of its enemy. Targeted civilian attacks, chemical weapons, and summary executions would be viable options.  There would be no need to quibble over the legality of an event such as Hiroshima, and nothing to stop a nation fighting in accordance with jus ad bellum from conducting similar operations.  Clearly, this does not lead to a desirable result and thankfully, it is not the case.

B.   Jus in bello

            The doctrine of jus in bello exists to regulate the means through which a conflict is carried out and derives its authority from customary international law, reinforced by both the 1949 Geneva Conventions and the 1907 Hague Protocols.[8]  Jus in bello applies regardless of whether a conflict is ‘just.’  Indeed, even in a situation where a nation illegally invades its neighbor in pursuit of an ethnic cleansing campaign, both parties to the conflict are still bound by jus in bello.  Both parties would thus be required to use appropriate means while attempting to accomplish their objectives.  What constitutes ‘appropriate means’ is discussed in more detail below. 

            The most significant difference between the doctrines of jus ad bellum and jus in bello is that jus ad bellum applies only to the collective.[9]  A state may be in breach of jus ad bellum, but an individual never is.  As a result, no specific body exists to punish jus ad bellum offenders.  There is no court, for instance, in which to try the state of Iraq for its 1990 invasion of Kuwait.  Instead, such violations are typically dealt with through sanctions, boycotts, and in some cases, U.N. sponsored force.  In contrast, jus in bello applies to individuals and with the formation of the International Criminal Tribunal for the former Yugoslavia (and subsequently other such courts) individuals who violate jus in bello can be put in jail.

            The significance of this should not be overlooked.  In creating the ICTY, the U.N. Security Council empowered an independent institution with the unilateral ability to reach inside a sovereign nation’s boundaries and prosecute an individual therein. 


Relying primarily on the four 1949 Geneva Conventions, two Additional Protocol of 1977, the Universal Declaration of Human Rights, and the 1907 Hague Protocol, the UN Security Council, in establishing the International Criminal Tribunal for the former Yugoslavia (“ICTY”) codified a framework with which to govern the acceptable means of conducting war, and created a jurisdiction capable of prosecuting individuals who violated that framework.   

The ICTY’s power was challenged in its first case, by Dusko Tadic when he became the first individual indicted for war crimes in the former Yugolavia.  Among other grievances, Dusko Tadic raised three objections which struck at the authority behind his indictment. First, he argued that the ICTY had not been lawfully created by the Security Council and was therefore without authority to issue any indictments.[10]  Second, he challenged the primacy of the ICTY’s jurisdiction over a national court.[11]  And third, Tadic argued that the ICTY only had subject matter jurisdiction over crimes against humanity if committed during an international armed conflict.[12]   Tadic reasoned that his alleged crimes did not occur during an armed conflict, and thus none of his actions fell within any of the Articles of the ICTY Statute.[13]

            As a result of the defense’s approach in Tadic, the case quickly became a seminal moment in the infant court’s existence, and an important test as to whether the international community could enforce jus in bello.  The Appeals Chamber rejected each of Tadic’s arguments in turn, thus legitimatizing its own existence and function.  In addressing Tadic’s third argument, however, the Appeals Chamber was forced to define what types of conflict qualified as ‘armed conflicts’ and distinguish from that definition banditry, rioting, and other lawless criminal behavior over which municipal courts have the exclusive right to adjudicate.  The Court’s decision on this matter is significant because it roughly outlines the types of situations to which jus in bello applies and thus created a minimum threshold which must exist, independent of the Accused’s actions, in order for the ICTY to have jurisdiction. 

III.Existence of an Armed Conflict

            Tadic pointed out that for the ICTY to have jurisdiction for the alleged crimes, it was necessary for the crimes to have been committed during an armed conflict.  The basis for this argument can be found in the language of Common Articles 2 and 3 of the Geneva Conventions.  The relevant language of Article 2 provides that. “the present Convention shall apply to all cases of declared war or of any other armed conflict which may arise between two or more of the High Contracting Parties.”[14] (emphasis added)  The language here restricts application of the Conventions to situations where an armed conflict is present.  Likewise Article 3 states, “In the case of armed conflict not of an international character… each Party to the conflict shall be bound to apply, as a minimum, the following provisions.”[15] (emphasis added) Again the threshold for applying the provisions of the article is the existence of an armed conflict, without which the Geneva Conventions do not apply.  The effect of this language is that the ICTY, which bases much of its jurisdiction on the Geneva Conventions, cannot prosecute individuals absent an armed conflict.[16]     

            Tadic conceded that there had been an armed conflict in Yugoslavia, but he argued that at the time of his actions in Prijedor – where he allegedly committed his crimes – an armed conflict no longer existed.[17]  As a result, the Appeals Chamber was forced to define not only what qualified as an armed conflict, but also determine its scope.

Addressing this problem, the Tadic Court found two factors relevant to determining when an armed conflict existed: (1) the intensity of the conflict, and (2) the organization of the parties present.[18]  Applying these factors provides a basis for determining when an armed conflict begins; it does not, however, help in determining the crux of Tadic’s argument, the boundaries of the conflict.  Armed conflicts come in various forms, and are often not a linear progression over an easily defined area. Instead, they can exist as sporadic outbursts of violence over discontinuous areas.  In order to define adequately how far an armed conflict extends, both the temporal and geographic scope of a conflict must be considered.[19]      

The Tadic Court took a broad view of both.  The Appeals Court held that an ‘armed conflict’ extends beyond the precise time and area where fighting occurs and that the conflict extends to the “whole territory under the control of a party, whether or not actual combat takes place until a peaceful settlement of the conflict is achieved.” [20]  This decision effectively ended Tadic’s jurisdictional challenge to the Court; it also had significant implications on the application of jus in bello.

Framing when and where an armed conflict exists affects the interplay between international criminal law and municipal law.  This is significant because the two systems of law address similar actions differently.  Behavior that that occurs under international criminal law can qualify for immunities from municipal criminal law.[21]  For instance, a participant in an armed conflict who kills an enemy combatant cannot be prosecuted for murder.  Without the context of an ‘armed conflict,’ the act of killing another individual – even if that individual is a soldier – is considered murder, and the offending individual can be brought before a local court and punished.[22]  This concept is called ‘combatants’ privilege’ and is explored in more detail below; for now it is enough to point out that  extending the scope of ‘armed conflict’ not only extends the ICTY’s subject-matter jurisdiction, it also expands the applicability of jus in bello.  In turn, this both provides criminal liability under international criminal law, and potentially extends immunity from municipal criminal law to individuals who participate in an armed conflict.[23]

Establishing the ICTY as a legitimate court in which to try individuals who violate jus in bello represented a significant step for international criminal law.  With the formation of the ICTY and specifically, the Tadic decision, jus in bello found the teeth it needed to regulate the means employed in an armed conflict.  Further, the Tadic decision provided a working definition for the scope of an ‘armed conflict’ which framed the application of jus in bello. In recasting the consequences for failing to use appropriate means during a conflict, the ICTY’s enforcement of jus in bello creates specific incentives not to violate the doctrine.  It also creates duties and privileges for both participants and civilians in a conflict.

