Extra-Territorial Asylum PDF
Extra-Territorial Asylum PDF
law
Traditionally, before the advent of human rights law, legal issues arising from
extraterritorial asylum were predominantly addressed in the context of
‘diplomatic asylum’, a term which refers to asylum in embassies or other
premises of a state located in the territory of another state.1 Legal discourse
on diplomatic asylum chiefly focused on the potential friction arising out of
grants of extraterritorial asylum between the state granting asylum and the
territorial state. Because extraterritorial asylum may constitute an affront to
the territorial sovereignty of the other state, it was seen to give rise to questions
of legitimacy under international law.
Both the maturation of human rights law and current policies of relocating
migration management warrant a legal restatement of the concept of extraterri-
torial asylum. Firstly, the various manifestations of pre-border migration
management question the extent to which existing discourse on diplomatic
asylum can be extrapolated to a more general theory on the legality of extra-
territorial asylum. Secondly, the present-day importance of human rights,
including the acceptance that human rights obligations may bind a state when
it is active in a foreign territory, require a determination of whether there can
be circumstances under which the petitioned state is under a human rights
obligation, vis-à-vis an individual, to grant protection and how such an obliga-
tion can be accommodated with possible concurrent and conflicting obligations
the petitioned state may have vis-à-vis the territorial state. In extraterritorial
situations, the scope of these protection duties is informed not only by the
duty of non-refoulement, but also involves the preliminary issue of whether
and under what circumstances the asylum-seeker should be granted the right
to physically bring himself within the territorial jurisdiction of the desired
state, for example by allowing him to present himself at the border of that
state. This is often referred to as the right to seek asylum, understood as the
right to relieve oneself from the authority of one country in order to be able
to request territorial asylum with the authorities of another.
The right to grant asylum, explored in section 4.2., is understood as the right
of the non-territorial state, vis-à-vis the territorial state, to confer asylum upon
an individual situated in the latter state. Although the relationship between
the state granting asylum and the state whose national is granted asylum is
currently scarcely addressed in international refugee law discourse – and for
a large part considered immaterial as a consequence of the principle of terri-
torial sovereignty coming to prevail2 – it remains of primordial importance
in extraterritorial situations, precisely because those situations are characterised
by the impossibility of the state addressed by the asylum-seeker to invoke
the shield of territorial sovereignty. Section 4.2. explores the extent to which
international law has recognised the institution of diplomatic and other forms
of extraterritorial asylum, how the institution of extraterritorial asylum involves
a reconciliation of potential conflicting claims of humanitarianism and terri-
torial sovereignty, and how the law on diplomatic and consular relations may
influence the legality and/or feasibility of grants of diplomatic asylum. It
should be noted here that this section deals only with grants of asylum within
the territory of another state. The other typical situation of extraterritorial asylum,
namely at sea, is addressed in chapter 6, which discusses questions of com-
peting state competences in the specific context of the Law of the Sea.
Under the right to obtain asylum, in section 4.3, it is examined under what
circumstances individuals have a right to obtain asylum from the non-territorial
state. This question concerns the right of asylum in its modern (human rights)
understanding: under what conditions can an individual claim entitlement
to protection? In situations where an individual requests protection from
another state than the one in which he is, a topical issue is whether inter-
national obligations protecting against refoulement have equal bearing in
territorial and extraterritorial situations and what the nature of the relationship
between the petitioned state and the individual must be to enliven human
rights obligations on the side of the former. This exercise mainly constitutes
2 This has now been confirmed in Article 1 of the United Nations General Assembly Declara-
tion on Territorial Asylum, see n. 6 infra and accompanying text.
Extraterritorial asylum under international law 115
The distinction between territorial and extraterritorial asylum has long standing
in international law. The notion of territorial asylum was traditionally under-
stood as the right of states to grant asylum to aliens on their territory, which
may be asserted vis-à-vis the pursuing state.3 In this vein, the right to grant
asylum has often been linked to the right to refuse extradition.4 In the Asylum
Case, the International Court of Justice equated the right of a state not to
extradite aliens present in its territory with the right to grant asylum and
confirmed that this right is a normal exercise of territorial sovereignty:
‘In the case of extradition, the refugee is within the territory of the State of refuge.
