1. Introduction
The invasion of Iraq by the United States of America in 2003 tested many existing legal norms and sparked much debate as to the relevance of the various components of international law, including the law of occupation. After the invasion, it quickly became apparent that the US (and coalition forces) would remain in Iraq for a period of time.Footnote 1 Questions regarding the application and implementation of the law of occupation inevitably arose. The answer to these questions became murky as the structure of the US’s post-invasion administration of Iraq emerged.
In the aftermath of the invasion and capitulation of the Iraqi armed forces, the permanent representatives of the US and the UK wrote to the President of the United Nations Security Council.Footnote 2 Generally, the intention of the letter appears to be to provide an update on the state of affairs in Iraq and to seek the support and involvement of the UN. From a law of occupation perspective, the most critical elements of this letter are as follows: (i) it does not refer to the US or the UK as occupying powers and the word occupation does not appear in it; (ii) it contains a declaration from the US and the UK that they will abide by their obligations under international law in post-conflict Iraq; and (iii) it refers to the creation of the Coalition Provisional Authority (CPA).Footnote 3 In respect of the CPA, the letter states:
[i]n order to meet these objectives and obligations in the post-conflict period in Iraq, the United States, the United Kingdom and Coalition partners, acting under existing command and control arrangements through the Commander of Coalition Forces, have created the Coalition Provisional Authority … to exercise powers of government temporarily.Footnote 4
In response to the letter, the Security Council passed Resolution 1483.Footnote 5 In contrast to the letter, the Security Council Resolution contains a statement (albeit perambulatory) affirming that Iraq is in a state of occupation and that there are occupying powers.Footnote 6 Specifically, the Resolution referred to the US and the UK as ‘occupying powers under the unified command of the “Authority”’.Footnote 7 Subsequently, the CPA engaged in a period of governance over Iraq,Footnote 8 exercising executive, judicial, and legislative authority.Footnote 9
The Security Council Resolution plainly affirmed the applicability of the law of occupation to the situation in Iraq. However, it did not explicitly address the relationship between the CPA and the law of occupation, or the legal relationship between the CPA and the occupying powers.Footnote 10
On first reading of Resolution 1483, the CPA appears to be merely an instrument of the US and the UK used in their occupation of Iraq.Footnote 11 For some commentators, Resolution 1483 is sufficient evidence of the fact that the CPA was under the authority of the US and the UK and no further investigation is required as to the characterization of the relationship between these parties.Footnote 12 Other commentators suggest that the CPA itself was the belligerent occupying power.Footnote 13 That is, it is not clear whether the CPA is part of the US state or is instead, a non-state actor.
The evident lack of clarity regarding the CPA and its status under law (both domestic and international) gives rise to two significant questions: (i) was the CPA, as a result of its relationship with the US, bound by the law of occupation; and (ii) now that the CPA has dissolved, is the US responsible for the wrongful acts of the CPA?
To understand the legal relationship between a state and a non-state actor during an occupation, both international courts and scholars have turned to the theoretical framework of ‘occupation by proxy’. The theory is that one state can occupy another through a non-state actor or agent. Instances of occupation by proxy can be found throughout history, for example, it is argued that Nazi Germany occupied several European states through a number of local puppet administrations during the Second World War.Footnote 14 Following the break-up of Yugoslavia, the International Criminal Tribunal of the Former Yugoslavia (in more than one instance) considered the ability of a state to occupy another through a non-state actor. Scholars have sought to expand on these decisions to create a legal framework for the theory of occupation by proxy.Footnote 15 I submit that the theory of occupation by proxy can be applied to the CPA and the US during its occupation of Iraq to ensure that the CPA, is in fact, bound by the law of occupation and that the US is responsible for any breaches of the law of occupation by the CPA.
To support this position, I will first examine the relationship between the CPA and the US through the lens of the law of state responsibility, the law of occupation and Resolution 1483. Next, I will establish a legal framework for the theory of occupation by proxy. Then, I will apply the legal framework to the CPA and the US in Iraq and conclude by exploring the potential implications in terms of responsibility. In order to limit the scope of this article, the relationship of the UK and the CPA will not be considered.Footnote 16
2. The legal status of the CPA and its legal relationship with the US
There are a number of avenues under international law by which to examine the legal status of the CPA and its legal relationship with the US. The first avenue that this article will examine is the law of state responsibility. The law of state responsibility contains the principles for determining when a state will be accountable for a breach of an international obligation. As part of this legal framework, the law of state responsibility also determines when a state will be responsible for the conduct of a person or entity through which a state may operate. The second existing legal framework that will be examined is the law of occupation. The occupation of Iraq is at the core of the relationship between the CPA and US, therefore, the law of occupation may theoretically govern this relationship. Finally, the legal status of the CPA and its relationship to the US will be examined in the context of Resolution 1483 where the Security Council expressly acknowledged the CPA.Footnote 17
2.1 The law of state responsibility
In certain circumstances, the law of state responsibility, as reflected in the International Law Commission’s Articles of State Responsibility (ILC Articles), will attribute the conduct of a person or entities to a state. In order to establish responsibility, the law of state responsibility seeks to determine the legal status of an entity or group of persons.
2.1.1 Organs and de facto organs of a state
No one entity can exercise all state functions at all times. The law of state responsibility recognizes this and explicitly acknowledges the role of organs of the state. Under the law of state responsibility, the act of an organ of a state will be attributed to the state.Footnote 18 Article 4(1) of the ILC Articles states that an act of any organ of a state will be considered an act of that state.Footnote 19
The ILC Articles provide little guidance as to what entity constitutes an ‘organ’. Article 4(2) of the ILC Articles states that ‘an organ includes any person or entity which has that status in accordance with the internal law of the State’.Footnote 20 Accordingly, an entity given the status of an organ under the domestic law of a state will be considered an organ pursuant to the ILC Articles. However, there may be instances where an entity is not afforded the status of an organ of a state under its domestic law but nonetheless, will be considered an organ of the state.Footnote 21 That is, an entity may be a de facto organ of a state.
