Treaties, States Parties and Commentaries
  • Print page
Commentary of 2020 
Article 91 : Successful escape
Text of the provision*
1. The escape of a prisoner of war shall be deemed to have succeeded when:
(1) he has joined the armed forces of the Power on which he depends, or those of an allied Power;
(2) he has left the territory under the control of the Detaining Power, or of an ally of the said Power;
(3) he has joined a ship flying the flag of the Power on which he depends, or of an allied Power, in the territorial waters of the Detaining Power, the said ship not being under the control of the last named Power.
2. Prisoners of war who have made good their escape in the sense of this Article and who are recaptured, shall not be liable to any punishment in respect of their previous escape.
* Paragraph numbers have been added for ease of reference.
Reservations or declarations
None
Contents

A. Introduction
3780  Escape gives rise to a dilemma between two contrasting ideas. The first is that attempts to escape may be viewed as a demonstration of patriotism and of the most honourable intentions, and not as a crime.[1] According to this precept, prisoners of war have a right, a moral duty, and sometimes, under the law of the Power on which they depend, even a legal obligation to escape.[2] The second idea is that an attempt to escape is a punishable breach of discipline vis-à-vis the Detaining Power, which it is under a strong obligation to prevent.[3] These contrasting ideas have been described as the ‘paradox of escape’, which is at the core of the legal framework on prisoner-of-war escapes.[4]
3781  During the Second World War, crimes committed against recaptured prisoners of war included torture, ill-treatment and even summary executions, as evidenced by subsequent war crimes prosecutions.[5] For instance, 14 German officials, acting under orders from Hitler, murdered 27 Allied air force personnel who had escaped from the Stalag Luft III camp in Sagan on 22 March 1944 and were recaptured a few days later.[6] After the war, the German officials were tried for war crimes by a British Military Court and sentenced to death.[7] With this context in mind, the drafters of the Third Convention felt the need to strengthen and further refine existing provisions on prisoner-of-war escapes.
3782  The legal framework regulating escapes and escape attempts consists of a series of articles (Articles 91–94) of the Third Convention. These provisions should therefore be read together. In essence, these rules state that successful escapes may not be punished (Article 91); unsuccessful escape attempts occasion disciplinary punishment only and possibly special surveillance as a preventive measure (Article 92); connected offences committed with the sole intention of facilitating escape and which do not entail any violence against life or limb occasion disciplinary punishment only (Article 93); and the Power on which the escapee depends must be notified of any recapture, provided notification of the escape has been made (Article 94). These provisions are complemented by other rules, such as Article 42, which addresses the possibility of using weapons to prevent an escape.
3783  The rules on prisoner-of-war escapes remain relevant today. In the ICRC’s experience of contemporary international armed conflicts, prisoners of war who escape or attempt to escape and are later recaptured are often in a particularly vulnerable position. They are notably at risk of being ill-treated; considered as criminals and inappropriately punished for their escape; subjected to excessive measures to prevent future escapes; transferred to special places of detention/military facilities outside prisoner-of-war camps; and/or denied ICRC visits. Collective punishment of prisoners of war, including of fellow prisoners who had not attempted to escape, has also been reported.[8]
Back to top
B. Historical background
3784  The principle that prisoners of war who successfully escape and are recaptured are not liable to punishment in respect of their escape is a longstanding rule of the law of war. The Lieber Code provided already in 1863 that ‘(i(f prisoners of war, having given no pledge nor made any promise on their honour, forcibly or otherwise escape, and are captured again in battle after having rejoined their own army, they shall not be punished for their escape, but shall be treated as simple prisoners of war’.[9] Similarly, the 1874 Brussels Declaration held that prisoners of war who are recaptured after succeeding in escaping are not liable to punishment for their previous acts.[10]
3785  The 1907 Hague Regulations and the 1929 Geneva Convention on Prisoners of War contained the same principle but specified in addition that it is only when prisoners of war succeed in rejoining their armed forces or in leaving the territory occupied by the armed forces which captured them that they are not liable to punishment if recaptured.[11]
3786  Taking into account comments by the ICRC, the Conference of Government Experts, convened in 1947 to review the 1929 Convention, further elaborated on the conditions under which an escape may be deemed successful, drawing a clear distinction between escape attempts, which may lead to disciplinary punishment, and successful escapes, which involve no punishment:
PW shall be considered as having successfully escaped:
1. on reaching neutral or non-belligerent territory, or territory not occupied, but under the authority of their own country or of an ally;
2. on rejoining their own armed forces or those of an allied Power;
3. on reaching the high seas;
4. on boarding, in the territorial waters of the DP [Detaining Power], a merchant vessel or warship flying the flag of their home country or of an allied Power, and not under the authority of the DP.[12]
3787  This text was redrafted at the 17th International Conference of the Red Cross in Stockholm in 1948[13] and adopted by the Diplomatic Conference in 1949 with several linguistic changes.