Complicating a uniform application of these duties and privileges has been the changing shape of conflict in the second half of the 20th Century.  The 1907 Hague Protocol, the Geneva Conventions, and much of the ICTY’s precedent to date were created with conventional forms of conflict in mind.  Increasingly conflicts have tended to be internal and non-conventional in character – i.e. guerrilla wars and sustained insurgencies.  It is not surprising then that the application of jus in bello to internal conflicts is often imperfect.  The following analysis seeks to define the implications of jus in bello as it applies to both the civilians and combatants in a conflict, and highlights some of the problems of applying the current approach to unconventional conflicts.

IV.Jus in Bello’s two duties

            During an armed conflict, jus in bello imposes two duties on combatants: a duty of proportionality and a duty to discriminate.   The interpretation of these two duties forms the basis of international criminal law and the ICTY has reached the majority of its judgments through interpretations of these duties.  This approach has served the Chambers well in establishing the criminal liability of actors in the Bosnian and Croatian conflicts of the early 1990’s.  These conflicts were relatively conventional in their character as the Bosnian, Croatian, and Serbian factions formed recognizable armies with a distinct hierarchy, and separate national affiliations.  Increasingly the ICTY has been called upon to judge the conduct of KLA guerrilla fighters in Kosovo.  The Kosovo conflict was decidedly unconventional in its character and the ICTY will likely have great difficulty applying the current duties and obligations to guerrilla actors there.  Specifically, the duty to discriminate becomes an unrealistic obligation when guerrilla combatants are held to the same standard as an organized army.  As a result, it makes it very difficult for guerrilla fighters to adhere to the laws of war. 

The duty to discriminate and the duty of proportionality both regulate the behavior of two groups of individuals in an armed conflict; (1.) those participating in a conflict – the combatants, and (2.) individuals who take no part in a conflict – the civilians.  In an armed conflict, international law recognizes only these two distinct types of individuals.[24]  Before exploring the nuances of these two categories, it is helpful to explain why they are important.

 In an armed conflict, civilians and combatants are entitled to a distinctly different range of privileges and burdened with a different range of duties. In order to determine what privileges to afford an individual or what duties an individual is bound by, it is necessary first to determine whether that individual is a combatant or a civilian.  At a glance this seems an obvious distinction; the line, however, can quickly become blurred.  This section defines the privileges and duties attached to both combatants and civilians, examine instances where the line between them blurs, and explores the language of the Geneva Conventions in order to determine who qualifies as a civilian and who qualifies as a combatant.

            War is not, therefore, a relation of man to man but a relation of state to state.[25]

            The above quote provides a basis from which to understand why the concept of combatant’s privilege has found its way into international law.  War has been traditionally viewed as a conflict between States,[26] not as a conflict between individuals.[27]  Thus, violence against appropriate targets – i.e. enemy combatants – which is not only sanctioned, but compelled by the State, does not result in individual criminal liability. Contrast this idea with a bank robber’s shootout with police following a heist.  Here the conflict exists between individuals, or at least between an individual and society.  In this scenario, it is the bank robber who is directly and criminally responsible should police officers be shot during the gun battle.  Although both the death of a police officer shot by a robber and the death of a combatant shot by an enemy soldier involve an intentional killing, the laws of war permit the soldier’s action, while the robber is subject to criminal liability for murder based on whatever municipal jurisdiction applies. 

            As combatants, individuals cannot be prosecuted for violence provided their actions conform to laws of combat.[28]  Further, captured combatants are entitled to POW status.  This is consistent with the law’s distinction between civilians and combatants; combatants may not be tried in the regular courts; they are immune from municipal criminal law and instead they are held until the cessation of hostilities.[29]  This privilege applies to combatants irrespective of whether the combatant is fighting for ‘just’ cause. [30] Significantly, the privilege to kill in certain situations with immunity implicitly carries the sobering caveat that combatants are in turn legal targets to enemy combatants. 

            In sum, combatants are privileged to engage in a sphere of violence during an armed conflict and are immune from prosecution stemming from their conduct.  If captured, they are entitled to POW status upon their detention.  Although civilians lack immunity from prosecution during an armed conflict, and thus have a duty to refrain from taking part in the hostilities, they gain the privilege not to be legal targets during a war.[31]

            These privileges and duties create an intriguing dynamic over the desirability of combatant versus civilian status.  Consider a situation where a group of individuals are alleged to have killed several soldiers.  Should those individuals be classified as a group of combatants they would become legal targets to enemy soldiers and could be shot unless they surrendered.[32]  Conversely, should those individuals be classified as civilians, they cannot simply be shot by opposing forces; once captured or having surrendered, they would be entitled to all the procedures, safeguards, and security of a State’s (or an organized insurgency’s) justice system; they would likely be put on trial for murder. 

In the latter case, if the group had access to a courtroom, they would choose a defense based entirely on the situation.  If the prosecution’s evidence for a murder trial is particularly weak,[33] the State has a liberal sentencing policy, or the conflict looks to be of indefinite duration –America’s ‘war on terror’ for instance – then the best defense may be to argue that the accused are civilians.  Alternatively, if there was a strong case for murder, a potential death penalty, or the conflict seems as though it was nearly over, the best defense might be to argue that the accused are combatants and not criminally liable. Because the advantages of both civilian and combatant status are contextual, there is rarely consensus over the scope of who qualifies as a combatant.[34]  Government agencies charged with prosecuting crimes seek to cast an accused status in a way which maximizes the punishment; the perpetrators seek the inverse.  While this becomes an intriguing legal question (and provides much work for lawyers), it does little to clarify who qualifies as what.

 In order to concretely pin down the basis for defining each category, a brief analysis of the Geneva Conventions and the two Additional Protocols is necessary.

            The first challenge in defining exactly who qualifies as a combatant, and who qualifies as a civilian is to determine whether the conflict to which they are an actor is an internal or international affair.[35]  The language used to define participants in each type of conflict is dissimilar.  Before examining this difference, it is important to establish where the sources of their respective definitions come from.          