A number of international courts and tribunals have grappled with the notion of de facto organs. A body of jurisprudence has emerged in respect of the test to be applied to determine if an entity will be considered a de facto organ of a state under international law. In the Nicaragua case,Footnote 22 the International Court of Justice (ICJ) applied the test of ‘complete dependency’ to determine if the rebel group, the contras, was a de facto organ of the US.Footnote 23 In this case, the Court held that the aid given by the US to the contras, although essential to the operation of the contras, was not enough to render the contras wholly dependent on the US.Footnote 24 In the Bosnian Genocide case the ICJ affirmed the threshold of complete dependency.Footnote 25 In this case, the Court stated that it is necessary to examine the facts and to ‘grasp the reality of the relationship’ when determining whether an entity is a de facto organ of a state.Footnote 26 In this instance, the Court found that the army of the Republika Srpska could not be described ‘as lacking any real autonomy’,Footnote 27 and accordingly, could not be classified as an organ of the Federal Republic of Yugoslavia (FRY).Footnote 28 Although the Court acknowledged the significant military, political, and logistical bonds between the army of the Republika Srpska and the FRY, it was quick to point out the division over strategy as well as the degree of independence exercised by the army of the Republika Srpska.Footnote 29 If the CPA is considered an organ or de facto organ of the US under the ILC Articles then any wrongful act of the CPA (including under the law of occupation) would be attributed to the US.Footnote 30
The CPA has been described as ‘all at once, an element of the Defense Department, a multinational organisation and a foreign government’.Footnote 31 At the time of its creation the legal status of the CPA under US domestic law was uncertain not only to members of the US Government but even to the head of the CPA, Lewis Paul Bremer III.Footnote 32 The lack of clarity regarding the legal status of the CPA derives, in part, from the uncertainty about how it was created.
US law permits an entity exercising federal power to be established by Congress (via an ‘Enabling Act’) or by the President exercising his or her executive power.Footnote 33 The CPA was not established pursuant to any Act of Congress.Footnote 34 Bremer was appointed as presidential envoy of President George W. Bush to Iraq on 6 May 2003.Footnote 35 On this basis it is argued that the CPA derived its authority from the appointment of Bremer as a presidential envoy.Footnote 36 This theory is strengthened by official US Government documentation which refers to the CPA as an entity of federal government.Footnote 37
However, there is an alternative theory which contends that the CPA was created by General Thomas Franks, the military commander of US forces in Iraq, pursuant to the laws of war.Footnote 38 In his message to the Iraqi people labelled ‘Operation Iraqi Freedom’, General Franks referred to the CPA as the entity that would be responsible for the governance of Iraq.Footnote 39 This theory was put forward by the Justice Department of the US in the case DRC, Inc. and Robert Isakson v. Custer Battles, LLC, et al., which argued that the CPA was established under the laws of war and occupation.Footnote 40 However, this argument was wholly rejected by the US District Court for the Eastern District of Virginia, which found that the CPA was not an instrumentality of the US Government under US law.Footnote 41 Although the judgment was overturned on appeal, the Appeals Court did not address the question of instrumentality.Footnote 42
Adam Roberts also dismisses the theory that General Franks created the CPA on the basis that it is not evident that the CPA was operating as an administrative body from the time of the General’s message.Footnote 43 Instead, the CPA assumed operations around the time that the US and UK wrote to the President of the Security Council in May.Footnote 44
In summary, there appears to be no conclusive evidence of the origins of the CPA’s authority and accordingly, whether the CPA is considered an organ of the US under US domestic law. Further, US courts have openly rejected the proposition that the CPA was an instrumentality of the US. However, even if not an organ of the US pursuant to US domestic law, the CPA may still be considered an organ of the US by applying the de facto organ test as set out in the Bosnian Genocide case.
It has been suggested that ‘the CPA was run by the US and took instruction only from Washington’Footnote 45 and that ‘in practice, the CPA was always treated as a part of the US federal government’.Footnote 46 The CPA’s ability to physically operate within Iraq was wholly dependent on the presence of US forces (over which it was not in command).Footnote 47 However, by way of contrast, although the US contributed funds to the CPA and its projects,Footnote 48 it ‘principally relied on Iraqi money to fund both reconstruction and Iraqi government operations’.Footnote 49 Furthermore, an examination of the decision-making process of the CPA suggests that although the CPA was subject to review and guidance by Washington, it often exercised a degree of autonomy and independence in its decision-making.Footnote 50
There does not appear to be any evidence that every single decision of the CPA was subject to rigorous oversight by the US administration. In respect of the use of Iraqi funds, James Dobbins noted that the CPA was unwilling to be constrained by Washington.Footnote 51 In other instances, the CPA resisted orders given by Washington and pursued its own policies. For example, against the wishes of Washington, the CPA conducted an enquiry into the ‘Oil For Food Programme’.Footnote 52 The CPA also rejected Washington’s order to close down Al-Jazeera’s operations in Iraq.Footnote 53 These case studies illustrate some of the CPA’s autonomy and independence from the US administration.
The degree of independence and autonomy of the CPA (both financial and strategic), suggests that the CPA would not meet the threshold of ‘complete dependency’ required by the Bosnian Genocide case and, therefore, would not be considered a de facto organ of the US pursuant to the law of state responsibility.
2.1.2 Proxies
As well as organs of states, the law of state responsibility recognizes other persons or groups of persons for whom a state will be responsible. These other persons or groups of persons are commonly known as ‘proxies’.
ILC Article 8 provides that ‘[t]he conduct of a person or group of persons shall be considered an act of a State under international law if the person or group of persons is in fact acting under the instructions of, or under the direction and control of, that State in carrying out the conduct’.Footnote 54 Could this article apply to the relationship between the CPA and the US?
2.1.3 Effective control and the Nicaragua case
The seminal case regarding the concept of a ‘proxy’ under international law is the Nicaragua case.Footnote 55 In the Nicaragua case, the ICJ examined whether the conduct of, and more specifically, a breach of international law by, the contras could be attributed to the US.Footnote 56 As a part of this assessment, the Court considered whether the contras were operating under the direction or control of the US. In doing so, the Court applied an ‘effective control test’.Footnote 57 Put simply, the Court found that for the US to be responsible for any breaches of international law by the contras, it would need to be established that the US was in effective control of the contras.Footnote 58 The Court outlined the financial assistance, training, planning and military support given by the US to the contras.Footnote 59 The Court also acknowledged the ‘general control’ by the US over the contras and the high level of dependency on the US by the contras.Footnote 60 However, notwithstanding these findings, the Court did not hold that the conduct of the contras could be imputed to the US as the US was not in effective control of the contras.Footnote 61
The ICJ decision in the Bosnian Genocide case suggests that for the purposes of attribution, the conduct of an entity must be considered on a case-by-case basis:
it must be shown that this “effective control” was exercised, or that the State’s instructions were given, in respect of each operation in which the alleged violations occurred, not generally in respect of the overall actions taken by the persons or groups of persons having committed the violations.Footnote 62
Consequently, it would be necessary to examine each individual act of the CPA to determine whether the conduct in question could be attributed to the US under the ILC Articles.
