Hostname: page-component-745bb68f8f-g4j75 Total loading time: 0 Render date: 2025-02-11T00:53:11.413Z Has data issue: false hasContentIssue false
  • English
  • Français

Between Consented and Un-Contested Occupation

Published online by Cambridge University Press:  24 October 2018

Kyo Arai
Kyo Arai*
Affiliation:
Professor of International Law, Faculty of Law, Doshisha University, Kyoto (Japan). karai@mail.doshisha.ac.jp.

Abstract

It has long been recognised that ‘non-consent’ is a fundamental element of the law of occupation. Under modern international humanitarian law (IHL), the consensual presence of foreign military forces is generally not seen as belligerent occupation.

However, if we accept the principle that the application of IHL should rely on the objective situation on the ground and not on the subjective judgment of the situation of parties to the conflict, it may be natural to diminish the significance of consent by the territorial states in relation to the application of the law of occupation. It may be somewhat harmful to deny such protection based solely on the existence of the territorial states’ consent without considering the relationship, in reality, between the occupier and the population in the occupied area. According to a teleological interpretation of IHL, especially when it is obvious that the latter has no allegiance to the former, the tense relationship between them should be regulated by the law of occupation.

This article discusses whether and how state consent could be a humanitarian ground to negate the legal protection for its own people, and highlights situations where the local population needs protection by the law of occupation (or comparable rules) in consensual military occupations.

Keywords

law of occupationconsentpacific occupationvalidity of consenthuman rights
Type
Articles
Information
Israel Law Review , Volume 51 , Issue 3 , November 2018 , pp. 365 - 388
Copyright
Copyright © Cambridge University Press and The Faculty of Law, The Hebrew University of Jerusalem 2018 

1. Introduction

From its original conception in the 1874 Brussels ProjectFootnote 1 and the Hague Regulations of 1899/1907,Footnote 2 the fundamental tenets of the law of occupation, while being supplemented by the Fourth Geneva Convention of 1949,Footnote 3 have remained unchanged to this day. In essence, the law of occupation ‘[authorises] the occupant to safeguard its interests while administering the occupied area, but also [imposes] obligations on the occupant to protect the life and property of the inhabitants and to respect the sovereign interests of the ousted government’.Footnote 4

However, the developments in international law during this time (such as the prohibition of war), the development of international human rights law, and especially the humanisation of the law of armed conflict,Footnote 5 have affected, in addition to specific rules within the law, the raison d’être of the law of occupation. The issue, then, is whether and, if so, how have such developments affected the non-consensual nature of belligerent occupation,Footnote 6 understood as the ‘absence of consent from the [territorial] State … as a precondition for the existence of a state of belligerent occupation’.Footnote 7

Does ‘occupation’ that takes place with the consent of the territorial state never constitute belligerent occupation and is therefore not subject to the law of occupation? It is not difficult to think of some situations which could cast doubt on this presupposition. First, there may be situations where a foreign army exercises power that is similar in nature to that of an occupying power notwithstanding the ‘consent’ of a government. The situation in Iraq after 2004 may be an appropriate example. Belligerent occupation of Iraq by the United States (US) and the United Kingdom (UK) was formally announced to have ceased on 28 June 2004, and the International Committee of the Red Cross (ICRC) and other institutions seemed to agree with this.Footnote 8 From that date the presence of those forces was seen as the result of ‘consensual’ deployment based on invitation by the Iraqi interim government. Surely, from a political perspective, this announcement of the end of occupation was necessary at that precise moment in order not to jeopardise the legitimacy of the newly established local government. However, the former occupying powers effectively retained a certain level of authority over the territory of Iraq during the following years, especially with regard to the maintenance of law and order. Moreover, since the number of detained Iraqi citizens later skyrocketed,Footnote 9 as did the number of deployed soldiers, the former occupying powers had more decisive influence on the Iraqi people than was the case during the one-year occupation until June 2004. Thus, it may be fair to say that when looked at from the perspective of the raison d’être of the laws of occupation, the UK and the US forces should have been regarded as an occupation force after 2004.Footnote 10

Another instance of the consensual presence of a foreign military comparable with belligerent occupation is found where the territorial state recognises the exclusive administration of part or all of its own territory by a foreign military. Such a case is generally differentiated from belligerent occupation, and is understood as ‘land-lease’Footnote 11 or ‘conventional occupation’.Footnote 12 However, such differentiation can be formalistic in certain cases. An illustrative example is the US presence in Okinawa (Japan) between 1952 and 1972.Footnote 13 Unlike the post-2004 Iraqi situation, in Okinawa during this period the exercise of any authority and control by the territorial state, Japan, was completely replaced by that of the US. In 1951 Japan signed a peace treaty with the Allied Powers.Footnote 14 While Article 1 of this treaty made clear that the state of war should be terminated,Footnote 15 Japan agreed (in Article 3)Footnote 16 that the US would have ‘the right to exercise all and any powers of administration, legislation, and jurisdiction over the territory and inhabitants’ of the Nansei Shoto, including Okinawa, which the US Army and Navy had controlled as an occupied territory since spring of 1945.Footnote 17 While Japan's ‘residual’ sovereignty over the Islands was proclaimed,Footnote 18 Okinawa was put under the effective and exclusive control of the US indefinitely,Footnote 19 with a self-governing body of the local people, which was strictly commanded by the US Civil Administration in Ryukyu Islands (USCAR), as set out in its Proclamation No 13.Footnote 20 A tense relationship between the US military and the inhabitants was not altered by the conclusion of the Peace Treaty; therefore the US occupation virtually continued from 1945 to 1972, although the last two decades were technically consensual.Footnote 21 The legal landscape worsened after 1952 as the application of the international law of occupation ceased and was not replaced by an alternative source of legal protection. Neither international law nor US or Japanese constitutional law provided protection and the inhabitants were left facing the discretionary exercise of power by the US militaryFootnote 22 for more than 20 years.Footnote 23

Above all, it should be recalled that ‘consent’ has the potential to be engineered at all times, as shown by precedents such as the Soviet intervention in Czechoslovakia and Afghanistan or the US intervention in Panama.Footnote 24 In these cases, ‘consent’ or ‘invitation’ to intervene and occupy the country may be given by a phony ‘government’ fabricated by the intervening state or a ‘local government’ that inadequately represents the population, or, at best, in situations where evidential problems inevitably exist. This is a powerful reason to legally evaluate the ‘consent’ of the targeted state in a vigilant manner.

That said, the term ‘sous l'autorité de l'armée ennemie’ (‘under the authority of the hostile army’) under Article 42 of the Hague Regulations suggests, prima facie, that cases where consent exists are excluded from the definition of belligerent occupation.Footnote 25 This is because, as the French text suggests, occupations that are subject to the law of occupation are intended to be conducted by an ‘enemy’ state against which the territorial state is waging ‘war’ as a matter of law.Footnote 26 More recently, the 2012 ICRC Expert Meeting on Occupation and Other Forms of Administration of Foreign Territory reconfirmed this non-consensual essence of occupation, although it also acknowledged that the validity of the consent of the territorial state could be difficult to ascertain in specific cases, taking account of the likelihood that, as discussed above, consent can be engineered.Footnote 27

This formulation, however, is such that the protection of individuals rests on the existence of consent by the territorial state, and this does not conform to the shift under international humanitarian law (IHL) away from securing the interests of states and towards emphasising the humanitarian protection of individuals. It also conflicts with the underlying idea that the applicability of IHL will depend on the prevailing facts on the ground, and not on subjective classifications of the situation in question by a party to the conflict. That said, if the consent of the territorial state still retains a significant impact on the application of the law of belligerent occupation, the question remains as to whether and to what extent that should be maintained.

Before delving into this issue, a preliminary point should be noted. Consent by the territorial state is relevant not only to the application of the law of occupation, but may also be considered in other contexts. Consent can be the basis for precluding wrongfulness under the law of state responsibility,Footnote 28 and a justification ad bellum for use of force (intervention by invitation).Footnote 29 In that sense, the existence of the consent of the victim state is nothing but the central element of the legality of ‘occupation’ of another country under the jus ad bellum. Although analyses from these points overlap and are sometimes confused with determining the application of the law of occupation,Footnote 30 they should be distinguished from the latter. Since the law of occupation can apply to any foreign military control over the territory of a state regardless of its legality ad bellum, wrongfulness ad bellum on the part of the intervening state has no implication for the application of the law of occupation.

2. Objectivisation of the Concept of Occupation

According to traditional international law, a consensual military occupation was not subject to the law of occupation. This was based on a legal assessment relating to the existence or non-existence of a state of war. Distinctions were made between belligerent and pacific occupations, and military operations that were carried out with the prior consent of the territorial state did not qualify as a state of war.Footnote 31 Thus, a consensual occupation that did not generate a state of war was characterised, by definition, as a pacific occupation, rather than a belligerent occupation to which the law of occupation did apply. Pacific occupations, however, were not always carried out in a pacific manner. Even coercive ‘occupations’ that lacked consent were characterised as pacific occupations, inasmuch as they were not accompanied by a formal state of war.

Following the Second World War and in accordance with contemporary international law, Common Article 2 of the 1949 Geneva ConventionsFootnote 32 stipulates that the Conventions apply to ‘all cases of declared war or of any other armed conflict’, thereby expanding the application of IHL to include ‘occupations’ that occur outside a state of war.Footnote 33 Legally this means that the distinction between belligerent and pacific occupation has dissolved. Furthermore, taking account of the occupation of Denmark by Germany during the Second World War, the Geneva Conventions, including the Fourth Convention, regulate ‘all cases of partial or total occupation of the territory of a High Contracting Party’, even if this ‘meets with no armed resistance’ (Common Article 2(2)).Footnote 34 As Kolb and Vité maintain, for the purpose of maximising humanitarian protection, Common Article 2(2) of the Geneva Conventions reversed the mode of evaluating the absence of consent.Footnote 35 According to this logic, it is necessary that the ‘peace’ be openly proclaimed so as to preclude the application of the occupation regime, while any other situation involving the deployment of troops on foreign territory is now presumed to be covered by Common Article 2(2).Footnote 36 Consequently, however, it became necessary to distinguish between occupation, especially when it is uncontested, to which the law of occupation applies, and other situations where foreign armed forces are deployed but the law does not apply. The consent of the territorial state may fill this role to some extent.

