Rethinking Occupation: The Functional Approach

Rethinking Occupation: The Functional Approach

[Aeyal Gross is a Professor at Tel-Aviv University, Faculty of Law.]

This is the first post of our Symposium on the Functional Approach to the Law of Occupation.

Questions regarding the existence of an occupation, and especially its end, came to the fore in 2004-2005 with Israeli pronouncements about the end of its occupation of Gaza, and UN, US, and UK statements about the end of the occupation in Iraq. In the years that followed, I found myself at various events where the question of whether Iraq or Gaza were still “occupied territories” was discussed, at times seemingly ad absurdum. Seated at a conference on “Occupations and Withdrawals” at the University of Glasgow in 2006 and listening to the discussions around me as to whether those territories were still occupied and whether “boots on the ground” are required for an occupation to exist, I felt I was attending a real life enactment of Felix Cohen’s “Heaven of Legal Concepts,” where legal concepts are “thingified” in a way that Cohen characterizes as “transcendental nonsense.”  [See: Transcendental Nonsense and the Functional Approach (1935)]. Some of the discourse on the existence (or absence) of “occupation” was, I thought, an example of legal analysis ignoring practical questions of value or positive fact. Instead, discussions took refuge in “legal problems” that, according to Cohen, can always be answered by manipulating legal concepts in certain approved ways that bar the way to “intelligent investigation of social facts and social policy.” While Israel had removed its settlers and permanent military presence from Gaza, and while the Security Council had proclaimed that the occupation of Iraq was over, the occupying powers continued to exercise extensive control over the daily life of the people residing in these territories. Some of my international law colleagues argued that these territories are no longer occupied, while others disagreed. Listening to this debate,  I began to think that “occupation” should be included in the category Cohen calls “magic solving words” – words that  are actually incapable of solving anything if we remain within the binary on/off framework of the traditional international law of occupation. Article 42 of the Hague Regulations determines that “[t]erritory is considered occupied when it is actually placed under the authority of the hostile army” and that “[t]he occupation extends only to the territory where such authority has been established and can be exercised.” The discussions of Gaza and Iraq illustrated to me that by relinquishing some of the control or by transforming it, occupants may attempt to absolve themselves of their responsibility by claiming that the territory is no longer occupied within the framework set out in Article 42.

Reflecting upon Cohen’s insights, I recalled that, in his legal realist suggestion, norms should not follow from abstract concepts but rather the opposite. For instance, rather than saying that a labor union can be sued because it is “a person” or “a quasi-corporation,” it should be said that a labor union is “a person” or “quasi-corporation” because it can be sued. Whereas the first approach is one coined in “transcendental” terms by asserting something that sounds like a proposition but cannot be confirmed or refuted by positive evidence or ethical argument, the latter avoids this circularity. To follow Cohen, then, we can address the “thingification” of occupation. As is well known, it is by virtue of the determination that the situation is or is not one of occupation that parties are assigned rights and obligations under international law. But an alternative to reliance on “heavenly legal concepts” and “transcendental nonsense” is  a “functional approach” that, in Cohen’s words, “represents an assault upon all dogmas and devices that cannot be translated into terms of actual experience” and from which “all sorts of empirical decisions are supposed to flow.”  “If the functionalists are correct,” argued Cohen, “the meaning of a definition is found in its consequences.” To apply Cohen’s approach to the law of occupation, then, when we ask whether there is an occupation we should consider whether the liabilities and duties of an occupier should be attached to certain acts. This is an ethical question that cannot be answered in purely legal terms since that would make it circular. Rather, we should consider the ethical character of the legal question and the conflicting human values in every controversy. This approach will prevent occupiers from relinquishing responsibility when control is transformed, and will ensure that as long as an occupying party continues to exercise some degree of control, it will continue to be held accountable. In the functional approach, legal decisions are not “products of logical parthenogenesis born of pre-existing legal principles but are social events with social causes and consequences. Law and legal institutions should thus be appraised in terms of some standards of human values.”