3788  This typology of successful escape was devised mainly on the basis of State practice.[14] For instance, during the Second World War, Germany considered that prisoners of war had successfully escaped only if they had reached a neutral territory or a territory Germany did not occupy.[15]
Back to top
C. Paragraph 1: The notion of successful escape
3789  The definition of successful escape adopted in Article 91(1) distinguishes three alternative – but exhaustive – situations that can individually amount to a successful escape; each of these is analysed separately below. Overall, for the purpose of Article 91(1), an escape is successful only if a prisoner of war has managed to escape from the control of the Detaining Power.[16] The mere fact of being able to leave the prisoner-of-war camp or to shake off immediate pursuers is not sufficient.[17] It is necessary for the escapee to join friendly forces, to leave the territory under the control of the adversary or its allies or to board a ship not under the control of the Detaining Power.[18]
Back to top
1. Subparagraph (1): Joining the ‘armed forces’
3790  The first situation in which the escape is to be considered successful is when the escapee has been able to join the armed forces of the Power on which they depend, or those of an allied Power. As soon as this is the case, the escapee is no longer a prisoner of war or a person hors de combat.[19]
3791  ‘Armed forces’ as referred to here are the subject of Article 4 of the Convention, which defines the conditions under which persons who fall into the power of the enemy are recognized as prisoners of war.
3792  Another possibility is that the escapee joins an allied Power. The term ‘allied Power’ is not defined in the 1949 Geneva Conventions. It should be understood as describing any co-belligerent State, whether or not a formal treaty of alliance exists, as opposed to a neutral Power.
3793  The verb ‘joined’ must be understood in a practical sense. It means physically reaching the armed forces of the Power on which the escapee depends or those of an allied Power, rather than being legally incorporated into those forces. The mere fact of reaching, for instance, the barracks of an allied Power, a small unit of friendly forces or a resistance group opposing the Detaining Power is sufficient.[20] However, prisoners of war who have merely joined up with others who have escaped cannot yet claim to have escaped from the enemy; the forces that they have succeeded in joining must also be outside the power of the enemy and beyond the reach of the opposing forces.[21]
3794  By contrast, the precise location of the friendly forces is immaterial for the purposes of this subparagraph. Thus, the escape is still successful even if the prisoner of war joins a unit that finds itself on the territory of the Detaining Power or of one of the Detaining Power’s allies or on a territory occupied by these enemy forces.
Back to top
2. Subparagraph (2): Leaving territory under the control of the Detaining Power and its allies
3795  Article 91(1), subparagraph 2, describes the second situation of successful escape. This is the case when the prisoner of war leaves the sovereign territory of the Detaining Power or of any of its allies, as well as any other territory that they occupy or otherwise control. In this case, the escapee finds themselves either: a) within any territory under the control of the Party on which they depend or of its allies, including territories occupied by them; or b) within the territory of a neutral, i.e. non-belligerent, State.
3796  Once a prisoner of war has left the territory under the control of the Detaining Power, or of an ally of that Power, the escape is successful even if shortly after that the former Detaining Power manages to extend its control over the area in which the former prisoner is located.