Internal conflicts are governed by Article 3 Common to the Geneva Conventions which applies “in the case of armed conflict not of an international character…”[36]  Internal conflicts are by definition not of international character and thus fit the criteria of Common Article 3.  Conversely, international conflicts are governed by Article 2 Common to the Geneva Conventions which applies the privileges and duties set forth in the Conventions to “all cases of declared war or any other armed conflict which may arise between two or more of the High Contracting Parties, even if a state of war is not recognized by one of them.”[37]  While the 1949 Geneva Conventions does not define the term “armed conflict” as either an international or internal conflict, Common Article 2’s use of the language “may arise between two of more of the High Contracting Parties” seems to foreclose this article applying to an internal conflict which occurs between one State and an insurgent group.[38]   Through the language of Common Article 2, international conflicts are governed by the entirety of the Geneva Conventions minus Common Article 3.  Adding a final layer of complexity to this issue are the two 1977 Additional Protocols.  Additional Protocol I develops and supplements Common Article 2 and therefore applies to international conflicts.[39]  Protocol I, however contains express language which expands the scope of Common Article 2 adding conflicts where “peoples are fighting against colonial domination and alien occupation and against racist regimes in the exercise of their right of self-determination.”[40]  (These are sometimes referred to as CARD conflicts.)   Protocol II develops and supplements Common Article 3 and restricts its scope from applying to from application to conflicts described by Protocol I (i.e., CARD conflicts).[41]  

Common Article 2 therefore governs international conflict, and through Additional Protocol I also extends to CARD conflicts.  Common Article 3 applies to internal conflicts except those which qualify as CARD conflicts.  Thus, specific types of conflict are governed by different Articles Common to the Geneva Conventions which use different language to define who is a combatant and who is a civilian.[42]  It is important to keep this in mind when examining this language.

A.   Who is a combatant?

1.     International conflicts

Article 4(A) of the Third Geneva Convention provides the basis for determining who qualifies as a combatant in an international conflict.  It provides POW status to :

(1) Members of the armed forces of a Party to the conflict as well as members of militias or volunteer corps forming part of such armed forces.


(2) Members of other militias and members of other volunteer corps, including those of organized resistance movements, belonging to a Party to the conflict and operating in or outside their own territory provided that such militias or volunteer corps, including such organized resistance movements, fulfill the following conditions :

            (a) that of being commanded by a person responsible for his subordinates;

            (b) that of having a fixed distinctive sign recognizable at a distance;

            (c) that of carrying arms openly

            (d) that of conducting their operations in accordance with the laws and customs of               war.


(3) Members of regular armed forces who profess allegiance to a government or an authority not recognized by the Detaining Power. [43]


This definition is further supplemented by Article 43 and Article 50.1 of Protocol I. Article 43 defines the term ‘armed forces’ as “all organized forces, groups, and units which are under a command responsible to the Party.”[44]  It also provides that ‘paramilitary or armed law enforcement units may incorporated into an armed force.[45] Article 50.1 defines civilians as any person who does not belong to one of the categories referred to in Article 4(A) (1), (2), (3), and (6) or in Article 43.[46] 

Together these Articles form the framework for delineating roles in international and CARD conflicts.[47]  International conflicts therefore benefit from affirmative definitions that describe who is a combatant and who is a civilian.  One could walk through a checklist of criteria provided by Article 4(A) and Article 43 to determine an individual’s status. Within this framework, for instance, a uniformed soldier easily qualifies as a combatant via Article 4(A)(1).  Similarly, a shopkeeper unaffiliated with any armed force would not fit the definition of Article 4(A) or Article 43 and thus qualifies as a civilian under Article 50.1.  These criteria are particularly useful, however, for determining less clear examples. Consider a regime collaborator; someone who voluntarily passes information to one side of a conflict. This individual is a much more ambiguous actor in a conflict, but his role can be determined by walking through the definition. 

            A collaborator could not qualify as combatant--an individual described by Article 4(A)(1).[48]  Although the collaborator may greatly facilitate the mission of the armed forces of one party to the conflict, a collaborator could not be called a ‘member of an armed force.’  Membership in an armed force is defined by Article 43 of Additional Protocol II, which requires an individual to be within some system of hierarchy, discipline, and command must be present to be considered part of an ‘armed force.’ Collaborators are not bound by army procedure, rules, or regulations; they do not report to superior officers, they would not enlist or serve a term as a regular soldier might.  An individual working within a formal intelligence gathering structure qualifying as a “system of hierarchy, discipline and command” would not qualify as a combatant under 4(A)(1) is more appropriately be labeled a spy.[49]  

            Likewise, a collaborator would not qualify as a combatant under Article 4(A)(2).[50]  Indeed, bearing arms openly and wearing a distinctive insignia are antithetical to a collaborator’s existence.  A collaborator’s survival would rely on blending in with the local population in order to collect authentic information.

            Similarly, Article 4(A)(3) fails to describe a collaborator.[51] This section applies to members of armed forces who profess allegiance to a party or government not recognized by the detaining power.[52]  Once again, the threshold requirement is membership in an armed force which a collaborator does not satisfy. 

            Article 4(A)(6) applies to “inhabitants of a non-occupied territory who… spontaneously take up arms to resist the invading forces…”[53]  Clearly this section could not be used to qualify a collaborator as a combatant.[54]  A collaborator, therefore, would not fall into any of the Article 4(A) categories, nor would Article 43’s definition of armed forces apply.  As such, in an international conflict a collaborator would fall within the definition of Article 50.1 and would be considered a civilian.[55]

Despite the existence of a mechanical framework for determining roles in an international conflict, there is still very often a lack of consensus about who qualifies as a combatant and who qualifies as a civilian. This ambiguity is the focal point of the U.S. government’s current policy of holding suspected Al Qaeida and Taliban fighters indefinitely without charge.  A full discussion of this exceeds the scope of this paper; at the heart of this debate, however, is the U. S. Government’s refusal to acknowledge individuals held at Guantanimo Bay and elsewhere as either combatants or civilians which in turn forecloses both POW status and the procedural mechanisms of the judicial system.[56]   Such individuals are labeled ‘unlawful combatants,’ and fall into this category if they participate in an armed conflict in such a way that their actions do not qualify them as a combatants.[57]        

2.     Internal Conflicts

            While there remains much debate over this topic when applied to international conflicts, the framework provided to regulate internal conflicts is much less specific.  As described above, internal conflicts are regulated by Article 3 Common to the Geneva Conventions and supplemented by Additional Protocol II.  Common Article 3 grants basic provisions to “persons taking no active part in hostilities.”[58]  Similar language is used by Additional Protocol II which states, “All persons who do not take a direct part or who have ceased to take part in hostilities are entitled to respect of their person… and shall in all cases be treated humanely.” and continues providing a distinct set of basic rights to persons “deprived of their liberty for reasons related to the armed conflict.”[59]

            These Articles form the privileges and duties afforded civilians and combatants – here labeled participants in an internal armed conflict.  The privileges and duties are largely the same, here, however;  there is no affirmative definition as to who qualifies for each group.  The language used to define each category is pinned to the idea of “participation” in an armed conflict.[60] 

            When realistically applied to internal conflicts, the word ‘participant’ creates significant ambiguity.   Temporally, internal conflicts are often sporadic clashes, they exist as isolated flare-ups where hit and run tactics are employed rather than sustained fighting.  The actors in an internal conflict often assume much more ambiguous roles.  Guerrilla fighter are often soldiers by night while leading regular civilian lives during the day. This lack of continuity undermines any practical application of the term ‘participant.’  First, it is not clear when a participant has “ceased to take part in hostilities” as described by Protocol II, Article 4(1). And second, it is very difficult to determine who counts as a participant to begin with.