My “epiphany,” whereby the endless discussions as to whether an occupation exists are just variations on transcendental nonsense, led me to suggest that we should adopt a  functional approach to occupation. As I consider in detail in my forthcoming book manuscript (The Writing on the Wall: Rethinking the International Law of Occupation), much of this discussion has been conducted in conceptual terms and in a polarized debate often detached from the reality of people’s lives and from decisions about the assignment of responsibilities.

Consider the 2008 Al-Bassiouni case, where the Israeli High Court of Justice (HCJ) held that Israel no longer occupies Gaza, given the abolition of the military government and given that Israeli soldiers are no longer present on an ongoing basis. Accordingly, it concluded, Israel is not bound by the duties of an occupying power concerning the supply of electricity and fuel, and may consequently limit deliveries. This holding was ostensibly derived from the legal concept of “occupation,” with the HCJ finding that the absence of a military government and an ongoing military presence imply the end of occupation. What is troubling about this decision from a functional perspective is that the HCJ’s decision showed that Israel had in fact the power to affect the supply of electricty and fuel in Gaza. But by holding that Israel is no longer the occupying force and thus no longer bound by the duty of an occupier in respect to the provision of those goods, the HCJ subverted the accountability that should  flow from the exercise of power.

True, as I mention below, the HCJ’s dictum in this case did leave room for moving beyond the conceptual approach toward a more functional one.Yet, its ratio decidendi remained within the bounds of the conceptual approach, and shows  how transformation in the form of control may be used to absolve an occupier of responsibility, even in relation to cases where power is still exercised.

Adam Roberts wrote in 1984 that “there is no use worrying excessively about the definition of occupation. The core meaning of the term is obvious enough; but, as usually happens with abstract concepts, its frontiers are less clear.” This approach is typical and suggests that, in most cases (“core”), it is easy to determine whether occupation exists, but some indeterminacy may prevail at the margins (“frontier”). Contrary to this position, I argue that indeterminacy may rather be at the heart of the law of occupation and should actually be considered a “core” component rather than confined to its frontiers. This indeterminacy should be understood as reflecting the gap between the law’s attempt to distinguish occupation from sovereignty on the one hand, and the much more complex reality – not only in Gaza and Iraq but in many, if not most, post-1945 occupations –on the other. This indeterminacy also paves the way for using and abusing the international law of occupation to legitimize the illegitimate, partly through the creation of a “pick and choose” regime concerning the laws that apply in “borderline” situations.