3797  The rights and duties of neutral Powers regarding escaped prisoners of war are set out in the 1907 Hague Convention (V).[22] According to Article 13(1) of that Convention, ‘[a] neutral Power which receives escaped prisoners of war shall leave them at liberty. If it allows them to remain in its territory it may assign them a place of residence.’ In essence, to fulfil their duty of neutrality, neutral Powers may not transfer the escaped prisoners of war either to the former Detaining Power or to the Party to which they belong, nor to their allies.[23] By contrast, the neutral Power is not obliged to actively impede the escapees from rejoining the Power on which they depend or its allies. They may assign them to a place of residence, but this is not an obligation, and the Hague Convention (V) does not elaborate on the situations when they may do so. It has been argued that assigning to a place of residence should be done only in exceptional cases under circumstances similar to those allowing the assignment of civilian refugees to a residence when they are unable to leave the territory of the neutral country.[24] In practice, neutral States have adopted different policies. During the First World War, for example, the Netherlands left escaped French prisoners of war free to return to their country, while Luxembourg assigned escaped prisoners of war to a place of residence.[25]
3798  It should be noted that if escaped prisoners of war decide not to rejoin their armed forces to continue fighting the enemy, they run the risk of being considered as ‘deserters’ by their State (and might be prosecuted under domestic law).[26] This does not, however, have any impact on the successful character of the escape under international humanitarian law, provided the escapees manage to quit the territory of the enemy. If recaptured by the enemy, they may still be detained as prisoners of war, although they may not be punished for the successful escape.[27]
Back to top
3. Subparagraph (3): Joining a ship in the territorial waters of the Detaining Power
3799  The third situation of successful escape is an exception to subparagraph 2, since, under international law, the territorial sea is considered to be part of the coastal State’s territory.[28] Even when escaped prisoners of war do not manage to leave the territory under the control of the Detaining Power because they find themselves in the territorial sea of the Detaining Power, the escape is nevertheless successful if they join therein a ship flying the flag of the Power on which they depend or of an allied Power.[29] Escape is successful in such a case provided that the said ship is not under the actual control of the Detaining Power, i.e. it has not been captured.[30] For the purpose of this subparagraph, it makes no difference whether the ship is a warship, an auxiliary vessel, a merchant ship, a fishing vessel or any other category of vessel.[31] Thus, even joining a civilian ship flying the flag of the Power on which the escapee depends, or of an allied Power is enough.
3800  ‘Joining a ship’ means physically getting on board.[32] This could be done by swimming and climbing aboard the friendly ship, by using a small craft to reach the friendly ship or by being rescued by a friendly ship. The wording used does not indicate that consent to an escapee’s presence on board would be needed.
3801  If a prisoner of war is in the territorial waters of the Detaining Power while on, for instance, a rowing boat, they cannot be considered as having successfully escaped yet, because they are still on the territory of the Detaining Power and are not under the protection of friendly forces.
3802  Subparagraph 3 does not include the situation in which the escaping prisoner of war boards a ship flying the flag of a neutral State in the territorial waters of the Detaining Power. Therefore, escape is not successful when a prisoner of war gets on board any category of ship flying the flag of a neutral Power in the territorial waters of the Detaining Power.
3803  Outside territorial waters, such as when an escaping prisoner of war reaches the contiguous zone, exclusive economic zone or the high seas,[33] the escape is to be considered successful as per subparagraph 2, since the escapee has left the territory under the control of the Detaining Power or of an ally of that Power. Admittedly, the text proposed by the government experts in 1947, which was formulated differently, provided for successful refuge on the high seas in a separate paragraph. It should be noted that according to the 1960 ICRC Commentary, ‘(a( prisoner of war on the high seas, for instance on a raft, without having been picked up by a ship, cannot be considered to have succeeded in escaping’.[34] However, this is not supported by a textual interpretation of Article 91 and leaves the question why the escape would be successful if the prisoner of war boards a ship outside the Detaining Power’s territorial sea but not if they are on a raft on their own? The ICRC considers that once a prisoner of war has reached the seaward side of the outer limit of the Detaining Power’s territorial sea – whether on a friendly or neutral ship or on a raft – the escape is successful since the escapee is outside of the territory under the control of the Detaining Power. This interpretation is more logical when Article 91 is read as a whole. It is also supported by some State practice.[35]
3804  Nevertheless, it should be recalled that, under the law of naval warfare, ‘hostile actions by naval forces may be conducted … on … the high seas’[36] and that the prisoner of war who has successfully escaped and is on the high seas is a combatant who may be attacked or recaptured.