            Determining when a participant has ceased to take part in hostilities is sometimes referred to as the revolving door standard.[61]  Do participants cease to be participants the moment they lay down their weapons and resume their lives as civilians?  Can insurgents who attack a police convoy at night only to resume their jobs as shopkeepers or farmers the next day are afforded the same protections granted to civilians?  Such an interpretation of this Protocol would provide be a huge advantage to insurgents who could then choose whether they wanted to be treated as civilians or combatants and pick up a pitchfork or a rifle according to their best interest.                        

              Further, such an interpretation would provide incentive for insurgents to use their civilian status as a cloak from which to strike their adversary.  Likewise, it would be in the best interest of the advisory in this situation to attack the civilian population which doubled as insurgents by night.  Such incentives destroy the line between civilian and combatant and would undermine the purpose of defining their roles to begin with.[62]   

            The Commentary from Protocol II notes that an individual is not entitled to civilian status while his participation in an armed conflict lasts, after which he no longer poses a threat to the regime, and therefore cannot be a target.[63]  While this does not specifically address when participation begins or ends, it does link an individual’s participation to the danger she poses to a regime.  It can not be said that the insurgent by night described above ceases to pose a threat to her advisory upon resuming her job as a farmer or shopkeeper.  This is particularly true within the sporadic temporal context of an internal conflict. 

            Commentary for Protocol I though not directly applicable to Protocol II or Common Article 3 discusses the revolving door standard with more candor.  It notes that

Any interpretation which would allow combatants as meant in Article 43 to “demobilize” at will in order to return to their status as civilians and to take up their status as combatants once again as the situation changes or as military operations may require, would have the effect of canceling any progress that this Article has achieved.  Undoubtedly the success of guerrilla operations depends on the requirements of flexibility and mobility…However, this concept of mobility could not be extended into the legal field…[64]  


The drafters of the ICRC Protocol I were quite explicit in rejecting the notion of the farmer by day, combatant by night, part-time combatant, and interprets the combatant/civilian distinction in a binary manner.  While this may solve the revolving door dilemma it places even more significance on demarcating what qualifies as participation in the first place.  Considering the reality of internal conflict, where part-time participants exist, legally foreclosing that category makes it is very important to define clearly what actions make someone a combatant and which do not.  No such clarity exists. 

            There is no clear demarcation between what activities qualify as participation and thus change an individual from a civilian to a participant.  Individuals who provide insurgents with shelter, housing, provisions, information, weapons occupy an ambiguous area in international law.  Whether they are combatants or civilians largely depends on the situation in which they operate.  In dealing this issue, the ICTY declined to draw a bright line between acts which qualify as participation and acts which do not.[65]  Instead, it adopted a contextual approach based on the individual and the situation.[66]  Even in cases where retrospect makes this delineation obvious, it is a problematic approach.  The reality of war, and particularly the reality of internal conflict, often forces individuals to makes decisions without the benefit of contextual factors.

Consider the position of insurgents who want to avoid violating jus in bello and exposing themselves to prosecution.   How is a guerrilla fighter, in the midst of an insurgency to know whether a spy providing military information to the opposing government has reached the necessary threshold of involvement to qualify as a participant in the conflict?  If the guerilla fighter is left to guess, and kills an individual who later is deemed to be a civilian, the guerrilla fighter is then exposed to criminal liability.  Reducing acceptable behavior to guesswork is not the purpose of jus in bello.

            The reality of internal conflicts, and all conflict, is that there are at least three categories of individuals.  (1) clearly legal target, (2) clearly illegal targets, and (3) ambiguous targets.  It is this last category which, the current framework – particularly as it applies to internal conflicts – is incapable of realistically defining and regulating.  The final section of this analysis turns to this problem.  First, it examines the requirements of both the duty to discriminate and the duty of proportionality, explores problems with the current framework, and finally suggests a new approach.

B.   Duty to Discriminate

            The duty to discriminate is based in the law’s binary approach to categorizing individuals.  This duty requires those taking part in an armed conflict to distinguish between civilian and combatants, and target only combatants.  The duty to discriminate derives its authority form the Geneva Conventions and the Additional Protocols.   For international conflicts, language in Protocol I explicitly forbids attacking the civilian population or any individual civilians.[67]  The Fourth Geneva Convention entitled “Relative to the Protection of Civilian Persons in Time of War,” outlines further protections enjoyed by civilians.[68]   Attacks are further limited by Protocol I to those which strictly further military objectives.[69]  Likewise with internal conflicts, language in Common Article 3, “protects persons taking no part in hostilities, including members of armed forces who have laid down their arms...”[70]  Common Article 3 succinctly lays out the duty to discriminate stating that it, “requires combatants during and armed conflict to avoid targeting civilians and target only enemy combatants. “[71]  Thus, in both international and internal conflicts, individuals who qualify as combatants are required to discriminate between civilians and combatants and have a duty to shape their behavior so as to avoid targeting civilians. 

            When civilians are killed because of an armed conflict, an immediate question arises were the civilians killed, the target of the attack.  If so, then the combatants responsible for their death are criminally liable.  If not, however, then the combatants responsible may avoid liability if it is established that the means they used to carry out the attack were proportional to the military objective they hoped to achieve.  In such a situation, the duty of proportionality would absolve combatants from liability.   

C.   Duty of Proportionality

            As the label suggests, the duty of proportionality obligates participants in armed conflict to shape the means employed during an attack to be proportional to the military objective they hope to achieve.  This links the resulting civilian casualties of a military action with the actions objective.[72]  Significantly, this duty does not necessarily criminalize the killing of innocent civilians.  Instead, it recognizes that when military forces go to war, the process of conducting war often places civilians in danger.  Military forces make mistakes and innocent civilians get caught in the cross-fire and are killed as a result of war.  In a 30 November 2005 decision, the ICTY noted this fact commenting that, “history confirms, regrettably, that wartime conduct will often adversely affect civilians.”[73]  The duty of proportionality can be thought of as channeling the inevitable consequences of war into narrowest possible avenues such as to avoid as much collateral damage as possible.  If a target is legitimate, and the narrowest means are employed to achieve the target, resulting collateral damage –including civilian death – is not criminally punishable. The following examples better illustrate this point.

 During an armed conflict, if a guerrilla organization launches an attack aimed at securing a town, and targets a police compound, the guerrillas would not be liable under international criminal law if several civilians are killed in the resulting crossfire, because the means employed are proportional to their objective.  The town here qualifies as a legitimate target, the means employed by the insurgents are bullets, aimed at police officers not civilians, and are proportional to the objective the insurgents sought.  The resulting civilian casualty from such an action therefore becomes a legally permissible ‘horror’ of war.