Consider, for instance, the Israeli occupation of the Occupied Palestinian Territory (OPT). Over the years, this occupation has been partly transformed into a form of control that seeks to reduce friction between the armed forces of the occupying country and the occupied population. Indeed, traditional “face to face” interaction is often deliberately avoided.  This is evident in developments such as the Oslo Accords and the subsequent establishment of the Palestinian Authority, the West Bank Wall, the “privatization” of checkpoints, and the Gaza “Disengagement.” These changes point not only to the need to consider how occupation may take the form of what Yoram Dinstein calls “remote control,” but also to the limits of the binary approach. I think that we can already see some recognition of the need to develop a functional approach to occupation in existing international case law. In the famous judgment of the Nuremburg Tribunal in the Wilhelm List “Hostages Trial” (1949), one may find the locus classicus according to which an occupant’s power is as great as his responsibility (the latter being to preserve order, punish crime, and protect lives and property within the occupied territory). By contrast, the Ethiopia-Eritrea Claims Commission (EECC), especially in the Aerial Bombardment decision (2005), may be seen as shifting to the opposite approach. In this case, addressing the Ethiopian military presence in the conflict’s western front (a presence described as “transitory”), the Commission held that not all the obligations of the Fourth Geneva Convention dealing with occupied territories can be reasonably applied to an armed force anticipating combat and present in an area for only a few days. At the same time, the Commission emphasized that customary as well as Fourth Geneva Convention duties relating to the protection of enemy civilians and civilian property apply in areas controlled by a state’s armed forces, even when combat occurs. In determining that not “all the obligations” of an occupant can reasonably be applied in the circumstances, the Commission suggests a position that differentiates between obligations on the basis of the capacity and power exercised by the occupying power, rather than on a formalist on/off definition of occupation.  In other words, the EECC did not seek in this decision to determine abstractly whether occupation existed, and then derive duties from this determination. Instead, it held that where Ethiopia exercised power as an occupier, occupation exists and duties follow. This determination avoids the circularity described by Cohen by determining that Ethiopia’s duties derive from its actual exercise of power, not from reliance on abstract concepts of “occupation” or “sovereignty.” Thus, whereas the Hostages case held that the powers of an occupier are as great as its responsibility, the EECC took the position that responsibility actually follows from the exercise of power, implying that the responsibility of an occupier is as great as its power. Similar approaches may be detected in some judgments of the International Criminal Tribunal for Yugoslavia’s (ICTY)  (especially Naletilic in 2003) and even, as noted, in the Israeli Supreme Court’s Al-Bassiouni case concerning Gaza. The HCJ held that Israel does not have a general duty to ensure the welfare of Gaza Strip residents or to maintain public order in the Gaza Strip under international law according to the  law of belligerent occupation in its entirety. And yet, while holding that Israel lacks any effective ability to enforce order or manage civilian life in the Gaza Strip, the HCJ stopped short of determining that Israel does not owe Gaza  residents any duties. It cited three sources of duties: those that derive from (1) the state of armed conflict; (2) “the relationship that was created between Israel and the territory of the Gaza Strip after the years of Israeli military rule in the  territory, as a result of which the Gaza Strip is currently almost completely dependent upon the supply of electricity from Israel” (what may be called “post-occupation duties”); and (3) “the degree of control exercised by the State of Israel over the border crossings between it and the Gaza Strip.” The latter, even if the HCJ did not elaborate upon it in the judgment, joins the EECC and ICTY decisions as pointing to an emerging  judicial recognition of what I call the “functional approach” to the law of occupation. Let me emphasize that I consider this judgment to have been decided wrongly. The HCJ should have determined that, even if Israel does not have the effective ability to manage all aspects of life in Gaza, the degree of control and power it does exercise over electricity and fuel, as the case itself proves, means that it should maintain the responsibilities of an occupying force in regard to their provision.

My epistemological position is that the question of whether or not occupation exists is already often indeterminate. Moreover, this indeterminacy, as such, often becomes a form of control, as I elaborate in my forthcoming book.  Situations as diverse as post-World War II Germany, Northern Cyprus, and Western Sahara also point to the sovereignty /occupation indeterminacy. I argue that the functional approach can address this reality better than a binary approach, and avoid the situation whereby indeterminacy results in a ‘pick and choose’ approach in favor of accountability.  Taking as an example Israel’s duties as occupier, a functional approach would also have implied, however, that Israel’s responsibilities would derive not only from the powers it can and does exercise, but also from the consideration of what other parties are exercising power. For example, in the context of education, Israel would not be held accountable for the curriculum in a school run by the Palestinian Authority or Hamas over whose daily running Israel has no effective control. At the same time, Israel would be responsible as an occupier for the way its control of movement (of both people and goods, including school books and supplies) affects education. Thus, the functional approach may be used to create rather than circumvent accountability, while recognizing that some actions are beyond the occupier’s effective control  and that, in regard to some functions, we may not actually want it to intervene.

Arguably the functional approach may encourage occupiers to exercise fewer powers so as to avoid responsibility. If this means that Palestinians would have more control over their own lives, I do not consider this per se a problem. Clearly, however, the functional approach must not mean that occupiers are relieved of their duties when there is no one else exercising them, or when the occupier’s behavior prevents a legitimate sovereign from exercising them. Article 43 of the Hague Regulations remains instructive here. Similarly, the functional approach does not deny that several different entities may be considered responsible in some cases. Mostly,  it should not be interpreted in a way that strengthens the “pick and choose” approach. Rather, this approach aims to ensure that powers exercising control, even in scenarios that do not look like “classic” occupation, are prevented from avoiding responsibility  for their actions by denying, transforming, or relinquishing part of their control.

To conclude, in developing the functional approach I suggest abandoning the binary approach to occupation and shifting instead to the view holding that obligations should follow from the exercise of power and control,  confronting realities that cannot be understood solely through a limited binary polarity of sovereignty vs. occupation. This shift from a “conceptual” to a “functional” approach is imperative if the law of occupation is to create accountability rather than impunity, provide for the protection of people living under a foreign occupation, and take account of political and technological transformations.