As a general comment, it would appear that the US often determined and co-ordinated the CPA’s policies. For example, on 23 May 2003, the CPA issued Order 2, decreeing the dissolution of certain entities in Iraqi society, including the army.Footnote 63 The decision to disband the Iraqi army appears to be the brainchild of senior members of the Defense Department.Footnote 64 The initial draft of Order 2 was prepared and approved by the Pentagon.Footnote 65 Bremer recounts that the process of disbanding the army and issuing Order 2 was ‘carefully coordinated … with the Pentagon’.Footnote 66 A similar process appears to have been undertaken in respect of the De-Ba’athification measures implemented by the CPA. The CPA was ordered to implement the De-Ba’athification of Iraq by US Government organs with the decision to do so being determined by the US before the invasion even began.Footnote 67 At least in the instances described above, there is a strong argument that the US was in effective control of the CPA.
Notwithstanding the two specific examples described above, it is not entirely clear how far and to what extent the US generally exercised effective control over the CPA. As put by James Dobbins, Bremer and the CPA’s instructions from Washington were ‘quite general and for the most part, oral … he received plentiful advice but little further direction’.Footnote 68 In addition, as we have seen earlier, the CPA exercised a degree of autonomy and independence from Washington when implementing certain policies and taking certain action.Footnote 69 This type of evidence lends weight to the argument that the US did not direct and enforce the conduct of the CPA and its control over the CPA was only overall in nature.
If the conduct in question was, by way of example, the De-Ba’athification of Iraq, then there is a very strong argument that the US was in effective control of the CPA. However, the same conclusion may not be reached in respect of each and every act of the CPA.
2.2 The law of occupation
Although, the CPA may not necessarily be considered an organ of the US pursuant to the laws of state responsibility, it is conceivable that the law of occupation could provide an alternative outcome.
The law of occupation is ‘anchored in international law generally, but more specifically in the lex specialis of international humanitarian law’.Footnote 70 Lex specialis is the maxim that law that is specific or special in nature can deviate from law that is general in nature.Footnote 71 Theoretically, by applying the principle of lex specialis, the law of occupation could ignore the law of state responsibility and provide a different outcome regarding the legal status of the CPA.
Article 29 of the Geneva Convention is entitled ‘Responsibilities’. It states that ‘[t]he Party to the conflict in whose hands protected persons may be, is responsible for the treatment accorded to them by its agents, irrespective of any individual responsibility which may be incurred’.Footnote 72 The commentaries to the Convention go on to add that the occupying state is charged with the responsibility of ensuring its agents comply with the applicable law.Footnote 73 The Convention clearly envisaged that an occupying state might utilize the services of agents to assist with the administration of the occupied territory.
The commentaries to the Conventions emphasize a broad interpretation of the word ‘agent’ in Article 29.Footnote 74 Agents are described as ‘those persons alone who owe allegiance to the Power concerned’.Footnote 75 The commentaries state that the term agent must be defined to include all persons in the service of the state, including ‘civil servants, judges, members of the armed forces, members of paramilitary police organizations, etc.’.Footnote 76 The commentaries dismiss any suggestion that state procedure or nomenclature should determine the state/agent relationship.Footnote 77 According to the commentaries, what matters is ‘where the decision leading to the unlawful act was made, where the intention was formed and the order given. If the unlawful act was committed at the instigation of the Occupying Power, then the Occupying Power is responsible’.Footnote 78
As already discussed in this article, not every decision of the CPA emanated from Washington and further, the CPA went directly against Washington’s orders in some instances. Put simply, it may be impossible to establish that in each and every instance, the CPA was an agent of the US pursuant to Article 29.
It is this independence and outright defiance of orders that distinguishes the CPA from other similar entities that an occupying power may establish to administer an occupied territory. The Convention clearly envisaged the use of agents in the administration of territories, however, it did not envisage that the agents would be independent of the occupying state. Accordingly, it cannot be concluded with utmost certainty that Article 29 could be applied to the relationship between the US and the CPA with respect to each and every action.
2.3 Resolution 1483 and the CPA
For some commentators, the legal status of the CPA (and therefore, the attribution of its conduct) is relatively straightforward. Resolution 1483 defines the CPA by reference to the ‘specific authorities, responsibilities, and obligations under applicable international law of these states as occupying powers under the unified command (“the Authority”)’.Footnote 79 It is argued that this statement is evidence of the fact that the CPA is merely an expression of the powers of occupation of the US and the UK. In other words, the CPA was a vehicle for occupation by the US.Footnote 80
The argument follows that the legal status of the relationship between the CPA and the US (and the UK) has been endorsed by the Security Council and consequently, is law.Footnote 81 The CPA would therefore, be considered part of the US state.
In contrast, other commentators such as Robert Kolb, argue that the Security Council did not expand or alter the law of occupation pursuant to Resolution 1483.Footnote 82 In Kolb’s opinion, the fact that Resolution 1483 specifically labels some states as occupying powers (that is, the US and UK) and other states as non-occupying powers is not a reconstruction of the law of occupation.Footnote 83 Kolb points to the fact that the statements made by the Security Council in respect of the occupying powers are perambulatory only.Footnote 84 This same argument could be applied to the Resolution’s characterization of the relationship between the US and the CPA. Such provisions in Resolution 1483 are perambulatory and not operative, thereby arguably giving less weight to their determinative or ‘legislative’ character.
2.4 Conclusion
The uncertainty surrounding the establishment of the CPA and from where its authority originated means that it is unclear whether, under US domestic law, the CPA was an organ of the US.Footnote 85 In addition, the requisite level of dependence required to establish de facto organ status may not have existed between the CPA and the US. Further, the degree of independence exercised by the CPA suggests that it would not always be found to be a proxy of the US pursuant to draft Article 8 of the ILC Articles, or an ‘agent’ of the US pursuant to Article 29 of the Geneva Convention. A novel reading of Resolution 1483 may overcome this uncertainty but this too is doubtful because the relevant statements are perambulatory only. Where does this leave the CPA? If the CPA is not an organ of the US (and by implication, a state actor) then, arguably, it has no obligations under the law of occupation. Although the US does have responsibility under the law of occupation, it would arguably not be responsible for the actions of the CPA in each and every instance (either under the law of state responsibility or the law of occupation). This appears to be an entirely unsatisfactory outcome in light of the underlying principles of international humanitarian law. This article submits that the theory of occupation by proxy could provide a different outcome.
3. The theory of occupation by proxy
Occupation by proxy is the theory that a state may occupy another state’s territory through another entity.Footnote 86 There are examples of this type of occupation throughout history. It is hypothesized that Nazi Germany occupied several European states through a number of local puppet administrations.Footnote 87 By way of example, after invading Denmark, Nazi Germany did not remove the local administration.Footnote 88 Instead it exerted control over the territory by controlling the local administration.Footnote 89 Arguably, the local puppet administration was a proxy of Nazi Germany.