What is the rationale for allowing consent by the territorial state to preclude the application of the law of occupation? As noted above, in accordance with traditional international law, such consent formally signalled the absence of a state of war and provided the justification for the non-application of the law of occupation. Additionally, there were (and still are) substantive grounds to exclude the application of the law of occupation when consent is given. First, the application of the law of occupation is unnecessary because the invitation of a foreign army through ‘consent’, rather than the law of occupation, could delimit and legally control the modus operandi of the said army, and give good guidance for the preservation of the status quo ante. Second, if the invited foreign troops assist the inviting government and therefore are intended to act as the latter's auxiliary force, the law of occupation does not purport to protect individuals against their own government.Footnote 37

That being said, relying on the principle that the application of IHL must now be determined by the prevailing facts on the ground,Footnote 38 the question arises of whether the consent of the territorial state is sufficient to indicate a lack of necessity in applying humanitarian protection provided by the law of occupation. In cases where military occupation is met with armed hostilities, it is relatively clear that the occupying force is hostile towards the population in the occupied area. On the other hand, in cases of occupation without armed resistance, consent is the sole criterion for classifying situations as non-belligerent. The question, therefore, is how and to what extent can ‘consent’ be interpreted as a ground to overturn the underlying conception of Common Article 2(2) that protection of the population is assumed to be necessary whenever a foreign military exerts control over a given territory.

3. The Significance of Genuineness and Validity of ‘Consent’

Many experts, including those convened by the ICRC during the 2012 Expert Meeting, agree that, for the law of occupation to be inapplicable, the consent of the territorial state must be ‘genuine, valid and explicit’.Footnote 39 These requirements are deduced from an indisputable rule of international law that an acte juridique which produces a specific conséquence juridique, like ‘consent’ to foreign military intervention, shall be the freely expressed will (volonté libre) of a capable subject.Footnote 40

There is also a substantive rationale. Ensuring the genuineness and validity of consent is required in order for the consent to be understood as a manifestation of the need for humanitarian protection of the population in determining whether the law of occupation applies to a given case. On this point, it can generally be presumed that the legally valid and genuine consent of the territorial state renders operations by the foreign military forces consistent with the interests of the governed population. This is especially so when, because of the democratic foundation of the territorial government, it is regarded as ‘in a better position’Footnote 41 than other states or international institutions to appraise the necessity of applying the law of occupation.

3.1. The Validity of Consent as an International Agreement

First, the (in)validity of ‘consent’ as an international agreement between intervening and occupied states should be considered through the lens of the Vienna Convention on the Law of Treaties.Footnote 42 The grounds for the invalidity of treaties enumerated in the Vienna Convention can certainly be referred to in the discussion of the validity of consent in the context of the application of the law of occupation.Footnote 43

In conformity with the law of treaties, when the consent of the territorial state was ‘procured by the threat or use of force in violation of the principles of international law embodied in the Charter of the United Nations’ (Article 52) or ‘by the coercion of a representative’ of the state (Article 51), it is invalid a priori and, at the same time, a foreign deployment under such consent should be regarded as belligerent occupation.Footnote 44

In that sense, for example, the coerced consent of President Hácha of Czechoslovakia to the Nazi invasion in 1939 was seen as an indisputable precedent of an invalid agreement,Footnote 45 which by implication means that Bohemia and Moravia should be understood to have been occupied territories.

However, a more nuanced evaluation applied to the Anschluss. With regard to the Austrian consent in 1938, the Nuremberg Tribunal denied that it had been given validly and further opined that even if it had, it would have been coerced and did not excuse the annexation. While it was made clear by the tribunal that Austria was the first victim of the Nazi invasion from the perspective of the jus ad bellum,Footnote 46 it remained debatable whether Austria was under belligerent occupation from 1938 to 1945.Footnote 47 The official thesis of the Austrian government, announced after the war, was ‘that the “annexation” was brought about not through the negotiation of two equal states, but through outside military pressure and a traitorous terror campaign by a Nazi, Fascist minority, and was finalised by a military occupation employing war measures’.Footnote 48 If Austria had been occupied territory, however, the restoration of its sovereignty would have been ensured as immediately as possible, as was the case with Allied states liberated by other Allies. Thus, this shows that the formal invalidity of consent based on the law of treaties does not automatically create a situation of belligerent occupation.

Furthermore, it is doubtful whether, under Article 46 of the Vienna Convention, manifest violations of a provision of internal law of fundamental importance relating to competence to consent could negate the legal effect of consent to foreign ‘occupation’ as a relative ground of invalidity.Footnote 49 An example is consent given by a ‘revolutionary’ de facto government, which is manifestly unconstitutional but which effectively controls the national territory.Footnote 50 By way of further example, the agreement concluded by the US and Iraq, the procedures for which were allegedly in violation of Iraqi constitutional law,Footnote 51 could end a belligerent occupation of Iraq.

The validity of consent for military control of its territory expressed by a sovereign state to a foreign state may first be ascertained in accordance with the law of treaties. However, the resulting (in)validity of the consent does not automatically determine the (in)applicability of the law of occupation for the territory concerned. As the above analysis demonstrates, even in cases where validity was doubtful through the lens of the law of treaties, such consensual foreign ‘control’ may be classified as belligerent occupation when other factors are considered. Admittedly, on the other hand, a consent that is valid as an international agreement can be prima facie evidence for the exclusion of the law of occupation; the limitation of this premise will be examined later in the article.

3.2. Who Can Consent?

The validity of consent given by a government whose political stability is fragile should be considered. From the classical legitimist view, it can be argued that only a de jure government is able to give consent to the presence of a foreign force.Footnote 52 However, in accordance with the theory that the classification of a situation for the purpose of the law of occupation must be made only on the basis of the prevailing facts, it is necessary to determine which entity would be entitled to permit the presence of a foreign military with reference to who has de facto effective authority over the territory in question.Footnote 53

3.2.1. The Effective Authority of the Government that Gives the Consent

Among other possibilities in the matrix of potential situations,Footnote 54 the most important to be examined here is whether ‘consent’ given by a de jure government which has lost, partly or totally, effective control over its territory can be considered to be valid. The most probable situation where this would be an issue is when the central government of a state invites a foreign armed force to ‘occupy’ part of its territory, which is effectively and exclusively controlled by rebel forces. Is the application of the law of occupation excluded in such cases? In the context of the legality ad bellum relating to intervention by invitation, such consent could be seen as a justification for the use of force, insofar as the government maintains effective control of a significant part of the remaining national territory.Footnote 55

In applying the law of occupation, on the other hand, a government in exile, which has lost effective control over the entirety of and has even been physically detached from its national territory, sometimes gives consent, thus precluding the application of the law of occupation by a foreign power, which is its ally.Footnote 56 A noteworthy example is the occupation of some Allied territories with prior ‘consent’ by the armed forces of other Allied states (Belgium, the Netherlands and Norway) in the process of liberation from Nazi occupation. These ‘occupations’ did not qualify as belligerent.Footnote 57 They were governed by civil affairs agreements and not explicitly by the law of occupation, although it has been pointed out that the Allies applied the analogous principles and rules of belligerent occupation as a minimum standard with the understanding that the earliest possible handover to a sovereign authority was the prime objective of the occupation.Footnote 58 The fact that governments in exile retained their authority to negate the ‘belligerent’ nature of the occupation demonstrates the formalistic nature of evaluating who has authority to consent for the purposes of classifying a conflict situation. At the same time, however, the fact that the Allies applied analogous rules in these cases is a strong indication that the effectiveness of the authority which consents to the deployment of a foreign army has a crucial impact on the necessity for applying the law of occupation, based on the relationship between the occupying force and the occupied territory/population.

As Ferraro has pointed out, focusing on the prevailing facts also implies not having to enter into sensitive, controversial, and often endless discussions about the legitimacy of the authority concerned. This, in his opinion, would conform to the principle of public international law according to which an entity must be considered the legal government of a state if it is independent and is in fact the effective government.Footnote 59 Nevertheless, commentators commonly emphasise that it is very difficult to determine the validity of consent, as it is always the result of a political process. Furthermore, such consideration has to be carried out on a case-by-case basis. For these reasons, the 2012 ICRC Expert Meeting was able to secure a ‘general view’ with a consensus only in the case of ‘failed states’: when foreign forces intervene in a ‘failed state’, consent must be presumed to be absent because nobody in that state could legitimately consent to foreign intervention.Footnote 60

It is undeniable that the validity of the consent and the effectivity of the government that gives it must be secured because of its significant humanitarian implications, among other factors. In the context of the applicability of the law of occupation, however, some formalistic considerations have a significant impact as, for instance, even governments in exile have given consent to the effect of precluding belligerent occupation. As Sassòli suggests, because the determination of an occupation is a question of fact, consent by an ineffective de jure government would not be seen as sufficient, while mere consent by authorities with de facto control against the will of the de jure authority is equally insufficient.Footnote 61 For the humanitarian benefit of the population, the more persuasive view is that the consent of both de jure and de facto governments is required to exclude the application of the law of occupation.