3805  An important caveat for the purpose of interpreting subparagraph 3 is that the central notion of ‘territorial sea’ (or ‘territorial waters’) is not defined today as it was in 1949. In 1949, the territorial waters were generally understood to extend only to 3 nautical miles from a nation’s coastline (the so-called ‘three-mile limit’).[37] Nevertheless, in the twentieth century, a number of States started to expand their territorial waters.[38] The notion of ‘territorial waters’ was precisely defined in 1982 in the UN Convention on the Law of the Sea. In accordance with that Convention, the territorial sea extends at most 12 nautical miles from the baseline, which is usually the mean low-water mark of a coastal State.[39] At the time of publication (2020), the issue remains unresolved as there is no State practice to assess which of the two understandings of the breadth of the territorial sea applies for the purpose of Article 91(3).
3806  In accordance with general rules of treaty interpretation,[40] it is submitted that Article 91 and the notion of territorial waters must be interpreted in an updated manner, taking into account contemporary State practice and agreements on the law of the sea, such as the UN Convention (i.e. up to 12 nautical miles). Concretely, this means that an escaping prisoner of war who chooses to escape at sea will either have to reach a friendly or allied ship in the territorial sea or get hold of a raft, rowing boat or other vessel to get out of the territorial waters.[41] Although this interpretation of Article 91 minimizes the chances of successful escape at sea because the ‘distance to impunity’ becomes longer (12 nautical miles instead of 3), it would be legally unsound to freeze the notion of territorial waters as it was in 1949.[42] This updated interpretation is not necessarily to the detriment of prisoners of war as it also implies that the escaping persons, as long as they are hors de combat, remain immune from attacks under the conduct of hostilities until their escape is successful, i.e. until they reach 12 nautical miles.
3807  It should be noted that Article 91 does not govern the scenario of an escaping prisoner of war who boards an aircraft registered to the Power on which they depend or belonging to an allied Power in the airspace of the Detaining Power, neither of the said aircraft being under the control of the Detaining Power. This is apparently an issue that the drafters did not consider specifically. Accordingly, it would be inappropriate to draw analogies with the situation on board a ship because the legal frameworks applicable at sea and in the air differ substantially. A reasonable conclusion is, therefore, that the rule under subparagraph 2 applies and that the escaping prisoner of war is not successful until they leave the airspace above the territory under the control of the Detaining Power or of an ally of that Power.
Back to top
D. Paragraph 2: The privilege of impunity
3808  The legal consequence of a successful escape as defined in Article 91(1) is that, if the escaped prisoner is recaptured, they may not be subject to criminal or disciplinary sanctions for that escape. This rule does not, however, grant impunity for offences committed during the escape.[43]
3809  The privilege of impunity provided for in Article 91(2) derives from the fact that prisoners of war commit no offence against international law by attempting an escape.[44] They may thus legitimately try to escape from their captors. Arguably, this ‘right’[45] indirectly derives from the privilege of combatancy.[46] This explains why, unlike prisoners of war, civilian internees who are recaptured after a successful escape are liable to disciplinary punishment.[47]
Back to top
Select bibliography
Bretonnière, Maurice, L’application de la Convention de Genève aux prisonniers français en Allemagne durant la seconde guerre mondiale (typewritten thesis), Paris, 1949.