            This concept was illustrated during the 1999 NATO air strikes against Serbian units in Kosovo.  During that campaign, a NATO airplane mistakenly dropped a bomb on a passenger train killing scores of civilians.  In considering whether to bring charges against NATO for war crimes, the ICTY Office of the Prosecution first had to determine whether the casualties resulting from the errant bomb were proportional to the military objective NATO sought to accomplish.[74]  Here, an investigation found that the accident had resulted from the plane operating at a very high altitude which made it difficult to distinguish a commuter train from a military target.[75]  The plane had been flying at such an altitude in order to avoid being shot by anti-aircraft guns.[76]  The Prosecution ultimately decided that no liability could be attached to the action.  The pilot had been pursuing a legitimate military target – what he thought was a military supply line – and his mistake resulted from an act of self preservation responding to the context in which he flew his plane. (i.e. over a war zone with hostile anti-aircraft guns) This example illustrates two significant points.  First, when pursuing a legitimate target, the law does not require an individual to jeopardize his or her life in order to employ the narrowest means possible to minimize potential collateral damage.[77]  Second, death of innocent civilians, even when it results from a gross mistake, does not necessarily imply a crime.[78]

            Contrast those examples with the following.  Consider again the first hypothetical involving an armed conflict and a guerrilla group.  If instead of launching an attack on the police station, guerrilla fighters pack bombs into their cars and drive them into the police compound destroying the entire block, the guerilla fighters responsible for the act become liable for their actions under international law.  The scope of the harm certain to occur from their action is far greater than reasonably necessary to achieve the objective of destroying the police station.  The duty of proportionality clearly is violated.

            These examples offer relatively clear delineations between legitimate and illegitimate targets and the proportionality of the means used in achieving them.  Not all instances are so clear cut and a problem arises in trying to determine exactly where the line between legitimate and illegitimate military targets and proportional and disproportionate means exists.  This question implicates both the duty to discriminate and the duty of proportionality and can be summed up as (1) what is a legitimate target, and (2) what are proportional means.

The first question, ‘what is a legitimate target,’ is the threshold issue.  If the answer here is negative, then the means employed become irrelevant; the action is illegal.  Conversely, if the target is legitimate, then the means are proportional if they are pursued as narrowly as possible. (i.e. you haven’t killed a fly with a hammer)  Answering the first question, however, is often difficult because the line between legitimate and illegitimate target is not clearly marked.  This is particularly the case in internal conflicts which do not benefit from the clear definitions laid out in Article 4(A) of the Third Geneva Convention.  Internal conflicts have only Common Article 3’s language providing protection to individuals who are not taking part in armed hostilities.  The line between legitimate and illegitimate targets thus hinges on what qualifies as participation in an internal conflict.  The remainder of this analysis addresses this question.   

Commentators, such as Rachael Kerr, have speculated that the line largely flexes depending on the political climate in which the event takes place.[79]   This approach, however, does not provide much clarity to fighters proactively hoping to shape their conduct to conform with the law.  William Walker offers a more concrete approach suggesting that operations are proportional when their ultimate goal does not relate to human beings but rather is conducted with a quantifiable military objective in mind.[80]  This definition suggests a Kantian notion of forbidding individuals to be used as means to an end.   While this is not a perfect approach, to understand better where the line between legitimate and illegitimate target exists it is helpful to briefly consider some of the moral underpinning jus in bello.

 Jus in bello is not meant to prevent civilian casualty; implicit in the concept of jus in bello is the acceptance that civilians will be adversely affected – and killed – during war.  The moral aims of the doctrine are twofold; (1) to protect innocent civilians from becoming the targets of military actions[81] and (2) to channel the use of deadly force as narrowly as possible around enemy targets.[82]  Considering these aims together, the enemy targets mentioned in the second goal can never be innocent civilians. This is William Walker’s point; even if the execution of 10,000 civilians would provide a distinct military advantage, and is carried out as narrowly as possible, it violates the duty to discriminate.  This suggests that the military advantage an action creates is not the determining factor in distinguishing a legitimate target from an illegitimate one.  More important in the analysis is the immediate threat a target poses to a combatant.  A combatant cannot target an individual to pursue some other end; a combatant can only target an individual if that individual poses a threat.  Thus, a police compound and the police officers in is, would be legitimate targets during a guerrilla war because the existence of a police compound poses a threat to the existence of a guerrilla group.[83]  Extending this suggests that the goals of jus in bello  are (1) to protect innocent civilians, and (2).to limit the use of deadly force to situations where the target presents an immediate military threat.  In those situations jus in bello privileges collateral damage that occurs provided the deadly force is employed as narrowly as possible.  The line between legitimate and illegitimate targets is linked to the threat the target presents more than to the military advantage which could be achieved.

            The Commentary on Protocol II, suggests this as well. In discussing what qualifies as  participation, the Commnetary states that,  “direct participation in hostilities” implyies “a sufficient causal relationship between the act of participation and its immediate consequences.”[84]  This suggests a correlation between the act and its foreseeable consequences not unlike the proximate cause requirement of Tort Law. Participation in hostilities therefore appears to require, casual proximity to the foreseeable consequences and an intent to damage the enemy.[85]

V.   Threat provides the context for defining the duties

            As stated above, when adjudicating between legitimate and illegitimate targets, the ICTY has adopted a contextual approach to determining the legality of a target.[86]  This suggests that the ICTY considers such things as the parties involved in a conflict and the threat an object poses to one Party in assessing whether that object is a legitimate target.  With respect to the ambiguous targets,[87] those whose legality is not entirely clear- this contextual approach creates at least two problems.  First, the threat a potential target poses varies greatly depending on the brutality of the regime opposing an insurgency, and second as pointed out above, it is very difficult for combatants to assess the threat of a target contextually without the benefit of retrospect.     

A.     Threat to insurgents varies with regime brutality and openness

            In an internal conflict, the danger an insurgent faces varies widely, depending on the regime opposing the insurgency.  The consequences of conduct by a regime collaborator to an insurgency correspondingly vary.[88]  In a particularly repressive regime,[89] a collaborator providing the government with a list of insurgents would be hard pressed to argue that he or she does intend to cause serious harm to the individuals informed upon.  In such a situation when the government arrests and executes the listed individuals, the collaborator would be the proximate cause of the individual’s death, posing a direct threat to his existence; further, the collaborator would possess the requisite mens rea – intent – so as to qualify as a participant in the hostilities.  This would not be the case under a more liberal regime where a collaborator could argue that he or she assumes the information would be used to investigate an alleged threat and the individuals listed would be detained as prisoners of war or receive a fair and transparent trial. 

The culpability of the two collaborators described above is dissimilar.  Targeting and killing a collaborator under a repressive regime seems much less reprehensible than targeting the same individual under a liberal government.  Recall from above, the moral aims of jus in bello; (1) to protect innocent civilians, and (2).to limit the use of deadly force to situations where the target presents an immediate military threat. The first collaborator is not innocent, he or she has deliberately sought to cause great harm to an individual. Further, his or her actions create an immediate military threat to an insurgency.  Targeting such an individual would not compromise the moral aims of jus in bello and targeting such an individual should be permitted.

There is a need for flexibility in the line between acceptable and unacceptable targets in different situations.  The reality of conflict makes this essential to the practical application of international law.  What is also needed, however, is to communicate more clearly the contextual factors which determine which acts qualify as participation and further, to make that information available to combatants during a conflict.  