3.1 The elements of the legal framework
This article submits that the theory of occupation by proxy consists of three distinct elements: (i) the conflict must be an international armed conflict; (ii) the proxy must occupy territory; and (iii) there must be a proxy relationship between the state and the occupying entity. This article will seek to establish a legal framework for each limb of the theory of occupation by proxy.
3.2 The nature of the conflict
The law of occupation as it is conventionally applied to states is found in four main sources of international law, namely, customary law, the Hague Regulations 1907, the Geneva Convention IV 1949, and Additional Protocol I to the Geneva Conventions.Footnote 90 The Hague Regulations, Geneva Conventions and Additional Protocol I apply only to international armed conflicts.Footnote 91 Accordingly, if a conflict is considered a non-international armed conflict (NIAC), the law of occupation as expressed in these sources will not apply. At face value, a non-state actor controlling territory of a state (which is that actor’s home state) would be a NIAC. However, the relationship between a controlling state and a proxy may render a conflict an international armed conflict (IAC) (as opposed to a NIAC).Footnote 92
The Appeals Chamber of the International Criminal Tribunal for the Former Yugoslavia (ICTY) addressed the internationalization of an armed conflict in the Tadić case. In this case, the Appeals Chamber held that the conflict between Bosnia Herzegovina and the army of the Republika Srpska, who were nationally Bosnian, was an international armed conflict, as a result of the overall control that the FRY exercised over the army of the Republika Srpska.Footnote 93 In other words, the relationship between the army of the Republika Srpska and the FRY internationalized what would otherwise appear to be a NIAC.
Where an occupation has been ‘internationalized’, at a minimum, customary international law, the Geneva Conventions, Additional Protocol I, and the Hague Regulations would apply to any occupation arising from the conflict.Footnote 94
3.3 When is a territory occupied?
In the context of the theory of occupation by proxy, proxies are non-state actors. Accordingly, it must first be established that non-state actors can occupy territory.
3.3.1 Can a non-state actor occupy territory?
Adam Roberts points to several historical events where organized non-state actors have occupied a state to support his contention that a non-state armed actor can occupy a territory.Footnote 95 Aside from historical illustrations, it is contended that Article 1 of Additional Protocol II of the Geneva Conventions provides confirmation that a non-state armed actor can control and by extension, occupy a territory.Footnote 96 Additional Protocol II states:
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 and to implement this Protocol.Footnote 97
Although Article 1 does not specifically state that a non-state armed actor can occupy a territory, it does acknowledge that non-state armed actors are able to render a degree of control over a territory.Footnote 98 For the purposes of Additional Protocol II, a non-state actor must be in sufficient control of a territory to allow it to carry out sustained military operations and to implement the Protocols.Footnote 99 The commentaries to Article 1 of Additional Protocol II link the ability of a non-state armed actor to control a territory with its ability to implement and apply the Protocols.Footnote 100
The position that a non-state armed actor can occupy territory is strengthened by recent judicial decisions. For example, in the Blaškić case, the ICTY Trial Chamber held that the Croatian Defence Council occupied territory.Footnote 101
It appears to be a relatively settled proposition under international law that a non-state armed actor may occupy territory. However, the same cannot be said for non-state non-armed actors (such as the CPA). Nevertheless, this article submits that there are analogous circumstances under international law that can lend weight to the argument that a non-state non-armed actor can occupy territory.
Since its inception, the UN has administered a number of post-conflict territories. There is a significant body of literature devoted to the question of whether the UN would be considered to be occupying such territories and if the law of occupation would apply to the UN in this capacity.Footnote 102 A definitive answer to this particular question is beyond the reach of this article. However, there is a strong argument that in instances where the UN is administering a territory without the consent of the sovereign then it could be considered to be occupying the territory.Footnote 103 Most interestingly for the purpose of this article, the UN has administered territory via a non-state non-armed actor.
Following the conflict in the FRY, the UN Interim Administration in Kosovo (UNMIK) was established. UNMIK was empowered by Security Council Resolution 1244 to administer the territory.Footnote 104 Responsibility for the security of Kosovo, however, was in the hands of NATO military authorities operating under the title Kosovo Force (KFOR).Footnote 105 In a similar fashion to the situation in Iraq, UNMIK undertook administrative functions on the one hand, while KFOR maintained ‘military control’ of territory. Although it is by no means settled that UNMIK was subject to the law of occupation,Footnote 106 this article asserts that this doubt does not stem from the fact that it was a non-armed actor or because it was reliant on KFOR for the maintenance of security in Kosovo but rather, from the fact that it operated under the auspices of the UN.Footnote 107 Accordingly, it may be argued that an entity that is of administrative nature (rather than armed in nature) can control territory (even if it relies on an another armed actor to assist with the exercise and maintenance of this control). By way of analogy, the CPA was an administrative body in control of Iraq in conjunction with the presence of US military forces.
3.3.2 When is a territory occupied?
If a non-state non-armed actor can occupy a territory, what legal test determines when the actor is occupying territory? Adam Roberts states that ‘one may hazard the opinion that the law on occupations ought to be applied fully in the event of an occupation by a non-state entity’.Footnote 108 Accordingly, the same test for determining when territory is occupied in the context of a state occupying another state would apply.