3.2.2. Occupation of Territories where Sovereignty is Disputed

If the determination of whether a specific area is under occupation is to be based upon objective facts, this should be construed solely on the basis of the effectiveness of control and not on the legal title of the territory in question. This is especially significant in respect of territories where sovereignty is disputed. On this point, the Eritrea-Ethiopia Claims Commission held that ‘neither text [of the Hague Regulations nor the Fourth Geneva Convention] suggests that only territory the title to which is clear and uncontested can be occupied territory’Footnote 62 because ‘the alternative [interpretation] could deny vulnerable persons in disputed areas the important protections provided by international humanitarian law’.Footnote 63 The Commission then concludedFootnote 64 that ‘territory that was peacefully administered by the other party to that conflict prior to the outbreak of the conflict’ should be considered occupied territory. This clearly shows that the existence of occupation is determined solely by the fact of concrete administration and not on the formal title of the territory.Footnote 65 This approach is favourable also for ensuring humanitarian protection, as the necessity for protection tends to correspond with such concrete situations on the ground, and it should not be affected or obscured by an exaggerated dispute over the legitimate title for a particular part of the territory.

3.3. Prior Consent

Additionally, in order to preclude the application of the law of occupation, it can be contended – by analogy with the validity of state consent as a circumstance to exclude wrongfulness – that the ‘consent’ of the territorial state must be given prior to the foreign deployment.Footnote 66 This is necessary in order to prevent consent of the territorial state from being secured after the intervention by coercion or engineering.

Also, from the humanitarian perspective prior consent must be given. Article 47 of the Fourth Geneva Convention provides that ‘[p]rotected persons who are in occupied territory shall not be deprived, in any case or in any manner whatsoever, of the benefits of the present Convention …’ ex post facto ‘by agreement concluded between the local authorities and the Occupying Power … .’ It is empirically observable that once foreign armed forces have established effective control over a certain part of the territory with immense military pressure, the government of that state cannot be said to be in a position to express freely its genuine consent for such military presence. In addition, Article 47 was introduced to address the widespread instances of engineering ‘sovereign consent’ through the installation of a puppet government by a foreign power.Footnote 67

It is obvious, however, that the existence of ‘consent’ by the local government is one of the decisive factors in ending belligerent occupation. Under the traditional law of war, belligerent occupation could be ended by the conclusion of peace. The problem here is that it is difficult to distinguish between, on the one hand, an independent entity (a government) which can give consent ex post facto to an occupation, thereby terminating the belligerent occupation as a matter of law, and a sham government which lacks such power, on the other. Thus, in order to avoid this common abuse as well as the deterioration in the humanitarian protection of the local population, one expert has emphasised that the end of an occupation would begin only with the complete withdrawal of the foreign forces, accompanied by the full reinstatement of the authority of the territorial government.Footnote 68 Based on this theory, the ‘consensual’ presence of former occupying forces with only nominal transfer of its authority to the local government would be regarded as continuing belligerent occupation.Footnote 69

Accordingly, consent for foreign deployment must be given in advance, in order to secure the genuine nature of the consent and, at the same time, to prevent vulnerable people from being compelled to renounce the humanitarian protection under pressure from the foreign forces.

4. How Far Can the Law of Occupation be Humanised?

Apart from cases of doubtful consent, can genuine and valid consent always exclude the application of the law of occupation? The answer would be ‘yes’ for Ferraro, who stressed that ‘it is doubtful whether one could find a single example of belligerent occupation – and the related application of occupation law – that has occurred with the consent of the host state’.Footnote 70 However, as shown below, there are reasonable grounds for asserting that the application of the law of occupation is both possible and beneficial for the population of the territory in some cases of validly ‘consented’ foreign presence.Footnote 71

4.1. The ‘Allegiance Theory’ of Occupation

Driven by the humanisation of IHL to expand the protective reach of the law of occupation, it may be natural for some commentators to weaken the role of consent by the territorial state. For instance, in a paper submitted to the 2012 ICRC Expert Meeting,Footnote 72 Bothe argued that:

[b]elligerent occupation is a situation where typically the tension of interests [between the occupant and the occupied] … exists … [E]ven the existence of a valid agreement does not automatically exclude a situation of occupation. There may still be situations where the same conflict of interests exists.

Benvenisti expanded this approach in a more systematic way.Footnote 73 Presuming that it ‘is the inherent conflict of interests between governments and those governed that the law of occupation seeks to regulate’, he concluded that:

[i]n principle, conflicts of interest between government and governed can arise also in administering ‘friendly’ territory or whenever the ruler is not accountable to the ruled. Since contemporary international law has also recognized the obligation to regulate the internal relations between a state and its citizens, it makes sense to expand the concept of ‘hostility’ to which Article 42 refers to cover other grounds for ‘foreignness’ between an administration and the people subjected to its rule … [A] teleological interpretation of the law of occupation as well as developments in general international law now provide a firm basis for applying the law of occupation beyond situations of clear enmity, to all circumstances in which non-allegiance characterizes the relationship between an administration of territory and the population subject to it.

Benvenisti here relied on the ‘teleological’ (re)interpretation held by the ICTY in Tadić of ‘protected persons’ under the Fourth Geneva Convention.Footnote 74 ‘Protected persons’ are defined in Article 4 of the Convention as ‘those who … find themselves, in case of a conflict or occupation, in the hands of a Party to the conflict or Occupying Power of which they are not nationals’ (emphasis added). The tribunal purposefully expanded this, relying on the preparatory work of the Convention, which indicates that the drafters intended to extend protection to include those who are refugees and thus no longer owe allegiance to their home countries.Footnote 75 According to the tribunal, the object and purpose of the Convention is to protect civilians to the maximum extent possible; therefore, it does not make applicability dependent on formal bonds and purely legal relations.Footnote 76 In line with this reasoning, the tribunal opined that the scope of ‘protected persons’ should encompass other non-allegiant situations between the army and the inhabitants.Footnote 77 Based on this reasoning, the tribunal concluded that Muslims and Croats in Bosnia Herzegovina should have the status of ‘protected persons’, even though they had the same nationality as the perpetrators, Bosnian Serbs. This interpretation has been reiterated by other Chambers of the ICTY and other international tribunals.Footnote 78

This reinterpretation by the ICTY applies also to civilians who have found themselves in occupied territories. That said, could the absence of allegiance be seen as an additional indicator for evaluating the relationship between the occupant and the occupied, along with the consent of the territorial state? Is it even possible or appropriate to redefine the concept of occupied area as one where such ‘protected persons’, who have no allegiance towards the governing force, are present? Surely, in some situations the local population has no allegiance to the foreign military power even if their government consents to its presence.Footnote 79 In situations where part or all of the national territory is under the direct control of a foreign army, or where the latter assumes a ‘security mission’ that directly encounters the local population, those people can hardly have ‘allegiance’ to the foreign army despite the government's consent. This reinterpretation of the concept of occupation could fill the protection gap and detach the necessity of humanitarian protection from the formalistic bond that is ‘nationality’ de jure.Footnote 80 Therefore, it could be a remarkable expansion of Meron's argument of the humanisation of IHL.Footnote 81

However, this theoretical development is refutable on several grounds. First, this redefinition – which Bothe and Benvenisti propose in order to humanise the law of occupation so that it now applies also to consensual ‘occupations’ if there is no allegiance among the local people – generates inconsistencies with the understanding of the concept of occupation as a coercive exercise of control over territory. According to Ferraro and the majority of experts convened at the 2012 ICRC Expert Meeting, as far as the consent is legally valid, military operations by a foreign state based on such consent of the territorial state do not generate a situation of IAC or belligerent occupation. Therefore the law of occupation, by definition, does not apply.Footnote 82

Second, in practice, a lack of allegiance would have to be established collectively, which is substantially more difficult than was the case in Tadić. In that case, a lack of allegiance could be inferred from the victims’ faith or ethnicity. This would make it hard for a detaining or occupying power not to apply the wholesale approach based on the formal bond of nationality.

Third, it does not follow naturally from non-allegiance to a foreign state in whose power the persons find themselves that the area where they reside should qualify as occupied territory. The ICTY's teleological interpretation in Tadić intended to fill the gap of humanitarian protection for certain people on the premise that an armed conflict exists, but not to establish a new threshold of the conflict classification. Thus, one cannot assume that any case of placing the population under foreign military control to which they have no allegiance is identified automatically as armed conflict and belligerent occupation to which IHL applies.

4.2. Humanitarian Protection in Consensual ‘Occupation’

That being said, those conceptual and practical difficulties faced by the ‘allegiance theory’ of Bothe and Benvenisti do not preclude the need to provide legal protection for humanitarian purposes to the local population in certain circumstances, even when there is consent prima facie. Even if belligerent occupation does not exist in a formal sense, people may find themselves in danger of a foreign military to which they have no alignment or allegiance. When such a foreign military force effectively and exclusively controls the area where such people live, legal standards of protection of the type and degree that correspond with the law of occupation would be necessary. Apart from the oversimplified ‘allegiance theory’, on what ground should those people under the control of a validly consented foreign military be classified for protection under the humanitarian rules of, or rules that are comparable with the law of belligerent occupation?

4.2.1. Quasi-occupation

Effective and direct control in consensual foreign military operations

First, when a consented occupying force exercises direct power over the population, substituting the local power (as in Okinawa), or assumes a mission to maintain law and order (as in Iraq after 2004), the risk of potential clashes of interest cannot be ignored, even if the foreign power is deployed with the consent of a democratic government. As a consented operation, it appears prima facie non-conflictual in nature. Yet, the ‘foreignness’ of the force and its soldiers seems to be inevitable. Such ‘foreignness’ can be prominent whenever a foreign force ‘directly’ affects the daily life of the population. Moreover, it is easy to assume that, as a matter of fact, the greater the power that such force wields (‘comprehensiveness’), the greater the risk the population faces. It may be that such clashes between the foreign forces and the population are not sufficiently resolved through the constitutional process of the territorial state or international human rights law. Instead, one can argue that such conflict of interest may be better regulated by certain rules governed by the law of occupation.

