When Blockade Goes Global: The Sumud Flotilla Interceptions and the Legality of Israel’s Gaza Blockade

In the early hours of 30 April 2026, Israeli naval forces boarded 22 vessels of the Global Sumud Flotilla near Crete, roughly 600 nautical miles from Gaza and apprehending approximately 175 civilian activists. A few weeks later, on 19 May 2026, the Israeli Navy intercepted a second wave of 54 vessels about 70 nautical miles west of Cyprus and roughly 250 nautical miles from Gaza, detaining almost 430 activists. Both operations were defended as enforcement of Israel’s naval blockade of Gaza, originally declared in January 2009. They also represent a significant escalation of Israel’s earlier 2025 enforcement actions in the Eastern Mediterranean, namely the June 2025 Madleen incident and the October 2025 Sumud Flotilla interception, which occurred respectively at around 70 and 42 nautical miles from Gaza.

Following their apprehension, many activists were reportedly mistreated, with Israeli National Security Minister Itamar Ben-Gvir publishing footage of activists kneeling with their hands tied. The images prompted significant responses from a wide range of governments including France, the Netherlands, Canada, Belgium, Germany and the United Kingdom. What these statements conspicuously avoided, however, was any characterization of the interceptions themselves as breaches of international law. Some, like Italy, gestured implicitly at the geographic dimension by emphasizing that the operations had taken place in international waters. The European Commission, while having earlier advised the Flotilla against sailing to Gaza, reiterated that freedom of navigation under international law had to be upheld, and called on Israel to respect international humanitarian law. It stopped short, however, of characterizing the interceptions themselves as unlawful. A group of States led by Spain and Türkiye went further, condemning the Israeli operations as “flagrant violations of international law and international humanitarian law” (here and here), while Poland characterized the “[f]orceful activity in international waters” as “a serious violation of international law, including the law of the sea.”

This post takes up the question that the diplomatic record has conspicuously declined to address: whether the interceptions near Crete and Cyprus – undertaken beyond the territorial sea several hundred nautical miles from Gaza on the asserted authority of Israel’s 2009 naval blockade declaration – were lawful under international humanitarian law. In doing so, this post shall not address questions related to international law of the sea, jus ad bellum, allegations of genocide or the legal treatment of the detained activists.

The law of blockade

First, it is worth briefly setting out the law of blockade as it stands today under international humanitarian law. Although the customary international law of naval blockade emerged from centuries of belligerent practice, its first authoritative codification came with the 1856 Paris Declaration Respecting Maritime Law, which required blockades to be effective, i.e. maintained by a force sufficient “to prevent access to the coast of the enemy.” Subsequent efforts at codification followed, notably the unsuccessful 1909 London Declaration Concerning the Laws of Naval War and the 1913 Oxford Manual of Naval Warfare. Today, the most authoritative restatements are in paragraphs 93 to 104 of the 1994 San Remo Manual on International Law Applicable to Armed Conflicts at Sea as well as in chapter 7.4 of the Newport Manual on the Law of Naval Warfare. Further support can be found in national military manuals.

A lawful blockade has five cumulative conditions: (1) It must be declared and notified to belligerents and neutrals; (2) It must be effective rather than nominal; (3) It must be impartially applied to vessels of all states; (4) It must not have starvation or denial of essential goods to the civilian population as its sole or principal purpose and must allow the passage of humanitarian relief subject to inspection (whether a blockade must in addition satisfy a proportionality test is contested and is addressed below) and;      (5) it must not bar access to neutral ports or coastlines. In casu, most of these are not seriously contested.

Two further features of a legal blockade matter for what follows. First, blockade is generally considered as a method of warfare reserved for international armed conflicts (IAC) only. The classic counter-example – Lincoln’s 1861 blockade of Confederate ports during the US Civil War – is rather the exception that proves the rule as the Union’s resort to blockade was understood as an implicit recognition of belligerency that brought IAC rules into play. Second, the law of occupation is itself a body of IAC IHL. Under Common Article 2 (2) of the Geneva Conventions, the latter apply to any “partial or total occupation of the territory of a High Contracting Party, even if the said occupation meets with no armed resistance”. The law of occupation therefore engages on its own terms, independently of how the active hostilities are classified.