The Brutality Index

One way this could be accomplished is by creating and publishing an index which accounts for contextual factors in a conflict.  Specifically, this would consider factors such as the violent nature of the regime opposing an insurgency, (a brutality rating), and the existence of a functioning civil society and alternative legal means for insurgents to express discontent.[90]  Repressive regimes which deal with dissent in an opaque draconian manner and stymie traditional channels of expressing discontent (newspapers, public forums, judicial review of governmental actions; petitions to legislatures, starting new political parties, becoming a political candidate) would receive a high Brutality Index.  Conversely, those that dealt with dissent humanely, openly, and responsibly would generate a low Brutality Index.  The Brutality Index would be compiled by human rights groups on the ground and the United Nations, which would produce a report outlining specific threats, created by the regime, to human rights and freedoms.[91]  The Index would then shape the line between legitimate and illegitimate targets.  Ambiguous targets such as regime collaborators would become acceptable legal targets under a particularly repressive regime while collaborators under a more liberal government would remain illegal targets.

            The relative power of a regime is vastly greater than that of an insurgency.  As a result, a repressive regime uniquely magnifies the threat posed by ambiguous actors in a conflict.  The result of creating a Brutality Index would be to link explicitly the line between legitimate and illegitimate targets to the intentional threat specific individuals pose 

A. Application to specific conflicts


Applying the Brutality Index to some tangible examples is helpful.  Consider the Iraqi insurgency.  As the U.S. invasion wound down and an interim government was appointed, the landscape in Iraq fundamentally changed.  Available alternative means of expressing dissent sprang up and diversified media channels opened, including pointedly anti-establishment channels such as al Jazira.  Further, a fledgling legal system began to develop to try both alleged criminals and individuals accused of terrorism.  As a result, the Brutality Index after the interim government took power would begin to fall.  The sphere of acceptable targets would correspondingly decrease.

Insurgent violence in Iraq did not reflect this.  Numerous attacks were carried out on individuals who were connected to the armed forces of the interim government.  Military contractors and government employees were targeted in an attempt to destabilize the regime.  Under the proposed framework these attacks would be decidedly illegal.  Further, insurgents would know that these acts were illegal.  The Brutality Index report would point out the availability of alternative means for insurgents to express discontent.  Dissidents could form political parties, file lawsuits, run for offices, vote, and use television, newspaper, churches, and mosques to communicate their grievances.  Additionally, the formation of a government that appears committed to pursuing the rule of law and would all contribute to a relatively much lower Brutality Index than had previously existed in Iraq.  A report might highlight areas where the interim government fell short in this regard, but in general the Brutality Index would be falling. Such a finding would be supported by the subsequent ratification of the Iraqi Constitution, and a democratic election of an Iraqi government.      

Of particular concern under this framework would be events such as Abu Graib, and other allegations that prisoners were tortured in Iraq.  These events suggest that ambiguous targets pose a much greater threat to insurgents than they otherwise would.  Such events would dramatically increase the Brutality Index of the regime responsible, here the United States, and the graphic evidence of mistreatment and torture would legitimize a broader range of justifiable targets.

Kosovo Liberation Army

The recent conflict in Kosovo provides another example with which to illustrate this framework.  Consider the Kosovo Liberation Army’s (KLA’s) efforts to resist Serbia’s ethnic cleansing campaign and to obtain independence.

In 1974, Tito grant of autonomy began a period of self governance in Kosovo unmatched in generations.  Albanians elected a local government, had their own police force, and sent their children to an Albanian University.  The Yugoslav regime tolerated a range of non-violent dissent in the province including the non-violent means typified by Jusuf Gervalla’s use of Albanian folk music to convey revolutionary sentiment. During this period, Yugoslavia would have enjoyed a low Brutality Index and Albanian insurgents hoping to break free from Yugoslavia would have been limited to a very narrow sphere of acceptable targets.  Albanian insurgents during this time period virtually did not exist, and those that were operating used

            Following Tito’s death, in 1981 Yugoslav authorities violently cracked down on the Albanian population and individuals were thrown in jail for acts as benign as passing out pro-Albanian pamphlets and fliers.  Eight years later in 1989, Yugoslav president Slobodan Milosevic revoked Kosovo’s autonomy dissolved the Albanian parliament, disbanded the local police force, and forbid the University of Prishtina from conducting classes in Albanian.  During the 1990’s, Albanians were routinely stopped and searched on the street.   Possession of any materials considered revolutionary resulted in imprisonment, beatings, or worse.

 The Yugoslav government’s Brutality Index would have been demonstrably higher in 1980’s than it had been a decade earlier and it would have continued to increase throughout the 1990’s. 

 The Kosovar Albanians reacted to the changing political landscape by organizing resistance to the regime which ultimately took the form of the Kosovo Liberation Army.   

            As attacks continued, the Serbian regime conducted a wide series of arrests.  Authorities threw individuals in jail, where they were beaten, held without trial, and sometimes killed for rumored involvement in the KLA.  As the government’s level of violence soared, an extremely wide scope of violence against ambiguous targets would to be justifiable, better reflecting the threat such individuals greater to the KLA.   


Practical Application

                To demonstrate this frameworks function, consider a Kosovar[92] growing up in the 1970’s.  The relatively low Brutality Index achieved by Tito’s regime would make violence towards ambiguous targets illegal.  Thus, if the Kosovar were to start an insurgency during the 1970’s and the dissent reached the requisite level to be considered an ‘armed conflict,’ targeting ambiguous actors, such as collaborators, would be a violation of jus in bello and the Kosovar would be liable for such actions under international law.

            During the 1980’s and early 1990’s, in the face of a rapidly rising Brutality Index, if an armed conflict began with insurgents, a much wider sphere of violence would be permissible.  Civilians who provided logistical support, information, or a launching point for the MUP, would become targets as their actions directly and intentionally threatened the existence of the KLA   A Kosovar targeting such individuals would not be liable for his actions under international law.

            Finally, during the beginning of 1998, in the wake of the Jashari Massacre where over 50 Albanian civilians executed by Serb police in Precaz, the sphere of justifiable violence would dramatically increase again, justifying a wide range of violence against ambiguous targets.  Collaborators and informants would become legally permissible targets and an insurgent who killed such individuals would not be criminally liable.  Indeed, just following the Jashari Massacre the KLA could publicly acknowledge the soaring Brutality Index and clearly articulate the types of individuals they intended to target as a result.  This would serve notice that the KLA intended to legally address the growing threat ambiguous actors created, and provide those individuals an opportunity to leave or risk the repercussions.    

The United States  

With this in mind, events such as the ‘torture memo,’ and the detention without trial of prisoners in Guantanamo Bay should give the United States pause for concern.  These events could reasonably be used to justify a broader sphere of violence against American forces both in Iraq and domestically.  The United States has a long standing record of freedom of expression, a range of different media outlets through which to express criticism, and a transparent and fair judicial system. The Brutality Index and the justifiable sphere of violence in this setting are extremely low.  Recent events, however, have witnessed severe inmate abuses.  Instances such as the notorious torture at Abu Graib, the indefinite suspension without trial of prisoners at Guantanamo Bay, and an August 2002 memo from the U.S. Justice Department finding that laws outlawing torture could not bind George Bush,[93] would all serve to raise the nation’s Brutality Index.  