Frey, Hans Karl, Die disziplinarische und gerichtliche Bestrafung von Kriegsgefangenen (thesis), Springer, Vienna, 1948.
Krähenmann, Sandra, ‘Protection of Prisoners in Armed Conflict’, in Dieter Fleck (ed.), The Handbook of International Humanitarian Law, 3rd edition, Oxford University Press, 2013, pp. 359–411.
Maia, Catherine, Kolb, Robert and Scalia, Damien, La Protection des Prisonniers de Guerre en Droit International Humanitaire, Bruylant, Brussels, 2015.
Martin, Paul E., ‘Note sur les prisonniers de guerre évadés sur le territoire d’une Puissance neutre’, Revue international de la Croix-Rouge, Vol. 26, No. 301, January 1944, pp. 62–69.
Scheidl, Franz, Die Kriegsgefangenschaft von den ältesten Zeiten bis zur Gegenwart, Ebering, Berlin, 1943.

1 - See Frey, p. 42.
2 - See e.g. United Kingdom, Manual of the Law of Armed Conflict, 2004, p. 187, para. 8.116.1; United States, Code of Conduct for Members of the US Armed Forces, 1955, Article III; United States, Law of War Manual, 2016, p. 618–619, para. 9.26.6. See also France, LOAC Manual, 2012, pp. 52–53; Philippines, LOAC Teaching File, 2006, p. 21 (referring to the existence of such a duty in some States); and Switzerland, Regulation on Legal Bases for Conduct during an Engagement, 2005, p. 31 (referring to the existence of such a duty in some States).
3 - See Werner, p. 326, and Rowe, p. 1034.
4 - See Werner, p. 326. See also Bretonnière, pp. 335–336, and Levie, p. 403.
5 - Maia/Kolb/Scalia, p. 440. See also United Kingdom, Military Court at Singapore, Sasa case, Reference of proceedings, 1948, para. 3.
6 - United Kingdom, Military Court at Hamburg, Stalag Luft III case, Judgment, 1947, p. 31. See also United States, Military Tribunal at Nuremberg, The Ministries case, Judgment, 1949, p. 308, and Von Leeb case, Judgment, 1948, p. 395.
7 - United Kingdom, Military Court at Hamburg, Stalag Luft III case, Judgment, 1947, p. 31.
8 - See e.g. UN Security Council, Prisoners of war in Iran and Iraq: The report of a mission dispatched by the Secretary-General, January 1985, UN Doc. S/16962, 22 February 1985.
9 - Lieber Code (1863), Article 78. During the Second World War, prisoners of war were punished for not following an order by the Detaining Power to swear on their honour that they would ‘not in any circumstances attempt to escape’; see United Kingdom, Military Court at Singapore, Fukuei case, Abstract of evidence, 1946, pp. 1–2.
10 - Brussels Declaration (1874), Article 28.
11 - Hague Regulations (1907), Article 8, para. 2, a contrario; Geneva Convention on Prisoners of War (1929), Article 50.
12 - Report of the Conference of Government Experts of 1947, pp. 211–212.
13 - Draft Conventions adopted by the 1948 Stockholm Conference, draft article 81, p. 84.
14 - Report of the Conference of Government Experts of 1947, p. 212.
15 - Maia/Kolb/Scalia, p. 436; Bretonnière, p. 359.
16 - Krähenmann, p. 404.
17 - See Scheidl, p. 449, and Pictet (ed.), Commentary on the Third Geneva Convention, ICRC, 1960, p. 446.
18 - According to one view, a different test may apply for the purpose of Article 42, which regulates the different topic of use of weapons against prisoners of war; see the commentary on Article 42, paras 2552–2554.
19 - See e.g. Switzerland, Regulation on Legal Bases for Conduct during an Engagement, 2005, p. 31.
20 - See e.g. Belgium, Law of Armed Conflict Training Manual, 2009, Part IV, p. 22: ‘Une évasion doit être considérée comme réussie lorsque l’évadé a pu, même momentanément, rejoindre ses forces armées ou quitter le territoire effectivement contrôlé par l’ennemi.’ (‘An escape must be considered successful when the escapee has been able, even for a brief moment, to rejoin their armed forces or to leave the territory effectively controlled by the enemy.’)