It is noteworthy that such tension between the population and the intervening forces may arise when UN peace operations are deployed based on the request or the consent of the territorial government. When the UN Security Council endorses such intervention under Chapter VII of the UN Charter,Footnote 83 and especially when robust authority to the use of force has been given, peace operations can be tinged with coercive elements, which are far more fierce for the population under such control.Footnote 84 The UN peace operations have assumed many non-coercive mandates, not only to monitor ceasefire agreements or national referendums, but also the maintenance of law and order in the country concerned. More broadly, the UN forces and its civilian components have been in charge of the comprehensive administration of a country in lieu of its government in post-conflict or transitional situations.Footnote 85 As for territorial administration by UN peace operations, its fundamental difference from belligerent occupation is noted,Footnote 86 while some scholars argue, based on the similarity between them, for the application by analogy or ‘partial’ application of the law of occupation in those situations.Footnote 87 At least, as Arai-Takahashi admits, the law of occupation, especially the Fourth Geneva Convention, can serve as a source of inspiration for rules concerning the system of detention and administration of justice in the context of UN territorial administration.Footnote 88

Occupation ultra pactum

In situations of consensual occupation the rights and obligations of the deployed forces are generally grounded on and derive from the ‘consent’ given by the territorial state.Footnote 89 Legality ad bellum of the consensual military operation will also be grounded on the consent of the territorial state. In this sense, the UN General Assembly Resolution 3314 (XXIX), on the definition of aggression, includes as an act of aggression:Footnote 90

the use of armed forces of one State which are within the territory of another State with the agreement of the receiving State, in contravention of the conditions provided for in the agreement or any extension of their presence in such territory beyond the termination of the agreement.

Jus ad bellum rules can surely function as some limitation on such an excessive exercise of the foreign military powers beyond the extent of the agreement.Footnote 91

However, the issue may not be completely resolved with this simple solution. The reason for this is the difficulty in inferring clear-cut ‘codes of conduct’ for the foreign forces only from sovereign consent. In some cases, especially when the deployment takes place without concluding a status of force agreement, such consent as such cannot provide sufficient guidance to regulate the conduct and authority of the foreign forces because of its simplicity or vagueness.Footnote 92 As in the case of Okinawa between 1952 and 1971, a territorial state may grant a carte blanche, delegating a comprehensive and exclusive right to control part of its territory to the other without any legal conditions. Additionally, ‘consent’ by or ‘agreement’ with the local government as such may not be sufficiently effective, in reality, to control and discipline the actions of the foreign forces.

Accordingly, foreign military control that could be justified ad bellum through sovereign consent, which is mostly unable to regulate the day-to-day conduct of the forces, would be better regulated by the law of occupation for the purposes of filling the humanitarian gap to the greatest extent possible. Similarly, it is noteworthy that in many cases of consensual occupation, the rules and principles of occupation are said to have applied by analogy.Footnote 93 This may demonstrate that the law of occupation can provide useful guidance for foreign military governance, even outside the traditional contexts of IAC or belligerent occupation.

Occupation in the context of a non-international armed conflict (NIAC) on foreign soil

When a foreign force has a valid invitation to intervene, its deployment to the territory of a state does not in itself constitute an IAC. It may, however, be involved in another kind of armed conflict: a NIAC with an organised armed group in that territory. Such an extraterritorial conflict with non-state armed groups may exist, according to the recent trend of state practice and commentators,Footnote 94 when the foreign forces fight with a certain degree of intensity with a well-organised armed group in the territory of another country, or when a NIAC spills over from the country of origin to another.

A NIAC that occurs in the context of a validly consented operation may be regulated by Common Article 3 of the Geneva ConventionsFootnote 95 and applicable customary rules. Even in that situation the foreign forces sometimes need to control part or all of the operational area; nevertheless, such international rules for NIAC do not contain an equivalent rule for occupation to that of IAC.Footnote 96 In a traditional internal NIAC, in which the governmental force fights domestically against a rebel force, even if the former regains territory that was previously controlled by the latter, the concept of occupation is unlikely to be relevant.Footnote 97 Governmental forces who reconquer national territory against the rebels will not consider themselves as occupiers, nor will rebels who control part of the national territory.Footnote 98 This is because the essence of belligerent occupation is that it should be exercised over foreign territory.Footnote 99 The authority of the governmental forces to control territory of its own is constitutionally grounded. Additionally, it is unthinkable that states would admit that rebels could have the authority to retain a part of its own territory. On the other hand, in the case of foreign armed forces involving NIAC with a local organised armed group, territorial control by the former is without such constitutional foundation, but based on superficial consent by the government; there are therefore more similarities with belligerent occupation.

Furthermore, from a humanitarian perspective, Sassòli's suggestionFootnote 100 – that the consent of both the ineffective de jure government and the non-state actor with effective control over the territory concerned is necessary to exclude the application of the law of occupation – should also apply in this context. Accordingly, it should be stressed that the consent of the central government cannot always negate the necessity for humanitarian protection granted under the law of occupation for the population of a formerly rebel-held territory which is effectively and exclusively reoccupied by the consensual foreign military.

4.2.2. The Insufficiency of Applicable Human Rights Rules

In these scenarios where neither IAC nor belligerent occupation exists and the law of occupation does not apply formally, what rules can regulate these cases of quasi-occupation? According to the case law of the European Court of Human Rights, a contracting party may owe obligations to ensure respect for human rights outside the national territory not only when the state effectively controls such foreign territory as required by Article 42 of the Hague Regulations, but also when the force has ‘decisive influence’ towards the local government, non-state entity, or population.Footnote 101 This demonstrates the possibility that human rights law obligations apply in situations short of occupation, and include consensual quasi-occupations. When such decisive influence is exercised by the ‘occupying’ force over the ‘territory’ and ‘population’ in cooperation or in parallel with the local government, both authorities have to assume ‘shared’ (concurrent) responsibility to respect and ensure respect for human rights under international human rights law.Footnote 102

Also, in the case of belligerent occupation, a ‘righting’ process – the process of invoking human rights law along with or instead of the law of occupation for the occupied territories – is said now to be under way.Footnote 103 Applying human rights obligations to situations of foreign control of territory, at first glance, seems favourable. However, this trend may be criticised for undermining the rights of people living under occupation, by ignoring the balance between military necessity of the foreign military and the protection of populations ‘built into’ the law of occupation. This is because, as Gross points out, paradoxical as it may seem, human rights may broaden the justifications for limiting rights beyond the scope of a strict interpretation of the law of occupation.Footnote 104 Stressing the human rights obligations of the occupying or quasi-occupying power is, without a doubt, important. Yet, at the same time, if such a delicate balance between the foreign military and the population is needed in a given situation, the law of occupation can better serve as an alternative to the applicable human rights norms in circumstances of quasi-occupation.

5. Concluding Remarks

IHL, especially that portion known as the ‘Geneva system’, has been described as one of the most remarkable manifestations of ‘anti-formalism’ in international law.Footnote 105 It is further advocated that IHL as such has been humanised, meaning not only has the denomination changed from the law of war, but also in substance, insofar as the emphasis of the law has moved from protection of the state's interest to that of individuals. As the Tadić judgment made clear, the object and purpose of the law is believed to be to protect civilians to the maximum extent possible.

Contextualising this transformation of the underlying philosophy of IHL in the evaluation of the significance of sovereign consent, which traditionally had been understood as a mere indication of the non-existence of a formal state of war, it has, since 1949, come to entail substantial meaning: the non-existence of an actual hostile relationship between the occupant and the occupied country. Furthermore, under contemporary humanised IHL, could sovereign consent be understood to show the occupant's friendly nature towards not only the occupied country but also the local population? Could it be understood as evidence that humanitarian protection by the law is not necessary for the population in this area? As a logical basis for these accounts, sovereign consent should maintain not only formalistic legitimacy, indicating that the government can represent the state, but also substantial indicia that the government is entitled to express the humanitarian or other needs of its population. In order to ensure the latter legitimacy, however, sovereign consent must be given with additional qualifications, which are not necessary for evaluating the validity of consent from other perspectives of international law. In modern international law, such a ‘substantial’ degree of legitimism is not required for proof of statehood or government representation.Footnote 106

Accordingly, it may be based upon a legal fiction to conclude that sovereign consent, by itself, is sufficient to demonstrate the lack of necessity for humanitarian protection of its population. Surely many fictions are found in the very foundation of IHL. Fundamentally, the legal definition of civilian/combatant status is based on Rousseau's fiction that soldiers fight each other as state organs, not as human beings.Footnote 107 Reasoning that the belligerent force is targetable at any time except when it surrenders or is hors de combat necessitates the fiction that soldiers always pose a military threat unless they are formally discharged. The protection of civilians from a direct attack is based on another fiction regarding their innocence. These fictions have been challenged in the face of the reality of modern warfare.Footnote 108 Above all, the presupposition that such sovereign consent may indicate a lack of necessity for humanitarian protection for the local population could be regarded as not only fictitious but also artificial in some cases.

Belligerent occupation is frequently misconstrued as an anomaly or even an aberration of war,Footnote 109 but, in reality, it is a natural phenomenon of war. Parties to a conflict usually spare no effort in penetrating and, if possible, taking possession of enemy-held territory.Footnote 110 Thus, however different the legal grounds on which these operations are based, they may occur not only in IAC but also in other contexts, such as consensual foreign military operations or extraterritorial NIAC with non-state actors. Since the traditional law of occupation presupposed declared wars in the past or inter-state armed conflicts after 1949, the occupation regime has been tightly connected with the existence of IAC, excluding its applicability to consensual occupation and NIAC. As occupation is a natural part and parcel of armed conflict, this exclusion may be seen as excessively formalistic and artificial, given the trend of the ‘anti-formalistic’ humanisation of IHL. The previous section illustrated possible cases where a consensual ‘occupation’ would be better regulated by the law of occupation. The traditional law of occupation may not cover those situations, while it is obviously necessary that non-allegiant populations trapped in such situations should be protected from an overpowering foreign military control.