The inception of the Gaza blockade and the Mavi Marmara

Israel imposed its naval blockade on Gaza on 3 January 2009, during Operation Cast Lead. The blockade was publicly declared, accompanied by a Notice to Mariners, and explained as a measure against Hamas’ weapons smuggling. It was controversial from the outset and – like today – was challenged by several humanitarian initiatives. On 31 May 2010 at approximately 70 nautical miles from Gaza, Israeli naval forces boarded a flotilla of six vessels carrying humanitarian aid, which had declared its intent to break the blockade a year before. During the boarding, nine activists were killed, and a dozen injured, prompting a major diplomatic crisis between Israel and Türkiye and four overlapping international inquiries.

The Israeli Turkel Commission concluded that the ongoing conflict was international in nature, that Gaza was no longer under occupation following the 2005 disengagement, and that the blockade was therefore lawful. The Turkish National Commission of Inquiry Report on the other hand concluded that the blockade was unlawful as Gaza remained under Israeli occupation by virtue of indirect effective control and that a State could not by definition blockade its own territory. The UN Human Rights Council Fact-Finding Mission on the Mavi Marmara incident reached a similar position, based on the earlier Goldstone Report’s finding that Israel remained the occupying power in Gaza. Finally, the UN Secretary-General’s Panel of Inquiry of September 2011 – the Palmer Report – adopted the Turkel Commission’s reading on occupation status and concluded that the blockade was lawful as a method of warfare, but criticised the boarding at approximately 70 nm from the blockade zone as “excessive” and the use of force as disproportionate.

The extensive scholarly debate of the time mapped the same fault lines, all of them turning on the contested status of Gaza, which the ICRC has consistently treated as remaining under Israeli occupation. On the sceptical side, Kevin Jon Heller argued that if Gaza was not occupied, the conflict could not plausibly be characterised as an IAC and the blockade therefore lacked underlying authority; Russell Buchan, Douglas Guilfoyle, and Andrew Sanger reached parallel conclusions by different routes, with Buchan and Guilfoyle adding that the humanitarian crisis in Gaza made the blockade unlawful in any event. James Kraska and Wolff Heintschel von Heinegg, defended the blockade’s lawfulness on the basis that the conflict was properly characterised as international. Finally, Yuval Shany took a nuanced position, accepting that the broader closure regime was likely unlawful but arguing that the naval blockade could be analysed separately.

Since the Mavi Marmara incident, however, two key events have taken place. First, the ICJ Advisory Opinion of 19 July 2024 on the Legal Consequences of Israeli Policies and Practices in the OPT, and the Advisory Opinion of 22 October 2025 on Israel’s obligations vis-à-vis the United Nations in the OPT, have further recalibrated the legal terrain on the occupation question. Second, Israel’s full-scale military operation in Gaza following the 2023 Hamas attack and its control over large portions of Gaza, have rendered any defence of the “no occupation since 2005” position increasingly untenable.

The incompatibility of occupation and blockade?

First, it must be determined whether a blockade of Gaza is possible under international law at all. Even granting that Israel’s occupation of Gaza triggers the IAC framework under Common Article 2(2), Mustafa Tuncer has recently repeated the argument that the law of blockade and the law of occupation are mutually exclusive within that very framework. Indeed, every authoritative source seemingly defines blockade as a method of warfare directed against an enemy’s coast. The 1909 London Declaration’s article 1 restricted blockades to “ports and coasts belonging to or occupied by the enemy.” The 1913 Oxford Manual’s article 30 used a similar formulation, as do the 2006 Australian Manual (§ 6.60), the 2013 German Manual (§ 1060) and the 2022 US Commander’s Handbook (§ 7.7.1). Tuncer reads this textual consistency as reflecting a structural truth. A state cannot blockade what it occupies, because occupation makes the territory its own administrative responsibility and Gaza’s coast is therefore no longer Israel’s enemy’s coast. In many ways this echoes the Turkish Mavi Marmara Report from 2010, which had already concluded that “a State cannot, by definition, blockade the borders of territory it occupies.”