A decision with regard to extradition implies only the normal exercise of the
territorial sovereignty. The refugee is outside the territory of the State where the
offence was committed, and a decision to grant him asylum in no way derogates
from the sovereignty of that State.’5
5 ICJ 20 November 1950, Asylum Case (Colombia v Peru), I.C.J. Reports 1950, p. 274. The Asylum
Case evolved around the question whether Columbia had legitimately granted asylum to
Dr. Victor Haya de la Torre, who was charged with the crime of military rebellion by the
Peruvian government, in its embassy in Lima. The incident gave rise to two further decisions
of the Court: ICJ 27 November 1950, Request for interpretation of the Judgment of November
20th, 1950, in the asylum case, I.C.J. Reports, p. 395 (declared inadmissible); and ICJ 13 June
1951, Haya de la Torre Case, I.C.J. Reports 1951, p. 71.
6 UN General Assembly, Declaration on Territorial Asylum, 14 December 1967, A/RES/
2312(XXII), Article 1.
7 See, more extensively, F. Morgenstern, ‘The Right of Asylum’, 26 BYIL (1949), p. 327.
8 Ibid, p. 328, Grahl-Madsen (1972), p. 30.
9 The 1954 Caracas Convention on Diplomatic Asylum mentions asylum granted in legations
(defined as any seat of a regular diplomatic mission, the residence of chiefs of mission,
and the premises provided by them), war vessels and military camps or aircraft; Convention
on Diplomatic Asylum (28 March 1954) 18 OAS Treaty Series No. 18, Article 1. Also see
Convention of Havana on Right of Asylum (20 February 1928) 132 LNTS 323, Article 2;
and Montevideo Treaty on Political Asylum and Refuge (4 August 1939), Article 2.
Extraterritorial asylum under international law 117
‘In the case of diplomatic asylum, the refugee is within the territory of the State
where the offence was committed. A decision to grant diplomatic asylum involves
a derogation from the sovereignty of that State. It withdraws the offender from
the jurisdiction of the territorial State and constitutes an intervention in matters
which are exclusively within the competence of that State. Such a derogation from
territorial sovereignty cannot be recognized unless its legal basis is established in
each particular case.’10
15 1928 Havana Convention, Article 2; 1939 Montevideo Treaty, Article 6; 1954 Caracas
Convention, Article XI.
16 1954 Caracas Convention, Article II.
17 Article VI of the 1954 Caracas Convention also covers individuals being sought by private
persons or mobs over whom the authorities have lost control.
18 The issue of diplomatic asylum was discussed by the UN General Assembly at its 29th and
30th sessions but the debate was inconclusive. A report of the Secretary-General on the topic
forwarded to the General Assembly mentioned that only seven of the 25 States which had
presented their views were in favor of drawing up an international convention on the topic;
see UN. Doc. A/10139. By its resolution 3497 (XXX) of 15 December 1975, the General
Assembly decided to give further consideration to the question at a future session, but this
decision was not followed up. In resolution 1400 (XIV) of 21 November 1959, the General
Assembly requested the International Law Commission to undertake the codification of
the principles and rules of international law relating to the right of asylum. Regarding
diplomatic asylum, the Commission concluded in 1977 that the topic did not appear at
that time to require active consideration by the Commission; see Yearbook of the International
Law Commission, 1977, vol. II (Part Two), para. 109.
19 Institut de Droit International, Session de Bath 1950 (Resolution I), ‘L’asile en droit inter-
national public (à l’exclusion de l’asile neutre)’, Article 3 (2). The third and most compre-
hensive part of the resolution was devoted to establishing rules on extraterritorial asylum.
The resolution recognizes and delimits the legality of extraterritorial asylum by laying down
that asylum can be given ‘à tout individu menacé dans sa vie, son intégrité corporelle ou
sa liberté par des violences émanant des autorités locales ou contre lesquelles celles-ci sont
manifestement impuissantes à le défendre, ou même qu’elles tolèrent ou provoquent. Ces
dispositions s’appliquent dans les mêmes conditions lorsque de telles menaces sont le
résultat de luttes intestines’. Grahl-Madsen questions whether this resolution must be seen
as lex lata; Grahl-Madsen (1972), p. 49.