As was also implied in the previous section, it should be noted that the law of occupation is not the sole source of ‘rules’ to which reference should be made. International human rights law or IHL rules applicable to NIACFootnote 111 can complementarily regulate the discretion of foreign forces executing consensual ‘occupation’. These areas of international law had not emerged as independent branches of the law when the law of occupation was shaped during the late nineteenth to the early twentieth centuries. The law of occupation, thus, can be seen as an earlier attempt to humanise the law of armed conflict, focusing on wartime situations of people under foreign control that was perceived as especially dangerous at the time. Under contemporary international law, on the other hand, IHL has expanded the scope of application to non-traditional situations, possibly including ‘consensual’ occupation, although the protection regime for such situations has been multi-layered in accordance with the development of international law.Footnote 112

Footnotes

The author is grateful to Professor Yutaka Arai, Brussels School of International Studies, University of Kent, and Mr Kosuke Onishi, PhD Candidate, Doshisha University, for their helpful comments and suggestions. All views expressed in this article and all errors remain those of the author. This work was supported by JSPS KAKENHI Grant Numbers 15K03147 and 18K01287, and Doshisha University Overseas Research Travel Allowance.

References

1 Project of an International Declaration concerning the Laws and Customs of War, Brussels, 27 August 1874, Proceedings of the 1874 Brussels Conference on the Draft International Convention on War: Protocols of the Plenary Sessions: Protocols of the Committee Delegated by the Conference: Annexes, 61.

2 Hague Convention (II) with respect to the Laws and Customs of War on Land and its Annex: Regulations concerning the Laws and Customs of War on Land (entered into force 4 September 1900), Martens Nouveau Recueil (ser 2) 949 (Hague II); Hague Convention (IV) respecting the Laws and Customs of War on Land and its Annex: Regulations concerning the Laws and Customs of War on Land (entered into force 26 January 1910), Martens Nouveau Recueil (ser 3) 461 (Hague IV).

3 Geneva Convention (IV) relative to the Protection of Civilian Persons in Time of War (entered into force 21 October 1950) 75 UNTS 287 (GC IV).

4 Eyal Benvenisti, ‘Occupation, Belligerent’ in Rüdiger Wolfrum (ed), Max Planck Encyclopedia of Public International Law (Oxford University Press 2009) para 1.

5 Meron, Theodor, ‘The Humanization of Humanitarian Law’ (2000) 94 American Journal of International Law 239CrossRefGoogle Scholar.

6 Tristan Ferraro, Legal Adviser to the International Committee of the Red Cross (ICRC), flatly concludes that ‘the existence of consent is simply incompatible with the institution of belligerent occupation’ (emphasis added): Ferraro, Tristan, ‘Determining the Beginning and End of an Occupation under International Humanitarian Law’ (2012) 94 International Review of the Red Cross 132CrossRefGoogle Scholar, 153; see also Dinstein, Yoram, The International Law of Belligerent Occupation (Cambridge University Press 2009) 35–37CrossRefGoogle Scholar; Koutroulis, Vaios, Le début et la fin de l'application du droit de l'occupation (Pedone 2010) 76–89Google Scholar.

7 Tristan Ferraro, ‘Expert Meeting: Occupation and Other Forms of Administration of Foreign Territory’, ICRC Report, March 2012, 10.

8 As to the applicability of the law of occupation in Iraq after 2004 see Dörmann, Knut and Colassis, Laurent, ‘International Humanitarian Law in the Iraqi Conflict’ (2004) 47 German Yearbook of International Law 293Google Scholar; Roberts, Adam, ‘The End of Occupation: Iraq 2004’ (2005) 54 International and Comparative Law Quarterly 27CrossRefGoogle Scholar; Carcano, Andrea, ‘End of the Occupation in 2004? The Status of the Multinational Force in Iraq after the Transfer of Sovereignty to the Interim Iraqi Government’ (2006) 11 Journal of Conflict and Security Law 41CrossRefGoogle Scholar; Wills, Siobhan, ‘The Legal Characterisation of the Armed Conflicts in Afghanistan and Iraq: Implications for Protection’ (2011) 58 Netherlands International Law Review 173CrossRefGoogle Scholar; Wills, Siobhan, ‘The Obligations Due to Former “Protected Persons” in Conflicts that have Ceased to be International: The People's Mujahedin Organization of Iran’ (2010) 15 Journal of Conflict & Security Law 117CrossRefGoogle Scholar.

9 Wall, Andru E, ‘Civilian Detentions in Iraq’ in Schmitt, Michael N and Pejic, Jelena (eds), International Law and Armed Conflict: Exploring the Faultlines: Essays in Honour of Yoram Dinstein (Martinus Nijhoff 2007) 413Google Scholar, 414 (‘Not only did detentions in Iraq continue after the transfer of sovereignty but they skyrocketed – from around 4,000 detainees in the spring of 2004 to over 11,350 less than one year later … [B]y December 2005 14,000 detainees were being held by Coalition Forces’).

10 Carcano (n 8) 58; Dörmann and Colassis (n 8) 307–11.

11 As to the difference between land-lease and belligerent occupation, see Strauss, Michael J, Territorial Leasing in Diplomacy and International Law (Brill 2015) 198–202CrossRefGoogle Scholar.

12 Roberts, Adam, ‘What is a Military Occupation?’ (1985) 55 British Yearbook of International Law 249CrossRefGoogle Scholar, 277.

13 For a concise report in English on this, see Caroli, Rosa, ‘The Other Occupation of Japan: The Case of Okinawa’ in Caroli, Rosa and Basosi, Duccio (eds), Legacies of the U.S. Occupation of Japan: Appraisals after Sixty Years (Cambridge Scholars 2014) 109, 109–29Google Scholar.

14 Treaty of Peace with Japan (with two declarations) (entered into force 28 April 1952) 136 UNTS 46.

15 ibid art 1(a) reads: ‘The state of war between Japan and each of the Allied Powers is terminated’.

16 ibid art 3 reads: ‘Japan will concur in any proposal of the United States to the United Nations to place under its trusteeship system, with the United States as the sole administering authority, Nansei Shoto, … Pending the making of such a proposal and affirmative action thereon, the United States will have the right to exercise all and any powers of administration, legislation and jurisdiction over the territory and inhabitants of these islands’.

17 It should be noted that the legal situation in mainland Japan was sharply different from that of the Islands. Okinawa and other small surrounding islands were landed and invaded by the US forces in April 1945. After the US established effective control of those islands, a military government was formed on the basis of the Hague Regulations. A few months later, on 14 August, the Japanese government announced its surrender to the Allied powers, and the ‘occupation’ started after its expression of acceptance of the allied forces’ presence on the mainland. This ‘consent’ of Japan is comparable with a ‘ceasefire’ agreement, which cannot exclude the application of the law of occupation, while, in reality, this occupation of mainland Japan was not seen as belligerent occupation: Ando, Nisuke, Surrender Occupation, and Private Property in International Law: An Evaluation of US Practice in Japan (Oxford University Press 1991) 81ffGoogle Scholar. Unlike the case of Germany to which the theory of debellatio could apply, the non-application of the law of occupation to Japan from 1945 to 1952, despite its application to Okinawa during the same period, may be seen as legally unsound.

18 Statement by John Foster Dulles, 25(1) The Department of State Bulletin, 17 September 1951, 452, 455 (‘Several of the Allied Powers urged that the treaty should require Japan to renounce its sovereignty over these islands in favor of United States sovereignty. Others suggested that these islands should be restored completely to Japan. In the face of this division of Allied opinion, the United States felt that the best formula would be to permit Japan to retain residual sovereignty, while making it possible for these islands to be brought into the U.N. trusteeship system, with the United States as administering authority’ (emphasis added)). In international law generally, the term ‘residual sovereignty’ is sometimes used to describe the right retained by a lessor state for leased territories, such as Guantanamo Bay or Hong Kong. O'Connell specifies the meaning of such sovereignty as the ultimate power to dispose of the territory: DP O'Connell, International Law, Vol 1 (Stevens and Sons 1965) 352, 354. Applying this to the context of art 3 of the Japanese Peace Treaty, the residual sovereignty retained by Japan was thought to be nothing more than ‘the right to expect that the United States will not transfer the Ryukyus, including Okinawa, to any third party’: House of Representatives, Committee on Armed Services (87th Congress 2d Session), Report No 1684, 16 May 1962, 5.

19 As a result of a series of tough political negotiations between the two countries later on, the reversion of the Islands to Japan was realised in 1972: Japan and United States of America, Agreement concerning the Ryukyu Islands and the Daito Islands (with Agreed Minutes and Exchanges of Notes), signed at Tokyo and Washington on 17 June 1971 (entered into force 15 May 1972) 841 UNTS 275. However, it was clear that in 1951 this ‘lease’ of the Islands was supposed to last for much longer, as can be seen from this private message to the US by the Emperor of Japan: ‘The Emperor further feels that United States military occupation of Okinawa … should be based upon the fiction of a long-term lease – 25 to 50 years or more – with sovereignty retained in Japan’: ‘Enclosure to Dispatch No. 1293, 22 September 1947 from the US Political Adviser for Japan, Tokyo, on the subject “Emperor of Japan's Opinion concerning the Future of the Ryukyu Islands”’ in WJ Sebald, ‘Emperor of Japan's Opinion concerning the Future of the Ryukyu Islands’, Okinawa Prefectural Archives, code 0000017550, http://www.archives.pref.okinawa.jp/wp-content/uploads/Emperors-message.pdf.