Generally speaking, the entry into and passing through by a foreign military force of a territory will not amount to occupation.Footnote 109 Whether a state is occupying another state is a question of fact.Footnote 110 The concept of ‘effective control’ has developed as the test for determining when a state is occupying a territory (not to be confused with the effective control test pursuant to the Nicaragua case).Footnote 111 Central to the notion of effective control is Article 42 of the Hague Regulations.Footnote 112 The Hague Regulations have been declared customary international law by the ICJ.Footnote 113 In response, various international courts and tribunals have continued to rely on Article 42 of the Hague Regulations to determine when a state of occupation does exist.Footnote 114
Article 42 of the Hague Regulations states that a ‘[t]erritory is considered occupied when it is actually placed under the authority of the hostile army. The occupation extends only to the territory where such authority has been established and can be exercised’.Footnote 115 By combining various jurisprudence and academic literature and on the basis of Article 42 of the Hague Regulations, Tristan Ferraro has put forward a theory of ‘effective control’ containing three distinct limbs: (i) the presence of foreign military forces without the consent of the sovereign of the territory; (ii) the ability of the foreign military forces to substitute its authority for that of the sovereign of the territory; and (iii) the inability of the sovereign of the territory to exercise any authority over the territory.Footnote 116 Limbs one and three of this test are described by Yoram Dinstein as prerequisites to a state of occupation.Footnote 117
Limb one of this formulation has been tested by the recent developments of modern warfare. It is suggested that a foreign army may be able to occupy a territory ‘virtually’.Footnote 118 For present purposes, this article assumes that occupation requires the physical presence of foreign troops.Footnote 119 The second element of limb one, namely, the lack of consent of the sovereign of the territory may prove particularly pertinent in instances where a state is occupied by a proxy and the proxy is a local puppet administration. It could be argued that the puppet administration is the sovereign of the territory and has consented to the occupation, thereby denying the applicability of the law of occupation. Accordingly, the consent required to displace this limb must be genuine and must not be coerced.Footnote 120
Much of the controversy and debate regarding when an occupation exists revolves around the second limb of Ferraro’s test, that is, whether the foreign military force can substitute its authority for that of the sovereign. As put by Benvenisti, this debate centres on the question of ‘was it necessary for the foreign force to actually “exercise its authority”, or was it enough for it to be “in a position” to do so’.Footnote 121 Benvenisti suggests that a territory should be considered occupied when ‘the foreign army is in actual control over enemy territory, and isin a position to establish, if it so wishes, an authority of its own over the population. It is irrelevant whether the army actually does so’.Footnote 122 It is suggested by some commentators, including Benvenisti, that the Armed Activities case established a test for determining occupation that required the occupying state to exercise ‘actual’ control over the territory and its population.Footnote 123 Benvenisti points to the following statement made by the ICJ as evidence of this, ‘the said authority was in fact established and exercised by the intervening state in the areas in question.’Footnote 124 The fact that the ICJ referred to the actual exercise by the intervening state in this instance appears to suggest that the ‘potential’ for the intervening state to exercise its authority is not enough to meet the threshold established under Article 42 of the Hague Regulations. Notwithstanding this judgment, it appears that the majority of scholarly and judicial opinion upholds the ‘potential’ authority test.Footnote 125
Finally, the third limb of Ferraro’s test requires the occupying state to displace the local sovereign. In other words, as stated by Dinstein, ‘there has to be signs of effective control radiating into the environs, in a manner supplanting the authority of the displaced sovereign’.Footnote 126 The importance of this limb was emphasized in the Armed Activities case where the Court stated that it must be found that the Ugandan armed forces in the Democratic Republic of the Congo were not only located in the DRC but had also displaced the authority of the Congolese government with their own authority.Footnote 127
The effective control test continues to be the legal cornerstone of the law of occupation. However, as this article will explore, it has been argued that this test is not the most appropriate test for determining when a non-state actor is occupying territory.
3.3.3 The functional approach
As discussed earlier, it would appear that the law applicable to an occupation by proxy in an IAC is the same law that applies when a state directly occupies another. The logical extension of this is that the test for determining when a state is occupying a territory through a proxy would be the same test for determining when a state is occupying another in the classical sense. Although this may be the most obvious conclusion, applying the effective control test to situations of occupation by a non-state actor proxy invariably encounters some difficulties due to the nature and inherent limitations of a non-state actor.
If strictly interpreted, a non-state actor may not exercise a sufficient degree of control to meet the effective control standard required by Article 42 of the Hague Regulations.Footnote 128 Although a non-state actor may be in a position to exercise authority over territory, depending on the actor in question, it may not have the resources or the know-how to implement this authority. Accordingly, a state of occupation would not be established. With this in mind, Tom Gal argues that a different test needs to be applied to instances of occupation by proxy involving a non-state actor.Footnote 129 One such test is the functional approach.
The origins of the functional approach lie in the commentaries to Geneva Convention IV. The commentaries suggest that the law of occupation should apply once a person ‘falls into the hands of the Occupying Power’.Footnote 130 In other words, the protections under the law of occupation will apply as soon as a person is in the control of a foreign force, notwithstanding that a state of occupation may not generally exist.Footnote 131
This type of approach is referred to in the commentary to Article 6 of Geneva Convention IV which deals with the beginning and end of the application of Geneva Convention IV.Footnote 132 The commentary to Article 6 of Geneva Convention IV advocates that the definition of an occupation under the Geneva Conventions should be given a broader interpretation than that under Article 42 of the Hague Regulations, that is to say, ‘[t]he Convention should be applied as soon as troops are in foreign territory and in contact with the civilian population there’.Footnote 133 Furthermore, the law of occupation should be applied in proportion to the degree that the foreign forces continue to exert any authority over the local population.Footnote 134 It does not require that a state of effective control is achieved for the law of occupation to apply. In a sense, it is the antithesis to the ‘all-or-nothing’ approach that is exemplified by the effective control test.
The functional approach has, to an extent, been adopted by some international judicial bodies. It is arguable that the functional approach imbues the stance taken by the ICTY in the Naletilić case. In the Naletilić case, the Trial Chamber distinguished between the applicability of the law of occupation in relation to the protection of property and the applicability of the law of occupation in respect of the rights of individuals.Footnote 135 The Trial Chamber found that the laws of occupation under Geneva Convention IV applied once an individual ‘fell into the hands of the occupying State’, irrespective of whether a state of occupation, in accordance with Article 42 of the Hague Regulations, existed.Footnote 136 The Trial Chamber also found that the law prohibiting deportation from occupied territory applied to an occupying power as soon as an individual was in the control of the alleged occupying power.Footnote 137 The Appeals Chamber did not reject this approach.Footnote 138
Generally speaking, proponents of the functional approach derive its legitimacy from the fact that it prevents a state from avoiding responsibility under the law of occupation, particularly during the invasion phase where effective control has not yet been established.Footnote 139 It has been argued that it may also be useful in instances of occupation by a non-state actor who may not be in effective control of a territory (within the meaning of Article 42 of the Hague Regulations) but who, nonetheless, exercises a degree of control over a territory and its population.Footnote 140
The strength of the functional approach in respect of non-state actors is that it takes into account the limited resources and potential deficiencies of a non-state actor.Footnote 141 Non-state actors are unlikely to have the same resources (whether financial or otherwise) or governance expertise to occupy a territory in the manner envisaged by Article 42 of the Hague Regulations. In order to regulate the conduct of a non-state actor towards the population of a controlled territory, a functional approach would, instead, apply the law of occupation to the extent that the non-state actor exercises control over the territory and the population.Footnote 142 Although the functional approach may resolve many of the issues associated with the general deficiencies of a non-state actor vis-à-vis a state, this article contends that it alone does not adequately address the application of the law of occupation in relation to a non-state actor proxy.