Extraterritorial asylum under international law 119
Asylum. This draft, neither of legally binding nature, followed closely and
elaborated upon the principles set out in 1954 Caracas Convention.20
The fact that a right to grant diplomatic asylum has not been recognized
outside Latin America does not preclude a state from offering refuge to persons
seeking shelter. It only implies that a grant of refuge remains subject to the
territorial sovereignty of the host state. This means that if the territorial author-
ities do not object to the grant of protection, the grant is perfectly legal. There
might be other situations in which a grant of asylum does, by its nature, not
derogate from the sovereignty of the territorial state. Thus, it has been con-
tended that to provide asylum to persons fleeing from mob-violence against
which the territorial authorities cannot offer protection, does not impinge upon
the prerogatives of the territorial state.21 From a similar rationale, it is stated
that to provide protection in situations of general political upheaval, in which
justice is not adequately administered, does not oppose the rule of non-inter-
vention.22
In other situations however, should the territorial state object to refuge
or demand surrender of the person requesting asylum, the extraterritorial state
will ordinarily not be entitled to grant asylum.23 The ICJ in the Asylum Case
underlined that ‘the safety which arises out of asylum cannot be construed
as a protection against the regular application of the laws and against the
jurisdiction of legally constituted tribunals’.24
Being subject to the territorial sovereignty of the host state implies, further,
that the state wishing to grant protection requires the consent of the territorial
state if it wishes to arrange for a safe-conduct out of the country. It has fre-
quently occurred that territorial states have refused to grant safe passage,
rendering the extent of protection dependant on the limited facilities diplomatic
missions have at their disposal.25 This can be problematic, especially if faced
20 The International Law Association discussed the topic of diplomatic asylum in close
connection to territorial asylum. For discussions and text of the draft convention, see
International Law Association, Legal Aspects of the Problem of Asylum, Part II: Report,
55 International Law Association Reports of Conferences (1972), p. 176-207.
21 Grahl-Madsen (1972), p. 46; Morgenstern (1951), p. 376; P. Porcino, ‘Toward Codification
of Diplomatic Asylum’, 8 New York University Journal of International Law and Politics (1976),
p. 446-447; R. Jennings and A. Watts (eds.), Oppenheim’s International Law, 9th Ed. (1992),
Vol. I, p. 1084. This appears also the view of the ICJ, see Asylum Case, p. 282-283.
22 Morgenstern (1951), p. 377. To this effect also the dissenting judges Read, Badawi Pasha
and Azevedo in the Asylum Case, p. 312, 320, 333-335.
23 Grahl-Madsen (1972), p. 46.
24 Asylum Case, p. 284. Note that, although phrased in terms of general applicability, this
remark was made against the backdrop of the 1928 Havana Convention.
25 Even under the Havana Convention of 1928 and the Montevideo Treaty of 1939, it was
disputed whether the territorial State was obliged to accede to a request for a safe-conduct
out of the country if diplomatic asylum was granted on proper grounds. The ICJ held that
the treaty obligations entered into by Peru did not mean that Peru was legally bound to
allow a safe conduct; Asylum Case, p. 279. Article XII of the 1954 Caracas Convention on
120 Chapter 4
All this does not detract from the fact that extraterritorial asylum is essentially
about reconciling the principle of territorial sovereignty with claims of human-
itarianism.30 It could be upheld that, if confronted with conflicting claims
of humanitarianism and state sovereignty, exceptional circumstances may make
it legitimate for diplomatic missions to refuse surrender. Morgenstern has
Diplomatic Asylum does contain an obligation on the side of the territorial State to allow
departure for foreign territory, except in situations of force majeure.
26 A notable example is the overflowing of West German embassies in Prague and Budapest
by East-German citizens in 1989, who demanded passage to the west. The embassy in
Budapest was forced to take out a lease on a nearby building to house the throngs of East
Germans arriving every day. The embassy occupations played a significant role in the
Hungarian decision to open its borders to the west, and with it, the fall of the Berlin Wall.
The situation of North Koreans attempting to reach South Korea by seeking asylum in
foreign embassies and consulates in China resembles the plight of the East Germans. The
number of North Koreans seeking diplomatic refuge has risen steadily since the mid 1990s,
resulting in the Chinese taking ever more security measures around embassy compounds.