20 USCAR Proclamation No 13, which established the local governing body, clearly provides: ‘The Government of the Ryukyu Islands may exercise all powers of government within the Ryukyu Islands, subject however to the Proclamations, Ordinances, and Directives of the United States Civil Administration of the Ryukyu Islands (Art 2)’: Civil Administration Proclamation No 13, 29 February 1952, 1 Okinawasha, Gekkan, Laws and Regulations during the U.S. Administration of Okinawa, 1945–1972 (Ikemiya Shokai & Co 1983) 112Google Scholar (Laws and Regulations).

21 Even after the entry into force of the Peace Treaty on 28 April 1952, it was provided that ‘[a]ll proclamations, ordinances and directives of the United States Civil Administration of the Ryukyu Islands and of United States authorities previously exercising military government in the Ryukyu Islands … shall remain in full force and effect to the extent’: Civil Administration Proclamation No 22, 30 April 1953, 1 Laws and Regulations 117. This may be regarded as clear evidence that the basic structure of the military control over the Islands did not change after 1952.

22 Measures taken by the US military for land acquisition should be mentioned as a notable example. After the 1907 Hague Regulations ceased to apply in Okinawa, the USCAR issued Ordinance No 109 (3 April 1953, 2 Laws and Regulations 49) by which it could unilaterally declare acquisition of land for temporary or indefinite use even without the consent of the landowners, who were left without any due process safeguards and faced physical enforcement by US soldiers in the event of resistance: Miyagi, Etsujiro, ‘The Land Problem (1952–1958)’ in Ota, Masahide, Miyagi, Etsujiro and Hosaka, Hiroshi (eds), A Comprehensive Study on U.S. Military Government in Okinawa (An Interim Report) (Chapter 2) (University of the Ryukyus 1987) 36, 36–65Google Scholar, http://ir.lib.u-ryukyu.ac.jp/bitstream/20.500.12000/13849/3/62041071-3.pdf.

23 A comparable situation may be that of Guantanamo Bay which, like Okinawa, was indefinitely leased to the US for use as a naval base pursuant to an agreement with Cuba. The US government, for a long time, claimed the inapplicability of the US Constitution and its laws to Guantanamo, because it is not US territory. It is well known that this tricky status of the base was used to preclude any legal regulation regarding the treatment of detainees held there, granting full discretion to the Executive Power. However, this assertion was rejected in Boumediene v Bush 553 US 723 (2008), in which it was held that the extent of US control over the Guantanamo Bay Naval Base entails the extension of constitutional rights to protect foreign nationals detained there. As to the difference between land-lease and belligerent occupation in relation to Guantanamo Bay, see Strauss, Michael J, The Leasing of Guantanamo Bay (Praeger Security International 2009) 98–103Google Scholar.

24 With regard to the Soviet intervention in Czechoslovakia and Afghanistan, many states criticised the genuineness of the consent of the respective governments. For Czechoslovakia, its representative stated in the UN Security Council that foreign troops had crossed the borders without the knowledge of the Czechoslovak authorities: Security Council Official Records, 23rd year, 1441st meeting, 21 August 1968, UN Doc S/PV.1441, para 137. For Afghanistan, the Soviet assertion that the Afghan government approved the invasion lost its credibility when the Afghan President was murdered two days after the invasion. With regard to the US intervention against Panama, similar doubt can be cast because of the complex set of facts pertaining to the allocation of power inside the Panamanian government, which was controlled by the de facto leader General Manuel Noriega: see Nolte, George, ‘Intervention by Invitation’ in Wolfrum, Rüdiger (ed), Max Planck Encyclopedia of Public International Law (Oxford University Press 2011)Google Scholar para 5.

25 Hague Regulations (n 2) art 42.

26 Unlike the French text, ‘the hostile army’ in the English text may seem to have a ‘functional/non-formalistic’ connotation, which is in conformity with the assertion in the later part of this article, but this is not significant, as English is not an authentic text of the Hague Conventions: Schindler, Dietrich and Toman, Jiri (eds), The Laws of Armed Conflicts: A Collection of Conventions, Resolutions and Other Documents (4th revised and completed edn, Martinus Nijhoff 2004) 56Google Scholar.

27 Ferraro (n 7) 20–22.

28 International Law Commission, Articles on Responsibility of States for Internationally Wrongful Acts, with Commentaries (2001), Report of the ILC 53rd sess, UN Doc A/56/10, 2001(II) Yearbook of the International Law Commission 26, art 20.

29 Nolte (n 24).

30 For instance, the Report of the Independent International Fact-Finding Mission on the Conflict in Georgia (vol II, September 2009) presented a confusing expression: ‘If … Russia's military intervention cannot be justified under international law, and if neither Abkhazia nor South Ossetia is a recognised independent state, IHL – and in particular the rules concerning the protection of the civilian population (mainly Geneva Convention IV) and occupation – was and may still be applicable’: ibid 311.

31 Dinstein (n 6) 31–32, 35.

32 Geneva Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field (entered into force 21 October 1950) 75 UNTS 31 (GC I); Geneva Convention (II) for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of Armed Forces at Sea (entered into force 21 October 1950) 75 UNTS 85 (GC II); Geneva Convention (III) relative to the Treatment of Prisoners of War (entered into force 21 October 1950) 75 UNTS 135 (GC III); GC IV (n 3).

33 Pictet, Jean S (ed), Commentary: IV Geneva Convention Relating to the Protection of Civilian Persons in Time of War (ICRC 1958) 21–22Google Scholar.

34 Since GC IV is ‘supplementary’ to Section II of the Hague Regulations (art 154), this expansion of the concept of armed conflict and occupation also affects the application of the latter rules.

35 Kolb, Robert and Vité, Sylvain, Le droit de l'occupation militaire: Perspectives historiques et enjeux juridiques actuels (Bruylant 2009) 112Google Scholar.

36 ibid 113; Kolb, Robert, Advanced Introduction to International Humanitarian Law (Edward Elgar 2014) 101CrossRefGoogle Scholar.

37 Nolte (n 24) para 26.

38 Ferraro (n 6) 134–36.

39 Ferraro (n 7) 21.

40 Daillier, Patrick, Forteau, Mathias and Pellet, Alain, Droit International Public (8th edn, LGDJ 2009) 208Google Scholar.

41 The logic here is borrowed from the theory of the ‘margin of appreciation’ relating to the application of human rights obligations: see ECtHR, Handyside v United Kingdom, App no 5493/72, 7 December 1976, para 48.

42 Vienna Convention on the Law of Treaties (entered into force 27 January 1980) 1155 UNTS 331.

43 Ferraro (n 7) 21.

44 In addition to the example of the Nazi ‘protectorate’ of Bohemia and Moravia, and the Anschluss (which are discussed below), Japan's invasion of French Indochina with the consent of the Vichy government in 1940–45 may be another example that warrants detailed examination. After the defeat of France and the establishment of the Vichy government, Japan decided to deploy armed forces to the Tonkin province (North Vietnam) in order to cut the assistance route from southeast Asia to the Chiang Kai-shek military in China. Japan, with harsh military pressure, convinced the Vichy government to agree to the deployment in 1940. The Vichy decision was resisted at times and agreed to at others by the local colonial governor of French Indochina, who, unlike his colleagues in Africa or other parts of the world, did not leave the Vichy side until the final period of the Second World War. Plus, the local French army opposed, incidentally but also quite seriously, the Japanese invasion at that time. During the following year Japan invaded the southern part of French Indochina, adopting the same tactic. After the Allied liberation of France, Japan gave French Indochina an ultimatum and became involved in a state of war in March 1945, initially in the formal sense. Before 1945, especially during 1940–41, Japan seemingly handled the agreements with the greatest of care so as not to be seen as a belligerent occupation, but we can trace some grounds to negate the legal validity of these Franco (Vichy)–Japanese agreements: the entitlement of the Vichy government to represent France, military threat by Japan, and the opposition of the local governor.

The Tokyo Trial of Major War Criminals of Japan recognised that these 1940–41 ‘invasions’ by Japan against French-Indochina represented the first step in the ‘push to the South’ policy, and such coercion upon the Vichy government constituted direct threats of illegal force: International Military Tribunal for the Far East, United States et al v Sadao Araki et al, Judgment of 24 November 1948, reproduced in Boister, Neil and Cryer, Robert (eds), Documents on the Tokyo International Military Tribunal: Charter, Indictment and Judgments (Oxford University Press 2008) 477–78Google Scholar, 498–99. See also Indictment, ibid 40–41. Subsequent trials of Japanese war criminals by France in Saigon, in which 230 suspects in 39 cases were charged, limited its scope to cases after March 1945. Presumably, this was based on the view of the parties that incidents relating to military invasion and ‘occupation’ from 1940 to March 1945 were not covered by a trial to adjudge ‘war crimes’.

For reference information, the French government has not declassified any documents relating to these Saigon trials, while the Japanese government has been given some of these documents by France and has declassified them. The above description of the Saigon trials is based on research by Professor Chizuru Namba, ‘Kokuritsu Kou-bunsho Kan Shozo no “Saigon Saiban” Kanren Shiryo ni tsuite’ (in Japanese with English summary) [‘Materials concerning the “Saigon Trials” in the Possession of the National Archives of Japan’] (2008) 41 Kitanomaru [Journal of the National Archives of Japan] 81–79.