As we have seen, the functional approach requires the law of occupation to be applied proportionally to the extent that a non-state actor exercises authority over the population. In situations where a non-state actor is acting as a proxy of a state, this article contends that an additional layer of analysis is required in order to determine the extent to which the law of occupation should apply.Footnote 143
It is submitted that the degree of control which a state exercises over the proxy should determine the degree to which the non-state actor can implement the law of occupation.Footnote 144 The greater the control a state exercises over its proxy, then the greater the resources the state can make available to the proxy to assist it with implementing the law of occupation. The logic is that if a state is ordering and directing the conduct of a proxy, then there is no reason why such orders and directions cannot be made in accordance with the law of occupation in its fullest application. This type of approach will arguably encourage states to take a more active role in the function and operations of their proxies, rather than merely using a proxy as a convenient tool for avoiding responsibility under the law of occupation.
This article acknowledges that a state may seek to avoid responsibility by minimizing its degree of control over a proxy in order to limit the application of the law of occupation. However, if the control is minimized, the conflict would be a NIAC and the law of occupation would not apply (i.e., the state would exercise less than overall control over the proxy). As the next part of this article will explore, in addition, in order for a state and proxy relationship to function coherently, there needs to be a level of control exerted by the state over the proxy.
Notwithstanding the benefits outlined above of a functional approach, as a matter of current law, it is not the test for determining when a state of occupation exists.Footnote 145 Accordingly, this article submits that in order to determine whether the CPA occupied Iraq as a proxy of the US, the appropriate approach is to determine whether the CPA was in effective control of the territory.
3.4 Is there a proxy relationship?
The final limb of the occupation by proxy framework submitted by this article is the existence of a proxy relationship. There is no legal definition of proxy under international law. However, as discussed in part two of this article, the term ‘proxy’ is commonly used in reference to the law of state responsibility, specifically, draft Article 8 of the ILC Articles which states that:
the conduct of a person or group of persons shall be considered an act of a State under international law if the person or group of persons is in fact acting on the instructions of, or under the direction or control of, that State in carrying out the conduct.Footnote 146
The commentary to the ILC Articles notes that Article 8 contains two distinct types of scenarios where the conduct of a person or groups of persons will be attributed to a state. First, the conduct of a person or group of persons will be considered an act of a state if the person or group of persons is in fact acting under the instruction of a state.Footnote 147 Second, the conduct of a person or persons will be considered an act of a state if the person or groups of persons are acting under the direction or control of a state in carrying out the conduct.Footnote 148 As Section 2 of this article has discussed, this test was most famously interpreted and applied in the Nicaragua case. However, in the context of the theory of occupation by proxy, an alternative test has been proffered.
3.4.1 Overall control and the Tadić case
In the Tadić case, the Trial Chamber considered the relationship between the army of the Republika Srpska, being nationals of the Republic of Bosnia and Herzegovina and the FRY. The Trial Chamber explicitly recognized the potential for a foreign state to occupy territory via de facto organs or agents of the state.Footnote 149 In determining the relationship between the army of the Republika Srpska and the FRY, the Trial Chamber applied the ‘effective control’ test in accordance with the Nicaragua case.Footnote 150 The application of this test was rejected by the Appeals Chamber of the ICTY.
The Appeals Chamber acknowledged the ‘effective control’ test enunciated in the Nicaragua case,Footnote 151 however, it found that the effective control test was applicable to ‘unorganised groups of individuals acting on behalf of States’.Footnote 152 When considering the degree of control over military and paramilitary forces, the Appeals Chamber found that a different test needed to be applied.Footnote 153 The Appeals Chamber stated that ‘it must be proved that the state wields overall control over the group, not only by equipping and financing the group, but also by co-ordinating or helping in the general planning of its military activity’.Footnote 154 Accordingly, the Appeals Chamber held that the army of the Republika Srpska was a proxy of the FRY.Footnote 155
The overall control test put forward in the Tadić case was affirmed by the Trial Chamber of the ICTY in the Blaškić case.Footnote 156 In this case, the Tribunal held that Croatia was an occupying power through its overall control of the Croatian Defence Council.Footnote 157 However, cracks in the ICTY’s jurisprudence began to form in the Naletilić case.
3.4.2 Naletilić – against the grain
In the Naletilić case, the ICTY Trial Chamber rejected the legal reasoning of the Blaškić case and stated that:
there is an essential distinction between the determination of a state of occupation and that of the existence of an international armed conflict. The application of the overall control test is applicable to the latter. A further degree of control is required to establish occupation.Footnote 158
The Trial Chamber then went on to describe the test of effective control as set out in the Hague Regulations. Some commentators have argued that the Trial Chamber erred in its rejection of the Blaškić case jurisprudence because the Trial Chamber had confused the two tests that make up the theory of occupation by proxy, namely, the test to determine if an actor was a state’s proxy and the test to determine whether an actor (or a state) was occupying territory.Footnote 159 Given that the Trial Chamber went on to describe and apply the Hague Regulations after rejecting the Blaškić case, this article submits that there is merit to this argument. Notwithstanding that there may be grounds to suggest that the Naletilić decision is an outlier, recent judgments of the ICJ have contributed to the uncertainty about the most appropriate test for establishing a proxy relationship in the context of occupation by proxy.
3.4.3 The International Court of Justice – turning its back on Tadić
The ICJ was required to consider the applicable degree of control required to impute a non-state actor’s wrongful act to a state in the Armed Activities case,Footnote 160 and again in the Bosnian Genocide case.Footnote 161 In the Armed Activities case, the Court considered whether the activities of Congolese paramilitaries could be attributed to Uganda in relation to the Court’s findings of the law of the prohibition against the use of force.Footnote 162 In reaching its conclusion that the conduct of the Congolese rebels could not be attributed to Uganda, the Court applied Draft Article 8 of the ILC Articles by having regard to the effective control test of the Nicaragua case.Footnote 163 Interestingly, when considering the law of belligerent occupation, the Court acknowledged the Democratic Republic of Congo’s reference in its submissions to ‘“indirect administration” through various Congolese rebel factions and to the supervision by Ugandan officers over local elections in the territories under the [Ugandan People’s Defence Force’s] Control’.Footnote 164 Some commentators point to this acknowledgement as evidence of the Court’s support for a theory of occupation by proxy akin to that described in the Tadić case. However, this article argues that this may be overstating the case. Despite acknowledging the concept of indirect administration, the Court goes on to refer to its earlier finding that there was no evidence that the rebel groups were under the effective control of Uganda.Footnote 165
In the Bosnian Genocide case (decided after the Armed Activities case), the Court considered the overall control test as expressed in the Tadić case, but dismissed it outright.Footnote 166 The Court’s reasoning for doing so was multifaceted. First, the Court queried the jurisdiction of the ICTY to make such a determination in respect of law of state responsibility (which is not a criminal law issue).Footnote 167 Second, the Court highlighted that the overall control test was applied by the ICTY in order to characterize an armed conflict as an IAC or NIAC (as opposed to attributing acts of genocide).Footnote 168 Finally, the Court opined that the overall control test:
has the major drawback of broadening the scope of State responsibility well beyond the fundamental principle governing the law of international responsibility: a State is responsible only for its own conduct, that is to say the conduct of persons acting, on whatever basis, on its behalf.Footnote 169
The Court supported the Nicaragua ‘effective control’ test for determining when a state will be responsible for the acts of its proxy.Footnote 170