The coexistence view and the functional approach to occupation

The incompatibility view, though at first sight textually grounded, has been resisted in scholarship reading the same texts in the opposite direction. The disjunctive formula in the 1909 London Declaration’s article 1 – “ports and coasts belonging to or occupied by the enemy” – can also be read as identifying two independent grounds on which a coast becomes blockade-eligible. The Newport Manual at para. 7.4.1 has further widened that formula, defining blockade as directed at “specified coastal areas that are under the sovereignty, occupation, or control of an enemy.” On this reading, sovereignty, occupation, and control are alternative rather than excluding qualifications: even if a territory is no longer effectively controlled by the enemy, blockade remains available so long as its coast still belongs to the enemy in sovereignty terms. Marco Longobardo therefore criticised the Turkish Mavi Marmara Report for asserting strict incompatibility “without … a sound argument”, and argued that the law of blockade and the law of occupation are “distinct legal regimes which could … coexist during a naval blockade of an occupied coastline.” Shany, commenting on the UN HRC Fact-Finding Mission report on the Mavi Marmara incident, accepted that occupation already permits the occupier to control access to occupied territory, but argued that blockade extends enforcement beyond the territorial sea subject to more stringent conditions. Finally, Heintschel von Heinegg took the position that occupation triggers the IAC framework under Common Article 2 GC, which in turn makes blockade legally available against occupied territory. ” 

As to whether Gaza is or remains occupied, the author now turns to the ICJ’s 2024 and 2025 advisory opinions. The Advisory Opinion of 19 July 2024 adopted in para. 92 what Aeyal Gross has characterised as the “functional approach” – now mainstreamed in the ICRC’s updated Common Article 2 commentary. Put plainly, the functional approach replaces a binary “occupied / not occupied” status question with a sliding scale of obligations calibrated to the degree of effective control exercised. Thus, residual occupation duties may persist after physical military withdrawal. Applied to Gaza in paras. 93-94, the Court stated that Israel “remained capable of exercising, and continued to exercise, certain key elements of authority over the Gaza Strip … despite the withdrawal of its military presence in 2005.” Crucially, Judge Iwasawa’s separate opinion observed that “the Court does not take a position as to whether Gaza remained ‘occupied’ within the meaning of the law of occupation after 2005.” Marko Milanović believed this was deliberate as the AO was “carefully crafted to retain a certain degree of ambiguity as to the status of Gaza, presumably in order to enable consensus on the bench.” The Advisory Opinion of 22 October 2025 built on this, with the Court concluding that Israel’s effective control over the Gaza Strip had “increased significantly” since 7 October 2023, so that “Israel’s obligations under the law of occupation have also increased significantly, commensurate with the increase in its effective control over the territory” (para. 86). Sean Watts called Israel a “Quasi-Occupying Power” in Gaza while Milanović’s reading of the 2025 AO is that the deliberate ambiguity on the occupation of Gaza does still persist, but that it now anchors Israel in a concrete affirmative duty to agree to and facilitate humanitarian relief under Article 59 of the Fourth Geneva Convention.

Therefore, under the coexistence view, occupation and blockade are distinct regimes that can apply in parallel: the Court’s finding that Israel exercises occupation-level effective control over Gaza engages Common Article 2(2) GC, triggers the international-armed-conflict framework, and brings blockade within the methods of warfare available to Israel. Even on the stricter incompatibility view, the conclusion holds. The Court adopted the functional approach precisely to avoid declaring Gaza “occupied” in the binary Art. 42 Hague Regulations sense, and Hamas still controls parts of Gaza, including portions of its coast. The blockaded territory is therefore not only what is occupied by Israel but also what is left under enemy control. The Israeli blockade is therefore, in principle, available.

The Humanitarian dimension

Even if blockade is in principle available as a method of warfare, that does not automatically make this blockade legal. On the humanitarian side, the two leading restatements diverge. Para. 102 of the San Remo Manual prohibits a blockade whose sole purpose is to starve the civilian population, or which causes disproportionate damage to civilians relative to the anticipated military advantage. While the Newport Manual accepts the former, it rejects the customary status of the proportionality test (§ 7.4.5) on the ground that blockade is a method of warfare rather than an attack within the meaning of Article 51(5)(b) AP I. Interestingly, most military manuals – including the US one – follow the San Remo Manual in substance. Finally, there remain a number of customary IHL obligations: the starvation prohibition (Article 54 AP I, Rule 53 of the ICRC Customary IHL Study), the humanitarian-relief obligations under Article 23 GC IV and Articles 70-71 AP I, and the Article 59 GC IV duty to agree to and facilitate relief in occupied territory.