45 Second Report on the Law of Treaties, by Sir Humphrey Waldock, Special Rapporteur, Report of the ILC 15th sess, UN Doc A/CN.4/156 and Add.1-3, 1963(II) Yearbook of the International Law Commission 36, 50, art 11. The Waldock 1963 commentary referred to other precedents of invalid agreements brought about by coercion of national representatives, including coercion by Japan of the Emperor of Korea and his ministers in 1905 to obtain their acceptance of a protectorate treaty: Agreement between Corea and Japan by which Japan assumed Charge of the Foreign Relations of Corea (entered into force 17 November 1905) 199 CTS 399. Although its validity was not questioned by the great powers at the time (including Britain, Russia and the US), a historical survey (Fukuju Unno, Kankoku Heigō (Iwanami Shoten1995) 157–61 (in Japanese)) suggests that it is difficult to find any substantial difference between the situations surrounding the Korea-Japan protectorate treaty of 1905 and that of President Hácha in 1939. As Waldock pointed out, coercion is usually a mixture of personal pressure on the individuals and threats against their people (ibid). This has intensified the debate on the validity of the 1905 Korea-Japan treaty between the two states up to the present day.

46 International Military Tribunal, United States et al v Hermann Wilhelm Göring et al, Judgment, 1 October 1946, 11 Trial of The Major War Criminals before the International Military Tribunal 192–94.

47 Clute, Robert E, The International Legal Status of Austria 1938–1955 (Martinus Nijhoff 1962) 11–22CrossRefGoogle Scholar.

48 ibid 12 (emphasis added). Additionally, Federal Chancellor Figl stated that ‘Austria was at that time occupied in a manner contrary to international law and robbed of its capacity to negotiate without (the Anschluss) being legitimized by legal title’: ibid (emphasis added). It is unclear whether this reference to ‘military occupation’ was intended to mean that the government claimed that the law of occupation should have applied, or if it just criticised the annexation as illegal ad bellum.

49 As for the distinction between ‘absolute’ and ‘relative’ invalidity of treaties, see von der Decken, Kerstin, ‘Article 42’ in Dörr, Oliver and Schmalenbach, Kirsten (eds), Vienna Convention on the Law of Treaties: A Commentary (2nd edn, Springer 2018) 789Google Scholar, mn 12–14.

50 Ferraro (n 7) 21–22.

51 ibid.

52 Ferraro (n 6) 153.

53 ibid.

54 As for other possibilities, the consent of a de jure and effective government does not pose any legal difficulty, while that of a non-de jure and ineffective government is legally and militarily meaningless. The other significant possibility in the matrix is that of consent of a non-de jure but effective authority which has effective control over part of the territory. Can such authority give consent to another country to deploy its armed forces in the territory over which it has achieved control? First, this intervention generates international armed conflict (IAC) between the intervening state and the territorial state, and it is widely recognisable that the law of occupation would apply to the situation, even if the intervening state denies it.

55 Nolte (n 24) paras 14–18.

56 However, in the cases of France, Indo-China and Denmark, there were no prior agreements, which were addressed only after the ‘occupations’ were established: Kelly, Michael J, ‘Non-Belligerent Occupation’ (1999) 28 Israel Yearbook of Human Rights 17Google Scholar, 29.

57 Dinstein (n 6) 37.

58 Roberts (n 12) 264; Kelly (n 56) 29.

59 Ferraro (n 6) 153–54.

60 Ferraro (n 7) 23. It should be noted that even this observation, although agreed by consensus, is not as solid as it may seem, as the experts simply indicated a solution for the extreme end of the spectrum. There exists, however, a vast ‘grey zone’ between an undisputedly legitimate and effective government on the one hand, and the complete anarchy of a ‘failed state’ on the other. In such a grey zone, of course, the solution will be achieved by case-by-case considerations. The taxonomy of situations has to depend on the balance of the parties’ power, effectivity or legitimacy, and the political consideration of the outer world. Above all, it seems almost impossible for the experts to reach an agreement on the definition of ‘failed states’. An interesting example of ‘political manipulation’ of a situation involving a ‘failed state’, motivated by the interests of regional powers, is the treatment of the Transitional Federal Government (TFG) after 2004 in Somalia, the clearest example of a ‘failed state’. As to the process that the TFG, which had not exercised any control over the territory, used to entrench its position vis-à-vis the international community as the only viable and recognisable government of Somalia, see Lieblich, Eliav, International Law and Civil Wars: Intervention and Consent (Routledge 2013) 165–69CrossRefGoogle Scholar.

61 Sassòli, Marco, ‘The Concept and the Beginning of Occupation’ in Clapham, Andrew, Gaeta, Paola and Sassòli, Marco (eds), The 1949 Geneva Conventions: A Commentary (Oxford University Press 2015) 1389, 1402–03Google Scholar.

62 Eritrea-Ethiopia Claims Commission, Partial Award, Central Front, Ethiopia's Claim 2, Decision of 28 April 2004, 26 Reports of International Arbitral Awards 155, para 29.

63 ibid para 28.

64 ibid para 27. It should be noted that the boundary dispute between the two countries was referred to in a separate procedure by the Boundary Commission established by the Algiers Agreement of 12 December 2000 (Agreement between the Government of the Federal Democratic Republic of Ethiopia and the Government of the State of Eritrea (entered into force 12 December 2000) 2138 UNTS 94), by which the Claims Commission was also established.

65 The applicability of GC IV in the West Bank was upheld by the International Court of Justice (ICJ) regardless of whether Jordan had any rights in respect thereof before 1967: Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, Advisory Opinion [2004] ICJ Rep 136, [93]. The United Kingdom took a broader view: ‘[I]f, during an armed conflict, a state takes military control of a territory it did not control before the conflict the Convention is applicable, whatever the underlying disputes about title’ (Marston, Geoffrey (ed), ‘United Kingdom Materials on International Law 1998’ (1999) 69 British Yearbook of International Law 443, 598–600Google Scholar (emphasis added)).

66 Crawford, James, State Responsibility: The General Part (Cambridge University Press 2013) 287CrossRefGoogle Scholar.

67 Pictet (n 33) 274–75.

68 Ferraro (n 7) 29.

69 Accordingly, Benvenisti argues that the occupation of Iraq ceased in April 2005, rather than June 2004, when the new government was approved by the Iraqi Parliament endorsed by Iraqi people through election: Benvenisti, Eyal, The International Law of Occupation (2nd edn, Oxford University Press 2012) 254–56CrossRefGoogle Scholar.

70 Ferraro (n 6) 153.

71 As a possible example of such ‘occupation’, the Syrian military presence, by agreement with Lebanon, from 1990 to 2005 may be noted. Western countries, in their statements in the Security Council after the adoption of UNSC Res 1559 (2 September 2004), UN Doc S/RES/1559, requesting ‘all remaining foreign forces to withdraw from Lebanon’, stressed how significantly the Syrian government had imposed its political will on Lebanon, although they did not directly refer to the classification of the Syrian forces as the occupying forces in the sense of IHL: ‘Security Council Declares Support for Free, Fair Presidential Election in Lebanon; Calls for Withdrawal of Foreign Forces There’, Press Release, SC/8181, 2 September 2004, https://www.un.org/press/en/2004/sc8181.doc.htm.

72 ‘“Effective Control”: A Situation Triggering the Application of the Law of Belligerent Occupation, Background Document by Professor Michael Bothe’, Annex I of Ferraro (n 7) 37.

73 Benvenisti (n 69) 59–60.

74 ibid; ICTY, Prosecutor v Tadić, Judgment, IT-94-1-A, Appeals Chamber, 15 July 1999 [164]–[168].

75 ibid [164].

76 ibid [166], [168].

77 ibid [169].

78 ICTY, Prosecutor v Aleksovski, Judgment, IT-95-14/1-A, Appeals Chamber, 24 March 2000, [151]–[152]; ICTY, Prosecutor v Mucić and Others, Judgment, IT-96-21-A, Appeals Chamber, 20 February 2001, [81]–[84]; ICC, Prosecutor v Germain Katanga and Mathieu Ngudjolo Chui, Decision on the Confirmation of Charges, ICC-01/04-01/07, Pre-Trial Chamber I, 30 September 2008, [290]–[291].

79 It may seem possible, as an extension of this interpretation, to negate the application of the law of occupation when the central government does not consent but the population does have allegiance to or sympathy with the occupying force with which they share the same ethnicity (eg, Crimea, South Ossetia, Abkhazia or Northern Cyprus). However, such a positive attitude of the local population towards the occupant could not be equated with state consent. As one of the functions of the law is to preserve the status quo of the occupied territory, the territorial state is a beneficiary of the law of occupation. Humanitarian protection of the local population is but another of its important functions. Thus, the law of occupation should not be excluded in such cases.

80 Hoffmann points out that this emphasis on ‘allegiance’ for interpreting IHL is inconsistent with the prevailing understanding of ‘nationality’ in general international law: Hoffmann, Tamás, ‘The Perils of Judicial Construction of Identity – A Critical Analysis of the International Criminal Tribunal for the Former Yugoslavia's Jurisprudence on Protected Persons’ in Jenkins, Fiona, Nolan, Mark and Rubenstein, Kim (eds), Allegiance and Identity in a Globalised World (Cambridge University Press 2014) 497Google Scholar.