21 - As a result, if escaping prisoners of war join a unit of their own armed forces which is about to be captured, it is not a successful escape, ‘as there is no actual interruption, not even a brief moment of being beyond the detaining power’s reach’; Krähenmann, pp. 404–405.
22 - See, further, Maia/Kolb/Scalia, p. 438. With regard to the obligations of the neutral State on whose territory an escaped prisoner of war has sought refuge, see also Martin, pp. 62–69.
23 - Maia/Kolb/Scalia, p. 438; Paul Fauchille, Traité de droit international public, Tome II: Guerre et neutralité, Rousseau, Paris, 1921, pp. 684–685.
24 - Maia/Kolb/Scalia, p. 438; Martin, p. 68. 

25 - Martin, p. 68.
26 - On the status of deserters, see the commentary on Article 4, section D.4.
27 - Article 91(2).
28 - Sarah Wolf, ‘Territorial Sea’, version of August 2013, in Rüdiger Wolfrum (ed.), Max Planck Encyclopedia of Public International Law, Oxford University Press, para. 2, http://www.mpepil.com. See also UN Convention on the Law of the Sea (1982), Article 2(1) (providing that the sovereignty of a coastal State extends to the territorial sea).
29 - See UN Convention on the Law of the Sea (1982), Article 91(1): ‘Ships have the nationality of the State whose flag they are entitled to fly. There must exist a genuine link between the State and the ship.’
30 - Under the law of naval warfare, with some exceptions of categories of vessels which are protected, a Party to an international armed conflict is entitled to capture enemy vessels, in particular within its own territorial sea. For details, see San Remo Manual on International Law Applicable to Armed Conflicts at Sea (1994), paras 135–140.
31 - UN Convention on the Law of the Sea (1982), Article 29, which reflects customary international law on this point defines ‘warship’ as a ‘ship belonging to the armed forces of a State bearing the external marks distinguishing such ships of its nationality, under the command of an officer duly commissioned by the government of the State and whose name appears in the appropriate service list or its equivalent, and manned by a crew which is under regular armed forces discipline’. See also UN Convention on the High Seas (1958), Article 8(2); Hague Convention VII (1907), Articles 2–4; and San Remo Manual on International Law Applicable to Armed Conflicts at Sea (1994), para. 13(g). The San Remo Manual, para. 13(h) and (i) respectively, defines an ‘auxiliary vessel’ as a ‘vessel, other than a warship [i.e. non-commissioned], that is owned or under the exclusive control of the armed forces of a State and used for the time being on government non-commercial service’ and a ‘merchant vessel’ as a ‘vessel, other than a warship, an auxiliary vessel, or a State vessel such as a customs or police vessel, that is engaged in commercial or private service’. See also the commentary on Article 14 of the Second Convention, para. 1520.
32 - See Report of the Conference of Government Experts of 1947, pp. 211–212, which referred to ‘boarding’ the said ship. For the full draft of the provision, see para. 3786.
33 - Article 86 of the 1982 UN Convention on the Law of the Sea states that the high seas comport ‘all parts of the sea that are not included in the exclusive economic zone, in the territorial sea or in the internal waters of a State, or in the archipelagic waters of an archipelagic State’.
34 - Pictet (ed.), Commentary on the Third Geneva Convention, ICRC, 1960, p. 447. See also, in this sense, Scheidl, p. 449, and Krähenmann, p. 404. See also Horst Fischer, ‘Protection of Prisoners of War’, in Dieter Fleck (ed.), The Handbook of Humanitarian Law in Armed Conflict, Oxford University Press, 1995, pp. 321–367, at 361, para. 727.