In October 2003, South Korea temporarily closed its consulate in Beijing where around 130
North Koreans had taken refuge.
27 According to the US Government, the decision to grant refuge to Cardinal Mindszenty
was taken ‘under highly exceptional and most unusual circumstances and on urgent
humanitarian grounds at a time of foreign aggression against Hungary’, see M. Whiteman,
Digest of International Law, Vol. 6 (1968), p. 451.
28 For an account, K. de Jonge, Dagboek uit Pretoria, Amsterdam: van Gennep, 1987.
29 Grahl-Madsen (1972), p. 49. According to Porcino, ‘the allowance of safe passage transforms
temporary refuge into permanent asylum’; Porcino (1976), p. 438.
30 Morgenstern (1948), p. 236.
Extraterritorial asylum under international law 121
stated that ‘[i]t probably cannot be maintained that asylum can never be
granted against prosecution by the local government’.31 Likewise, Grahl-
Madsen does not rule out that in a case of ‘the most compelling considerations
of humanity’, heads of diplomatic missions may refuse to surrender a per-
son.32 And Riveles, in discussing the fate of the Durban Six, discussed below,
has argued that ‘[i]t should be recognized that a State has the permissible
response of granting temporary sanctuary to individuals or groups in utter
desperation who face repressive measures in their home countries.’33
In Oppenheim’s International Law, it is also mentioned that ‘compelling
reasons of humanity may justify the grant of asylum’.34 It refers, amongst
others, to the judgment in the Asylum Case, in which it was stated that:
violated international law’, since the events at the US embassy were considered
to constitute an attack on the entire diplomatic corps in Iran and consequently
any embassy was entitled to assist American personnel.3637 Another example
is the case of the Durban Six, which evolved around six prominent members
of the South-African anti-apartheid movement who had been served detention
orders and who sought refuge at the British consulate in Durban in 1984. The
British consulate complied with their request and promised it would not force
them out of the consulate, although the authorities also made clear that they
would not intervene on their behalf with the South African authorities and
that they could not stay indefinitely. After the embassies of the United States,
France, the Netherlands and the Federal Republic of Germany had later denied
to offer sanctuary, the Durban Six decided to depart voluntarily resulting in
the immediate arrest of five of them in front of the consulate building.38
But a majority of legal opinion appears to agree that the practice of diplom-
atic asylum is not uniform and of too inconsistent character to constitute a
rule of international custom.39 Some countries, such as Japan,40 reject the
doctrine altogether, and other countries – and this point is also illustrated by
the divergent responses of Western governments in the case of the Durban
Six – apply different and sometimes arbitrary considerations in choosing to
grant asylum. Neither is there the required conviction among states that the
practice reflects a norm of international law.41 Even though countries such
as the United States and the United Kingdom have on occasion granted refuge
in opposition to demands of the territorial state, both countries have denied
that there exists a legal right to that effect.42 The rather inconsistent and
36 For the Canadian position, see L.H. Legault, ‘Canadian Practice in International Law during
1979 as Reflected Mainly in Public Correspondence and Statements of the Department of
External Affairs’, 18 Canadian Yearbook of International Law (1980), p. 304-305. For a legal
comment, see: C.V. Cole, ‘Is There Safe Refuge in Canadian Missions Abroad?’, 9 IJRL (1997),
p. 662.
37 Legault (1980), p. 304-305.
38 For a factual background and legal appraisal see Riveles (1989), p. 139-159.
39 Morgenstern, in particular, has forcefully rejected the proposition that a right to grant
diplomatic asylum is part of customary law; Morgenstern (1948), p. 241-246. Also see Porcino
(1976), p. 445-446; Sinha (1971), p. 238; B. Gilbert, ‘The Practice of Asylum in Legations
and Consulates of the United States’, 3 AJIL (1909), p. 585; A.M. Rossitto, ‘Diplomatic
Asylum in the United States and Latin America: A Comparative Analysis’, 13 Brooklyn
Journal of International Law (1987), p. 114.
40 See eg the position of Japan in General Assembly discussions on the desirability of a
convention on diplomatic asylum. UN Doc. A/C.6/SR.1506 (1974), statements of Mr. Yokota
from Japan.