81 Meron (n 5) 256–60.

82 Ferraro (n 7) 21.

83 Charter of the United Nations (entered into force 24 October 1945) 1 UNTS XVI.

84 Many commentators argue that Chapter VII operations categorically trigger the application of the law of armed conflict, including the law of occupation. However, this is not always sound, nor does it always reflect the ‘realistic’ attitude of the parties concerned. In some cases during the 1990s – such as NATO's bombing of Bosnian Serbs or the Quick Reaction Force in Somalia supporting peacekeepers on the ground, both of which were authorised by a Security Council Chapter VII resolution – the participating countries tried to negate the application of the law of armed conflict and those requests were seemingly accepted by the UN: Lepper, Steven J, ‘The Legal Status of Military Personnel in United Nations Peace Operations: One Delegate's Analysis’ (1996) 18 Houston Journal of International Law 359Google Scholar, 368; The Judge Advocate General's Legal Center and School (United States Army), International and Operational Law Department, ‘Chapter 18, The Law of War (LOW)’ in Operational Law Handbook (Publication JA-422) (1998). It should be also noted that, according to art 2(2) of the Convention on the Safety of United Nations and Associated Personnel (entered into force 15 January 1999) 2051 UNTS 363, it can be argued that Chapter VII operations do not necessarily qualify as ‘an enforcement action under Chapter VII of the Charter of the United Nations in which any of the personnel are engaged as combatants against organized armed forces and to which the law of international armed conflict applies’. A quick answer is this: in cases where the intervening force authorised by the Security Council engages in organised non-state armed groups, the potentially applicable rules, if any, would be the law of ‘non-international’ armed conflict, as this armed conflict is not fought between states; thus the application of the Convention, which is mutually exclusive vis-à-vis the law of armed conflict, is not excluded.

85 In West New Guinea (1962–63), El Salvador (1991–95), Cambodia (1992–93), Eastern Slavonia, Croatia (1996–98), Timor-Leste (1999–2002) and Kosovo (1999–present).

86 Dinstein (n 6) 37; Arai-Takahashi, Yutaka, The Law of Occupation: Continuity and Change of International Humanitarian Law, and its Interaction with International Human Rights Law (Martinus Nijhoff 2009) 604–05CrossRefGoogle Scholar.

87 Irmscher, Tobias H, ‘The Legal Framework for the Activities of the United Nations Interim Administration Mission in Kosovo: The Charter, Human Rights, and the Law of Occupation’ (2001) 44 German Yearbook of International Law 353Google Scholar; Stahn, Carsten, The Law and Practice of International Territorial Administration: Versailles to Iraq and Beyond (Cambridge University Press 2008) 467–79CrossRefGoogle Scholar.

88 Arai-Takahashi (n 86) 605. As for the Australian army's experience of ‘applying’ GC IV to operations in Somalia, see also Kelly, Michael J, Restoring and Maintaining Order in Complex Peace Operations: The Search for a Legal Framework (Kluwer Law International 1999)Google Scholar.

89 It is for that purpose that states deploying force usually conclude a status of force agreement with the territorial states: Rowe, Peter, ‘Historical Developments Influencing the Present Law of Visiting Forces’ in Fleck, Dieter (ed), The Handbook of the Law of Visiting Forces (Oxford University Press 2018) 13–32Google Scholar.

90 UNGA Res 3314 (XXIX) (14 December 1974), UN Doc A/RES/3314(XXIX), para 3(e) (emphasis added).

91 The non-observance of trivial conditions found in the consent by itself could not necessarily place such conduct outside the limits of said consent. The Special Rapporteur of the ILC at the time, James Crawford, referred to ‘a requirement to pay rent for the use of facilities’ as an example of such trivial conditions and non-observance of which ‘would not transform the visiting force into an army of occupation’: Crawford, James, The International Law Commission's Articles on State Responsibility: Introduction, Text and Commentaries (Cambridge University Press 2002) 165Google Scholar fn 347. However, it should be noted that non-observance of some other conditions may destroy the legal foundation of the consent and constitute an act of aggression according to General Assembly Res 3314 (XXIX).

92 Also, it is difficult to find consistency of state practice that can be evidence of customary international rules. This is largely because such consensual deployment has been conducted on a case-by-case basis.

93 Kelly (n 56) 29.

94 ICRC, Commentary on the First Geneva Convention: Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field (Updated Commentary, Cambridge University Press 2016) 167–74Google Scholar.

95 It should be noted that Additional Protocol II (Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of Non-International Armed Conflicts (entered into force 7 December 1978) 1125 UNTS 609) is not understood to apply to such a scenario, as its art 1(1) provides: ‘This Protocol … shall apply to all armed conflicts … which take place in the territory of a High Contracting Party between its armed forces and dissident armed forces or other organized armed groups’ (emphasis added).

96 The ICRC customary international humanitarian law project recognises the ‘convergence’ of rules applicable to IAC and NIAC in probably most parts of the law, but carefully avoids referring to ‘occupation’ in the context of NIAC. For instance, in its Rule 129, sub-para A, it provides that ‘[p]arties to an international armed conflict may not deport or forcibly transfer the civilian population of an occupied territory …’ (emphasis added), while in sub-para B, a slightly differentiated expression is found: ‘Parties to a non-international armed conflict may not order the displacement of the civilian population, in whole or in part, for reasons related to the conflict …’ (emphasis added): Henckaerts, Jean-Marie and Doswald-Beck, Louise (eds), Customary International Humanitarian Law, Vol 1: Rules (Cambridge University Press 2005, revised 2009) 457–62CrossRefGoogle Scholar.

97 Obviously, in the event that the rebels successfully achieve statehood and are internationally recognised as such, the central government will be regarded as an invader or occupying power. Examples include Pakistan in 1971 vis-à-vis Bangladesh, and Serbia in the 1990s vis-à-vis Bosnia and Herzegovina or Croatia.

98 Kolb (n 36) 37.

99 Baxter, Richard R, ‘Ius in Bello Interno: The Present and Future Law’ in Moore, John Norton (ed), Law and Civil War in the Modern World (Johns Hopkins University Press 1974) 518Google Scholar, 531 (reproduced in Vagts, Detlev F and others (eds), Humanizing the Law of War: Selected Writings of Richard Baxter (Oxford University Press 2013) 271Google Scholar).

100 Sassòli (n 61) 1402–03.

101 ECtHR, Ilascu and Others v Moldova and Russia, App no 48787/99, 8 July 2004, para 392; ECtHR, Catan and Others v Moldova and Russia, App nos 43370/04, 8252/05 and 18454/06, 19 October 2012, para 150.

102 See Heijer, Maarten den, ‘Shared Responsibility before the European Court of Human Rights’ (2013) 60 Netherlands International Law Review 411Google Scholar. For a comprehensive account of state practice of shared responsibility, see Nollkaemper, André and Plakokefalos, Ilias (eds), The Practice of Shared Responsibility in International Law (Cambridge University Press 2017)CrossRefGoogle Scholar.

103 In particular, such a trend is marked by two ICJ cases: Legal Consequences of the Construction of a Wall (n 65) and Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v Uganda), Judgment [2005] ICJ Rep 168.

104 Gross, Aeyal, The Writing on the Wall: Rethinking the International Law of Occupation (Cambridge University Press 2017) 338–96CrossRefGoogle Scholar.

105 Koskenniemi, Martti, ‘Occupation and Sovereignty – Still a Useful Distinction?’ in Engsdahl, Ola and Wrange, Pål (eds), Law at War: The Law as It Was and the Law as It Should Be: Liber Amicorum Ove Bring (Brill 2008) 163Google Scholar, 169.

106 For an interesting argument relating to the validity of sovereign consent with the ability to protect its own people, based on the theory of the responsibility to protect, see Lieblich (n 60) 173–208.

107 As for the influence of Jean-Jacques Rousseau's Du Contrat Social to the development of IHL, see Draper, GIAD, ‘The Development of International Humanitarian Law’ in UNESCO (ed), International Dimensions of Humanitarian Law (Henry Dunant Institute, UNESCO, Martinus Nijhoff 1988) 67, 68–69Google Scholar.

108 Recent debates on the ‘obligation to try to capture enemy soldiers before killing’ question the validity of such a fiction in extreme cases: see Goodman, Ryan, ‘The Power to Kill or Capture Enemy Combatants’ (2013) 24 European Journal of International Law 819Google Scholar; Jens David Ohlin, ‘Recapturing the Concept of Necessity’, Cornell Legal Studies Research Paper No. 13-90, https://ssrn.com/abstract=2230486 or http://dx.doi.org/10.2139/ssrn.2230486. Similarly, debates on elements of the direct participation of civilians in hostilities have reflected the doubts of some people over another fiction about the protective reach for civilians against direct attack: see Melzer, Nils, Interpretive Guidance on the Notion of Direct Participation in Hostilities under International Humanitarian Law (ICRC 2009)Google Scholar.

109 Dinstein (n 6) 1.

110 ibid.

111 Possibly a ‘code of conduct’, which should be included in any status of forces agreement, may function in a similar way.

112 The relationship between this ‘human-centric’ reinterpretation of the law of occupation and the issue of self-determination is not explored in this article, but the following summary remarks: ‘In many documents, foreign occupation, along with colonialism and racially discriminatory regimes, has been stigmatised as a major infringement of the people's right of self-determination. However, from a normative perspective, the schemes of belligerent occupation have developed as a safeguard for the sovereignty of the occupied state against conquest or forceful annexation by the occupant. This means that the law of occupation as such can preserve the precondition under which people in an occupied territory exercise their right to freely determine their political status, as far as, under this law, (a) sovereignty and title to an occupied territory are not vested in the occupying power, (b) the occupying power is entrusted with the management of public order and civil life in the territory under its control, and (c) occupation is temporary, and may neither be permanent nor indefinite’: Gross (n 104) 18. The reinterpretation of the law of occupation asserted in this article could be understood as an extension of such normative safeguards to deceptive practices of consensual military occupation of foreign territories, which are excluded from the protective scope by a formalistic interpretation of the law of occupation. This is, needless to say, subject to adherence to the basic tenets of the law of occupation referred to above. On the other hand, it should be recalled that, as a matter of fact, foreign military rule as such, whether consensual or coercive, could affect the people's self-determination in an adverse way. Additionally, as Gross rightly argues, seeing occupation as ‘neutral’ may actually legitimise new forms of what should be considered illegal, and this danger is almost unavoidable (ibid 20–21). Such dilemma inherent in the law of occupation applies even to the ‘humanised’ law of occupation.