            As these events accumulate, a broader sphere of violence would be justifiable.  Ambiguous actors facilitating United States activities become a much greater threat to an insurgent if a captured insurgent is not subsequently treated as either a prisoner of war or a criminal provided access to the judicial system.  It would become harder and harder for informants and collaborators aiding the United States to argue that they did not intend for their actions to cause great harm to insurgents.  In such a situation, they would become combatants and legal targets



A Brutality Index would create a disincentive for failing to provide due process to suspected insurgents, closing alternative avenues available for expressing dissent, and resorting to violent methods of law enforcement.   

Confronted with an internal conflict, a government would find it advantageous to maintain open societies where individuals could express discontent freely.  Dealing with dissent responsibly and humanely would generate a very low brutality index and thus greatly reduce the sphere of violence that the international community considered justifiable.  Likewise, a transparent criminal system would become desirable in order to demonstrate to the international organizations producing the brutality index report that a low index is appropriate.

Problems with the suggested approach

            Several problems potentially could arise under this approach and they are addressed in detail below.  First, the framework could be criticized as providing too many advantages to insurgents.  This criticism, however, only stands when an insurgency opposes a violent regime.  Otherwise, the Brutality Index provides no benefit to the insurgent whatsoever.  In situations where an insurgency does benefit from this framework, and is able to target a wider sphere of individuals, the advantages the insurgency gains are warranted.  A regime’s uniquely disproportional advantage in terms of resources, manpower, weaponry, and military effectiveness justifies providing an insurgency with some legal advantages when the regime’s power is abused.

1. Private Constriction of Civil Society

Another dilemma which must be addressed, is the fact that in an internal conflict government is not the only institution capable of engaging in targeted violence against dissent and stifling civil society.  Private parties can effectively do so as well.  Should this justify a greater level of violence against the government?  Against the media?  The 2004 American presidential election was rife with accusations that media conglomerates had refused to air programs critical of the incumbent government seeking reelection.  Clearly in this example such behavior could not justify a greater sphere of violence against anyone.  It is not difficult to imagine a different scenario, however, where the answer is not as clear cut.  During part of the 1990’s, for instance, private Kosovo newspapers would not report KLA activity or discuss the group, having been directed by a local political party not to do so.

2. Private acts of violence

 Similarly, a problem exists where brutality towards individuals is carried out privately or simply condoned by the government.  Consider the Southern United States in the first half of the twentieth century where community discrimination and violence towards African Americans was largely not addressed by the state mechanisms established to protect Southern blacks.  A similar system of abuse and discrimination was apparent in Kosovo throughout the 1990s.  Albanians were routinely beaten, harassed, and humiliated by Serbs without any means of official recourse. 

            Both of these examples implicate problems properly measuring a government’s role in the private suppression of dissent or in private brutality.  A convenient solution is not to measure the government’s role at all but to build private acts of brutality or suppression into the brutality index.  Thus, regardless of who suppresses media outlets or violently targets a specific part of the population, if the government does not appropriately address such behavior, it would be reflected in the brutality index. Holding the government accountable for inaction in protecting its citizens create the right incentives.  Building private acts into the calculation of a brutality index would accomplish this regardless of whether or not those private acts were originally sponsored by the government.  The government would have an incentive to root out private oppression; it also would lack an incentive to “contract out” its own oppression.


            If the justifiable sphere of violence varies depending on the brutality of a specific regime, drawing the line between justifiable violence still creates a challenge. It is much easier to assess retrospectively how brutal a regime was and much harder to determine how brutal a regime is.   

            As described above, drawing any useful line between legal and illegal targets is difficult.  The current system centers on whether an individual’s activities qualify as participation in an armed conflict, without clearly laying out what participation entails. This term, ‘participation’ is an equally difficult measure to use during a conflict, however, defining a legal target in terms of participation also suffers from not being a realistic standard for a guerilla conflict—at least as it has been interpreted. 

            The reality presented by a guerrilla conflict can make activities that normally fall outside the scope of participation as dangerous to a guerrilla as an enemy soldier. When information  provided to a regime results in hundreds of individuals being rounded up by the government and executed, it is necessary for an insurgency to attempt to eliminate the source of the information.  Regardless of the legality, this course of action must be followed if the insurgency is to survive.  It becomes problematic when the traditional lines distinguishing legal and illegal targets are applied in situations where an insurgency cannot realistically both comply and survive.   This disconnect reduces the effectiveness and legitimacy of international law governing insurgency.  The reality presented in many conflicts, specifically those in which insurgents oppose a violent regime, is such that regime informants, spies, and collaborators can seriously damage an insurgency.  If neutralizing that threat immediately makes an insurgent a war criminal, there is little subsequent incentive to abide by any other laws of war.

            A flexible line, described above, considers the reality confronting a guerrilla and adjusts acceptable behavior accordingly.  While the problem of proactively contemporaneously drawing a clear line between legal and illegal targets remains a concern, a similar concern exists under the established laws which currently suffer by failing to consider the reality with which a guerrilla fighter is faced.








[1] Henry H. Perritt, Jr., Fire for the Colors: The Inside Story of the Kosovo Liberation Army's Fourth Generation War ch. 9A (forthcoming 2006) (manuscript dated 24 Oct. 2005 at 11, on file with author) [hereinafter "Fire"].

[2] While jus ad bellum was traditionally applied to nations, the changing face of conflict in the 20th Century forced the doctrine to expand to encompass other collective groups such as insurgent groups, guerrilla organizations, etc.  This analysis uses the term “nation” and acknowledges that it is not exclusive.

[3]“Self defence” used in this context is specifically defined in Article 51 of the U.N. Charter

[4] See UN Charter Art. 1

[5] Fire at 12

[6] id

[7] Orna Ben-Naftali and Keren R Michaeli, We Must Not Make a Scarecrow of the Law : A Legal Analysis of the Israeli Policy of Targeted Killings, 36 Cornell Int’l L.J. 233, 255. (2003)

[8] Geneva Conventions of 12 August 1949 for the Protection of War Victim, Aug. 12, 1949, 6 U.S.T. 3114. The Geneva Conventions consist of : Geneva Convention (I) for the Amelioration of the condition of Wounded and Sick in the Armed Forces in the Field, Aug. 12 1949, 6 U.S.T. 3114, (hereinafter Geneva Convention I); Geneva Convention (II) for the Amelioration of the condition of the Wounded, Sick, and Shipwrecked Members of the Armed Forces at Sea, Aug. 12, 1949, 6 U.S.T. 3217, (Hereinafter Geneva Convention II); Geneva Convention (III) Relative to the Treatnet of Prisoners of War Aug. 12, 1949, 6 U.S.T. 3316, (Hereinafter Geneva Convention III); Geneva Convention (IV) Relative to the Protection of Civilian Persons in Time of War, Aug. 12 1949, 6 U.S.T. 3516, (Hereinafter Geneva Convention IV); Hague Convention No. IV Respecting the Laws and Customs of War on Land, Oct. 18 1907


[10] Major Ian Corey, The Fine Line Between Policy and Custom : Prosecutor v. Tadic and the Customary International Law of Internal Armed Conflict,166 Mil L.Rev 145, 149 (2000)

[11] Id.