3.4.4 Which authority prevails?
Eyal Benvenisti submits that the appropriate test for determining whether an occupying proxy’s conduct will be imputed to a state is the effective control test.Footnote 171 Benvenisti particularly supports the findings made by the ICTY in the Naletilić case.Footnote 172 Benvenisti contends that ‘[t]his further degree of control is required because the occupant is responsible not only for its own acts (and by extension for the acts of its proxies), but also for the acts of third parties that operate in the territory’.Footnote 173
It would appear that the commentaries to the Geneva Conventions also support this position. As already noted in this article, Article 29 asserts that ‘what is important is to know where the decision leading to the unlawful act was made, where the intention was formed and the order given’.Footnote 174 This seems to imply that the conduct of an agent pursuant to Article 29 is to be considered on a case-by-case basis, rather than by reference to any general overall control that the state may exercise over the agent.
By way of contrast, other commentators have supported the theory of ‘indirect effective control’. The concept of indirect effective control deems a state to be an occupying power pursuant to the law of occupation when a proxy, over whom it has overall control, occupies another state’s territory.Footnote 175 In the view of such commentators, the overall control exercised by a state over a non-state actor ‘reflects a real and effective link between the group of persons exercising the effective control and the foreign state operating through those surrogates’.Footnote 176 This test of overall control is justified on the basis that it closes any ‘gaps’ in the law of occupation.Footnote 177
3.4.5 Resolving the conflict
This article submits that, as it stands, the theory of occupation by proxy is not a zero sum game between the ICTY authority and ICJ authority. The question that actually needs to be answered is whether the law of state responsibility should be the test for determining whether a non-state actor is a proxy of a state for the purposes of the theory of occupation by proxy. Although the term ‘proxy’ is most often associated with the law of state responsibility, it does not automatically follow that the law of occupation needs to apply the same formula for determining proxy status.Footnote 178 The theory of occupation by proxy can develop its own test for proxy status, separately from the law of state responsibility.
The ICJ dismissed the overall control test in the Bosnian Genocide case as it considered it inappropriate for the law of state responsibility. This article does not dispute this. Crucially, in the Bosnian Genocide case the ICJ did not take a position as to whether the overall control test is appropriate for determining whether a conflict is a NIAC or IAC.Footnote 179 Nor did it apply or consider the framework for occupation by proxy.
On this basis, this article contends that the overall control test (as expressed in Tadić) is the appropriate test for determining whether a conflict is an IAC or NIAC and also, whether a non-state actor is the proxy of a state pursuant to the theory of occupation by proxy.Footnote 180
3.4.6 Can the overall control test apply to non-state non-armed actors?
Notwithstanding the findings above, it could be argued that the Tadić test of overall control is not applicable to non-armed non-state actors. The overall control test articulated by the Appeals Chamber in Tadić refers only to non-state armed groups and not to other types of non-state actors such as the CPA. However, it would appear that the Chamber’s rationale for distinguishing between armed groups and other actors (such as individuals) is two-fold: armed actors are groups of persons; and armed actors are organized.
It does not appear to turn on whether the group in question is armed or not armed. The Appeals Chamber stated that it is necessary to distinguish between:
individuals acting on behalf of a State without specific instructions, from that of individuals making up an organised and hierarchically structured group, such as a military unit or, in case of war or civil strife, armed bands of irregulars or rebels.Footnote 181
The Chamber only refers to military units or rebels as examples of an organized group. Although not an armed actor, there is strong evidence to suggest that the CPA was highly organized. For example, there was a clear chain of command within the CPA and its personnel were subject to a set of procedures and policies.Footnote 182 Accordingly, this article submits that the theory of occupation by proxy put forward in the Tadić case is not limited to armed groups only and could theoretically apply to non-state non-armed groups.
4. Applying the legal framework of occupation by proxy to the CPA
The first limb of legal framework for the theory of occupation by proxy established in part three requires that there is an international armed conflict. There is no doubt that the US invasion of Iraq was an international armed conflict. Accordingly, this limb of the legal framework will not be examined in depth as part of this analysis. Instead, this analysis will focus on limbs two and three of the legal framework.
4.1 Did the CPA occupy Iraq?
To determine whether the CPA exercised effective control over Iraq, this article will apply the test for occupation enunciated by Tristan Ferraro (as described earlier in Section 3.3.2).Footnote 183 As a part of this analysis, this article will accept that limb three of the effective control test, that is, the inability of the sovereign to exercise any authority over the territory was clearly the state of affairs in Iraq.
Limb two of the Ferraro test requires the foreign military forces to substitute its authority for that of the sovereign.Footnote 184 The CPA clearly exerted authority over Iraq and its people. Regulation 1 of the CPA states that the CPA ‘shall exercise powers of government temporarily in order to provide for the effective administration of Iraq during the period of transitional administration’.Footnote 185 The orders made by the CPA during its reign clearly demonstrate its governmental powers and function. For example, Order 7 of the CPA amended the Iraqi penal code.Footnote 186 This type of power during an occupation is specifically provided for in Article 64 of Geneva Convention IV which regulates any amendments to the penal code of an occupied territory by an occupying power.Footnote 187 Pursuant to Order 1, the CPA removed a number of public officials from office as a part of the De-Ba’athification program undertaken by the CPA.Footnote 188 Again, this type of behaviour is regulated by Article 51 of Geneva Convention IV.Footnote 189 Accordingly, it is clear that the CPA displaced the authority of the sovereign, therefore satisfying limb 2 of the Ferraro test. However, it does not automatically follow that limb 1 of Ferraro’s test has been met.
Limb 1 of Ferraro’s test requires the presence of foreign military forces without the consent of the sovereign of the territory. There appears to be little argument over the fact that the CPA and foreign troops were present in Iraq without the consent of the sovereign. What is significant, however, in the case of the CPA is that these foreign troops were not ‘CPA’ troops. The CPA was not an armed actor and was instead reliant on state military forces (namely, the US and to a lesser extent, the UK) to secure and control Iraq. To add to the complexity, Bremer and the CPA were not in direct command of the vast majority of US military personnel in Iraq.Footnote 190
Although there are difficulties associated with the fact that the CPA was reliant on US military forces (over which it had no control) in order to maintain its military control over Iraq, this article contends that this does not prevent the CPA from being found to be in effective control of Iraq. Even though the CPA did not directly control these forces, it was able to leverage its relationship with the US in a manner which allowed it to effectively govern Iraq during its reign. In this regard, the cases of UN territorial administration (as described earlier in this article) act as a useful prism through which to view this potential roadblock. In conclusion, when viewed holistically and contextually, this article submits that the CPA occupied Iraq.