Obviously, I do not suggest  that the functional approach answers all questions. Consider, for example, the hostilities between Fatah and Hamas in 2006-2007. Should we argue that it was Israel’s responsibility, under Article 43, to exercise force in order to restore “public order”? Or should we consider that events happening within Gaza were outside the “function” Israel was fulfilling and thus its intervention was not required, and was possibly even prohibited? I am happy to leave this question, which proves Cohen’s point that ethical questions cannot be answered in purely legal terms, with  readers and with my fellow bloggers.

Print Friendly, PDF & Email
Articles, International Human Rights Law, Middle East
Notify of
Gidon Shaviv

Some points I didn’t quite understand:

1. Effective control need not have any relationship to occupation. Let us assume a hypothetical situation where Israel never ever occupied Gaza – and yet still blockaded it, and controlled the borders. Would they still be ” responsible as an occupier for the way its control of movement (of both people and goods, including school books and supplies) affects education”?

2. Wouldn’t this approach encourgae the occupied not to assert control? Wouldn’t the Palestinians now have a good reason not to take responsibility on schools/generating electricity?

3. Can there never be a state where no one holds the responsibiity to perform a duty? 

David Prater
David Prater

Enlightening piece and questions.  Just as a lay response to the questions posed, here are my observations: 1) One of the principle elements of sovereignty is what Stephen Krasner called interdependence sovereignty, or the ability to control the flow of ideas and goods across a state’s borders.  If Israel displaces the government in Gaza in managing the flow of goods/people/ideas across borders, then as a ‘binary’ matter Israel would be occupying Gaza – as you note, that leads to the result of Israel being responsible for treating all aspects of life in Gaza under the laws of occupation.  Professor’s Goss’ point is that Israel’s obligations should only extend so far as they have control.  If there was some legal regime that covered such a situation – effective control over a territory but no actual presence – such a legal regime would reasonably apply.  However, there being so such regime, the law of occupation is by default what should apply.  But, it should be done on an issue-by-issue basis, to determine who is in control and therefore who is responsible for ensuring governmental functions. 2) No doubt that the government in Gaza uses Israel as a scapegoat for its own inability to govern, and… Read more »

Eliav Lieblich

Great post. I think the functional approach to occupation correlates with the functional approach to sovereignty as manifested in concepts like the responsibility to protect. One thing that is not clear to me: if exercising power “externally” over a territory – absent “boots on the ground” – can give rise to duties of the external power towards the affected population, doesn’t the result simply merge with extra-territorial application of human rights law? In the latter context, it is already accepted that economic sanctions, for instance, should be assessed according to human rights considerations.


[…] לכיבוש – רעיון שמפותח בספר שאני עובד על השלמתו. הפוסט הראשון בסימפוזיון הוא שלי בו אני מציג את הרעיון. החל ממחר יופיעו פוסטים נוספים […]

Noam W
Noam W

Great piece, really one of the best I have read in a long time on this subject.

I was wondering, vis-a-vis the chance that teh occupier will relinquish responsibilities. You suggest that if this means that the occupied population will have more freedom to act, that is a good thing. This works very well, I think, with negative rights. How would it work with positive rights that require affirmative action on the side of either the local power or the occupying power, if neither wants the responsibilty?

Aeyal Gross

Thanks to all for their thoughtful comments. I will do my best to address them. @ Gidon – your first question is addressed to a large extent in Sari Bashi’s post. Without addressing now the questions of the duties of a “blockading” party I think that as you say effective control may not have a relation to occupation, but then duties may be different. (Presumably duties in blockade are more limited, negative and have to do with the armed conflict situation). But I realize in a way this opens the question of why do we need the framework of occupation at all, and can we avoid the binary approach altogether (see Valentina Azarov’s post and my reply to it that will be posted tomorrow). For now I will stick with the analysis that occupation does require a foreign military displacing the legitimate government, (entering its shoes or preventing it from being able to operate properly) and that the special scope of duties derive from that. I think in order to answer your question further we would have to consider what duties does a “blockading” party have and where they are different from that of occupier and what different purposes does… Read more »