35 - United States, Law of War Manual, 2016, p. 613, para. 9.25.1.1: ‘[I]f the POW reaches neutral territory or the high seas, he or she will have escaped successfully.’ The United Kingdom, Military Manual, 1958, para. 238, note 3, supported this interpretation: The effect of Art. 91 [of the Third Convention] is that the escape is to be considered successful not only if the prisoner rejoins the armed forces of the State with which he was previously serving, or those of its allies, but also if he has left the territory of or occupied by the Detaining Power, e.g., by reaching neutral territory. If he were picked up by a ship of the Detaining Power outside its territorial waters it would seem that he would not be liable to disciplinary punishment having already made a successful escape by leaving its territory. However, the United Kingdom, Manual of the Law of Armed Conflict, 2004, p. 184, para. 8.109, is more cautious, stating that the escape of a prisoner of war who ‘joins a vessel belonging to his own, an allied, or neutral state beyond the territorial sea of the detaining power … will be deemed successful. The position of an escaper who leaves that territorial sea under his own auspices, eg, on a raft, is unclear.’
36 - San Remo Manual on International Law Applicable to Armed Conflicts at Sea (1994), para. 10(b).
37 - The three-mile limit was allegedly determined in accordance with the ‘cannon-shot rule’ developed by Cornelius van Bynkershoek, De Dominio Maris Dissertatio, Oxford University Press, 1923, p. 44. See also Kinji Akashi, Cornelius Van Bynkershoek: His Role in the History of International Law, Martinus Nijhoff Publishers, Leiden, 1998, p. 150.
38 - For a detailed history, see Kai Trümpler, ‘Article 3 – Breadth of the Territorial Sea’, in Alexander Proelss (ed.), United Nations Convention on the Law of the Sea: A Commentary, C. H. Beck/Hart/Nomos, 2017, pp. 34–42, at 36–39.
39 - UN Convention on the Law of the Sea (1982), Part II, Section 2. See also ICJ, Territorial and Maritime Dispute (Nicaragua v. Colombia), Judgment, 2012, para. 177: ‘Whatever the position might have been in the past, international law today sets the breadth of the territorial sea which the coastal State has the right to establish at 12 nautical miles. Article 3 of UNCLOS [UN Convention on the Law of the Sea] reflects the current state of customary international law on this point.’
40 - See Vienna Convention on the Law of Treaties (1969), Article 31(3)(a)(b) and (c). The ordinary meaning of ‘territorial waters’ today is understood as extending at most 12 nautical miles from the baseline. Moreover, a treaty has to be interpreted in light of relevant State practice and rules of international law applicable in the relations between the Parties (such as the 1982 UN Convention on the Law of the Sea and relevant customary norms).
41 - Regarding escape on the high seas, see para. 3803 of this commentary.
42 - For a different interpretation, see Liesbeth Lijnzaad, ‘Swimming Into Deep Water: Article 91(3) of the Third Geneva Convention and the Successful Escape of Prisoners of War by Sea’, in Stephan Marquardt and Steven Blockmans (eds), The European Union’s Contribution to International Peace and Security: A Legal Practitioners’ Perspective, Brill, forthcoming 2021, arguing that adopting the contemporary size of the territorial sea in this provision implies increased efforts are required to escape and that this interpretation is consequently contrary to the object and purpose of the Third Convention.
43 - See the commentary on Article 93, section C.
44 - See Lieber Code (1863), Article 77. See also e.g. Canada, LOAC Manual, 2001, p. 10-10, para. 1039(20), and Nigeria, IHL Manual, 1994, para. 41.
45 - See e.g. Belgium, Law of Armed Conflict Training Manual, 2009, Part IV, p. 20; France, LOAC Manual, 2012, pp. 52–53; and Israel, Manual on the Rules of Warfare, 2006, p. 34.
46 - See Hague Regulations (1907), Article 1, and Additional Protocol I, Article 43(2). The combatant’s privilege involves the right to participate directly in hostilities and therefore to immunity from prosecution for lawful acts of war.
47 - Fourth Convention, Article 120. On the rationale for disciplinary punishment for unsuccessful escape, see the commentary on Article 92, section C.2.