4.2 Was the CPA a proxy of the US in the context of the theory of occupation by proxy?
As noted above, this article argues that the ‘overall control’ test is the most appropriate test for determining proxy status pursuant to the theory of occupation by proxy. In the Tadić case, the Appeals Chamber noted that specific instructions for particular acts are not required to establish that a state had overall control of a proxy.Footnote 191 In determining the nature of the relationship between the FRY and the army of the Republika Srpska, the Appeals Chamber highlighted the ‘general coordination, direction and supervision’ of the activities of the army of the Republika Srpska,Footnote 192 together with the financial, logistic and other support utilized by the army of the Republika Srpska from the FRY.Footnote 193
As already noted, the CPA was heavily reliant on Iraqi money for its reconstruction efforts,Footnote 194 nevertheless, the CPA still received funding from the US.Footnote 195 In some circumstances, the CPA actively ignored orders from Washington.Footnote 196 Nevertheless, Washington’s fingerprints are all over several of the key policies implemented by the CPA. For example, the De-Ba’athification of Iraq originated in Washington before the invasion even began.Footnote 197 Memos between the head of the CPA, Paul Bremer, and the Secretary of Defense, Donald Rumsfeld regarding proposed policies are evidence of the general co-ordination between the US and CPA regarding the governance of Iraq.Footnote 198 Further, the CPA’s ability to physically operate within Iraq was wholly dependent on the presence of US forces (over which it was not in command).Footnote 199 Accordingly, it is reasonable to conclude that the US was in overall control of the CPA.
5. Closing the gap
The application of the law of occupation has been continually avoided by states.Footnote 200 Since the ratification of the Geneva Conventions, states have gone to great lengths to dismiss any claims that they are occupying a territory and, therefore, are bound by the law of occupation.Footnote 201 Such has been the state of affairs in the post-Second World War period, that the case of Iraq is said to have awoken the law of occupation from its slumber.Footnote 202 However, the use of non-state non-armed actors (such as the CPA) as a tool for administration of an occupation arguably could provide another means for a state to avoid responsibility under the law of occupation.
As with non-state armed actors, the use of a vehicle such as the CPA could mean that gaps form within the chain of responsibility. Where the degree of control over a proxy required for occupation by proxy is different to the degree of control required to impute responsibility, there is a ‘gap’ in the law.Footnote 203 That is, although the law of occupation would apply to the state and its proxy, the law of state responsibility would not necessarily attribute responsibility for the proxy’s conduct to the state. As identified by Gal, in such situations no one would be liable for a breach of the law of occupation.Footnote 204 As has already been seen with private military companies,Footnote 205 this gap in the law may encourage states to outsource the administration of territories to the private sector in order to limit a state’s responsibility under international law.
The most obvious way to close this gap would be to align the tests for proxy status – effective control would be the test for occupation by proxy as it is for ILC Article 8 under the law of state responsibility. However, this would mean that the theory of occupation by proxy would apply in fewer situations and therefore, the protection of civilian populations would be weakened. As an alternate solution, Gal suggests that the gap could be closed by applying the historic doctrine of belligerency to non-state actors which would ‘internationalize’ the conflict and confer certain obligations on the non-state actor.Footnote 206 Although this proposition is somewhat persuasive, there could be a simpler solution available under international law.
In discussing the applicability of Article 29 of the Geneva Convention, this article drew on the maxim lex specialis – that is, law that is specific or special in nature – can deviate from law that is general in nature.Footnote 207 If the theory of occupation by proxy was considered part of the law of occupation, it could theoretically deviate from the law of state responsibility and ascribe responsibility to a state for its proxy in certain circumstances. Article 29 is evidence of the fact that the drafters of the Conventions had envisaged a specific regime of responsibility for states under the law of occupation.
There are two conditions precedent to this potential solution: (i) the theory of occupation by proxy must be considered part of the law of occupation so that the maxim of lex specialis can apply; and (ii) the theory of occupation must evolve to expressly address the responsibility of a state for its proxy.
In respect of the first condition, it is conceivable that, in future, the theory of occupation by proxy could become customary international law and by extension, part of the law of occupation.Footnote 208 In respect of the second condition, in the Tadić case the Appeals Chamber did discuss the theory of occupation by proxy with specific regard to responsibility and attribution, not just to the internationalization of an armed conflict.Footnote 209 However, as already discussed, in the Bosnian Genocide case the ICJ held that the overall control test was ‘unsuitable, for it stretches too far, almost to breaking point, the connection which must exist between the conduct of a State’s organs and its international responsibility’.Footnote 210 The forcefulness of the ICJ’s dismissal of the overall control test in the context of state responsibility has arguably weakened the credibility of the overall control test. However, it is important to note that the Bosnian Genocide case was not determining proxy status and state responsibility in the context of occupation by proxy. Accordingly, it could be argued that it is open to another international court or tribunal to apply the Appeals Chamber’s reasoning in Tadić in respect of responsibility where there is occupied territory.
For the moment, this argument remains purely hypothetical. Therefore, it is conceivable that this ‘gap’ still remains and the US would not be responsible for any breach of the law of occupation by the CPA.
6. Conclusion
This article began by proposing two questions: was the CPA, as a result of its relationship with the US, bound by the law of occupation; and second, now that the CPA has dissolved, is the US responsible for the wrongful acts of the CPA? The uncertainty about how the CPA was created, coupled with its instances of independence from, and defiance of the US administration, means that the answer to these questions is hazy. This article has suggested that the theory of occupation could be applied to the CPA/US relationship and further, could provide clarity about the CPA’s obligations under the law of occupation and potentially, US responsibility for the CPA’s actions. However, the theory of occupation has been dealt a blow by more recent judgments.
Although clearly expressed in the Tadić case, subsequent cases (particularly the Bosnian Genocide case) have created uncertainty about the thresholds and parameters of the theory. As a result, the use of the CPA to administer occupied Iraq may have allowed the US to quarantine part of its risk to an entity that has long disbanded. The implications of this for the law of occupation are profound. Where a state’s risk pursuant to the law of occupation is minimized, its incentive to comply with the legal regime is reduced. Accordingly, a CPA style territorial administration may be the new frontier for ‘occupation’ of territory by a state.