[12] Id.

[13] Id.

[14] Geneva Convention I, II, III, IV, Art. 2 (Hereinafter Common Article 2)

[15] Geneva Convention I, II, III, IV, Art. 3 (Hereinafter Common Article 3)

[16] Statute of the International Criminal Tribunal for the Former Yugoslavia (as amended through 19 May 2003 by UN Security Council Resolution 1481, Art. 2-5.

[17] Prosecutor v Tadic, Appeals Chamber Judgment, 15 July 1999, para. 66.

[18] Id.

[19] Id.

[20] Id. at para. 70

[21] Fire at 39

[22] Id.

[23] Id.

[24] Michael Schmitt, Humanitarian Law and Direct Participation in Hostilities by Private Contractors or Civilian Employees, 5 Chi. J. Int’l L. 511, 522 (2005)

[25] Jean-Jacques Rousseau 1 On Social Contract 90 (1762)

[26] And more recently expanded to include conflicts within States.

[27] Nathan Berman, Privileging Combat? Contemporary Conflict and the Legal Construction of War, 43 Colum. J. Transnat’l L. 1, 9 (2004)

[28] Berman, supra note 27, at 9

[29] See Berman supra note 29, at 9 & n. 15

[30] Berman, supra note 29 at 9

[31] Note : a distinction here between being targeted and being killed. Civilians cannot be the target of an attack, however, they can be killed during a conflict without any violation of international law, provided the proximate cause of their death was an attack with a proportional objective to the civilian damage it caused.  

[32] in which case they would become POW’s and held for the duration of the conflict

[33] For instance, it is not clear who fired the bullet; ambiguous evidence as to identity; alibi; possible self-defense

[34] See Berman at 10-14.

[35] Common Article 2

[36] Common Article 3

[37] Id.

[38] This assumes the latter is not a High Contracting Party

[39] Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Armed Conflicts, June 8, 1977, 1125 U.N.T.S. 3 [hereinafter Protocol I]

[40] Id.

[41] Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of Non-International Armed Conflicts, June 8, 1977 1125 U.N..T.S. 609 [hereinafter Additional Protocol II]

[42] Specifically, internal conflicts are regulated by Common Article 3 and supplemented by 1977 Additional Protocol II.  Meanwhile, the Geneva Conventions minus Article 3 govern international conflicts and are supplemented by 1977 Additional Protocol I which broadens the scope of international conflict to include “armed conflicts which peoples are fighting against colonial domination and alien occupation and against racist regimes in the exercise of their right of self-determination.”  These are often shortened to CAR conflicts

[43] Geneva Convention III, Art. 4

[44] Protocol I Art. 43

[45] Schmitt, supra note 24, at 523  (Note article 4(A) (6) of the Geneva Conention 3 is not relevant to this discussion.)


[47] Hereinafter referred to together as international conflicts.

[48]Article 4(A)(1) : Members of the armed forces of a Party to the conflict as well as members of militias or volunteer corps forming part of such armed forces.

[49] see Schmitt at 57-58

[50] Aricle 4(A)(2) :  Members of other militias and members of other volunteer corps, including those of organized resistance movements, belonging to a Party to the conflict and operating in or outside their own territory provided that such militias or volunteer corps, including such organized resistance movements, fulfill the following conditions :

            (a) that of being commanded by a person responsible for his subordinates;

            (b) that of having a fixed distinctive sign recognizable at a distance;

            (c) that of carrying arms openly

            (d) that of conducting their operations in accordance with the laws and customs of  war.


[51] Members of regular armed forces who profess allegiance to a government or an authority not recognized by the Detaining Power.

[52] Geneva Conventions III Article 4(A)(3)

[53] Geneva Convention III Article 4(A)(6)

[54] Conceivably, a Serb resident of a Serb village resisting an organized KLA movement into the village with his hunting rifle could qualify here.  Such an visible collaborator may also qualify under 4(A)(2).   

[55] Protocol I Art. 50.1 : defines civilians as any person who does not belong to one of the categories referred to in Article 4(A) (1), (2), (3), and (6) and in Article 43

[56] Berman supra note 27, at 44.

[57] Berman supra note 27, at 37 – an individual who does not belong to any of the categories of Article 4(A) but is still attempts to kill enemy soldiers would qualify

[58] Common Article 3

[59] Additional Protocol II Art. 4(1) and Art. 5(1)

[60] Note: while this article refers to participants and not combatants, the two terms will be used interchangeably here as the difference between them is not relevant to this paper

[61] Schmitt supra note 24, at 536

[62] Schmitt supra note 24, at 535

[63] Yves Sandoz, Christophe Swinarski, and Bruno Zimmernam, eds, Commentary on Additional Protocols of 8 June 1977 to the Geneva Conventions of 12 August 1949  p. 1943 (ICRC 1987) [hereinafter ICRC Protocols Commentary]

[64] Berman, at 53 citing ICRC Protocols Commentary at 1677-78 

[65] Tadic, Appeals Chamber Judgment, 15 July 1999 at para 616

[66] Id.

[67] Additional Protocol I Art. 51.2

[68] Geneva Convention IV

[69] Additional Protocol I Art. 52.2

[70] Common Article 3, para. 1

[71] Common Article 3, para. 1

[72] Fire at 114

[73] Prosecution v Limaj, Trial Chamber Judgment 30 November 2005, para. 210


[75] id.

[76] Id, at 202

[77] Id. at 201-202

[78] Id. at 201

[79] See id at 175-206

[80] William Walker, The International Law Applicable to Guerrilla Movements in Internal Armed Conflicts : A Case Study of Contra Attacks on Nicaraguan Farming Cooperatives, 21 N.Y.U.J. Int’l L. & Pol. 147, 156 (1988)

[81] This is accomplished by the duty to discriminate

[82] Accomplished by the duty of proportionality

[83] Presuming the armed conflict has developed to the point where the police are making some efforts to stamp out the guerrilla activity

[84]ICRC Protocols Commentary at P 4787

[85] Schmitt supra note 24, at 87

[86] Supra at 21

[87] Regime collaborators, spies, informants, weapon suppliers, etc.

[88] An individual who funnels information to the regime concerning the operations and structure of the insurgency

[89] imagine North Korea with an active insurgency

[90] This index could, for instance be modeled after Arend Lijphart’s assessment of democracy in his work Patterns of Democracy.  

[91] See Annex A 

[92] Here meaning Kosovo Albanian

[93] Mike Allen and Dana Priest, “Bush Authorizes Torture” The Washington Post Wed. Jun 9, page A03