In casu, on the San Remo reading, the blockade undoubtedly fails the proportionality test once one weighs the IPC’s August 2025 famine confirmation, Israel’s complete blockage of aid from 2 March to 18 May 2025, the expulsion of UNRWA, the ICJ’s provisional-measures Orders in South Africa v Israel, and the ICC’s November 2024 arrest warrants for the war crime of starvation. Even on a Newport reading, according to which “a denial of the blockading power’s consent to relief consignments will be arbitrary only in exceptional circumstances in which the very survival of the civilian population is at stake …”, the abovementioned facts make it difficult than to conclude that the current blockade is indeed unlawful.

When blockade goes global: geographic limits to enforcement?

Beyond the humanitarian conditions, and even if one accepts a very strict interpretation to the “sole purpose” test above, the geographic-discipline question poses additional challenges for the Sumud Flotilla interceptions. James Farrant had characterised the earlier Mavi Marmara’s boarding at approximately 70 nautical miles as “over-zealous” while Sondre Torp Helmersen went so far as to argue that blockades “cannot be enforced outside territorial waters”. Originally, the 1909 London Declaration in arts. 17-20 permitted lawful enforcement only within the “area of operations” of the blockading force. Today’s restatements are silent on any geographic limits to enforcement, however. The Newport Manual treats intent to breach as the operative trigger for interception (§ 7.4.7) without specifying any upper limit on the enforcement radius, while para. 96 of the San Remo Manual likewise leaves the distance to be determined by military requirements. While this intent-based trigger might be one of the strongest arguments open to Israel in the current case – and the Palmer Report treated precisely such declared intent as bearing on where the Mavi Marmara could lawfully be intercepted – intent and distance answer two distinct questions: which vessels may be stopped, and from how far.

In the absence of any textual provision, state practice points firmly in one direction and every blockade or blockade-equivalent operation since 1945 has conformed to some form of geographic discipline in terms of enforcement distance: The 1950-53 Wonsan blockade was enforced within roughly 60 nautical miles; the 1966-75 Beira Patrol operated 20-45 miles from Beira; the Indo-Pakistani war of 1971 confined interception to the East Pakistan and Karachi coasts; Operation Pocket Money against Haiphong harbour in 1972 was a mining of harbour waters; and finally the 2006 Israel–Lebanon blockade always remained operationally close to the Lebanese coast. Operations conducted under separate legal headings such as the French interdiction against the FLN (1956-62), the Iran–Iraq exclusion zones during the tanker war (1981-88), Sri Lankan operations against the LTTE, and the Saudi-led operations against Yemen since 2015, were never characterised as formal blockades. Even the WWI British “distant blockade” of Germany during the First World War was never a formal blockade, but operated via expanding contraband lists and reprisals. The same is true for the 1962 Cuban quarantine. Finally, an interesting comparison can be made with current US operations against Iran. The CENTCOM declaration of 13 April 2026 limited blockade enforcement to “the Gulf of Oman and Arabian Sea east of the Strait of Hormuz” – a clear geographic limit. When CENTCOM expanded visit-and-search worldwide on 16 April, as Mark Nevitt has clarified, it made sure to do so under the regime of contraband, not blockade. There is, in short, no precedent in eighty years of post-WWII state practice for blockade enforcement at the distances of the Sumud Flotilla operations, which are more akin to a quasi-global visit-and-search regime than blockade enforcement.

Conclusion

Three conclusions can be made in relation to Israel’s blockade and the Crete and Cyprus interceptions. On the occupation question, the ICJ’s functional approach leaves blockade in principle available under both the coexistence and incompatibility views. On the humanitarian conditions, the August 2025 IPC famine confirmation, UNRWA’s expulsion, the ICJ’s provisional measures, and the ICC’s starvation warrants not only fail San Remo’s proportionality test but may even reach Newport’s high “arbitrary denial” threshold. On geographic discipline, eighty years of post-1945 state practice supplies no precedent for enforcement at 250 nautical miles, let alone 600 nautical miles. To this author, the Crete and Cyprus operations cannot be considered as blockade enforcement. When blockade goes global, it ceases to be a blockade at all.