41 Also Porcino (1976), p. 445.
42 In surveying the 19th and 20th century practice of diplomatic asylum by the United Kingdom
and United States, Sinha observes that although both countries have on occasion authorized
asylum on humanitarian grounds, neither country has claimed that there is a right to grant
diplomatic asylum. Sinha (1971), p. 212-217. For the American position, see further Rossitto
(1987), p. 111-135.
Extraterritorial asylum under international law 123
43 For some views: W.M. Reisman, Sovereignty and Human Rights in Contemporary Inter-
national Law, 84 AJIL (1990), p. 872-873; A. D’Amato, ‘The Invasion of Panama Was a Lawful
Response to Tyranny’, 84 AJIL (1990), p. 516-524; L. Henkin, ‘An Agenda for the Next
Century: The Myth and Mantra of State Sovereignty’, 35 Virginia Journal of International
Law (1994), p. 115-118; S. Chesterman, Just War or Just Peace? Humanitarian intervention and
international law, Oxford University Press (2001); P. Hilpold, ‘Humanitarian Intervention:
Is There a Need for a Legal Reappraisal?’, 12 EJIL (2001), p. 437-467.
44 Porcino has argued that the protection of human rights warrants recognition and codification
of the right to grant asylum, by emphasising that the grant of asylum in an embassy is
only a ‘passive’ infringement on the rights of the sovereign: the foreign state does not enter
the territory of the sovereign state uninvited, and does not apply force or behave aggressive-
ly towards the territorial state. Accordingly, the state granting refuge makes only limited
incursion on the host state’s sovereign prerogatives: Porcino (1976), p. 446.
124 Chapter 4
under what circumstances human rights may displace the obligations of the
sending state vis-a-vis the territorial state.
Historically, the bond between diplomatic asylum and the privileged position
of diplomatic envoys was stronger than it is in current times. Early scholars
such as Grotius explained the legality of diplomatic asylum from the fiction
of exterritoriality45 – holding that the ambassador’s premises are inviolable
for they are outside the territory of the host state and placed in that of the
sending state.46 One of the first to reject this fiction was van Bynkershoek,
in positing that the ambassador’s immunities are functional and that the
ambassador’s premises may not be used to offer refuge to criminals.47
Although no longer seen as valid, the fiction of exterritoriality has on occasion
resurfaced in defending the legality of diplomatic asylum.48 Other writers
45 In the law on diplomatic relations, the terms extraterritoriality and exterritoriality are used
interchangeably, although the latter term appears to prevail.
46 Hugo Grotius, De Jure Belli ac Pacis, 1625 (translated by F.W. Kelsey, Oxford: Clarendon
Press, 1925), Book II, Chapter 18, Section IV, para 5: ‘[B]y a similar fiction, ambassadors
were held to be outside the limits of the country to which they were accredited. For this
reason they are not subject to the municipal law of the State in which they are living.’
47 C. van Bynkershoek , De Foro Legatorum Liber Sigularis, 1744 (translated by G.J. Laing.
Oxford: Clarendon Press, 1946), Chapter XVI, p. 79-80.
48 Dissenting in the Asylum Case, Judge Alvarez posited that the fiction of exterritoriality is
the basis for diplomatic asylum in Latin America and that accordingly, asylum is considered
not to intervene in the sovereign prerogatives of the host State; Asylum Case, p. 292. This
view is difficult to reconcile with the fact that even in Latin America, diplomatic asylum
is deemed legitimate in limited circumstances only.
Extraterritorial asylum under international law 125
54 Articles 41, paragraphs 1 and 3 VCDR and Articles 55 paragraphs 1 and 2 VCCR.
55 Articles 22, paragraph 1 VCDR and 31, paragraph 2 VCCR.
56 The original draft of the VCDR had held that the premises of the mission shall be used
solely for the performance of the diplomatic functions, whereas the final text merely prohibits
use which is incompatible with those functions. See Summary records of the ninth session,
Yearbook of the ILC 1957, vol. I, p. 143 at para. 55. Where the original text could have been
interpreted as prohibiting asylum on diplomatic premises per se – only allowing for deroga-
tion in case of special agreements; the standard of incompatibility is obviously more lenient.
57 B.S. Murty, The International Law of Diplomacy, The Diplomatic Instrument and World Public
Order, Dordrecht/Boston/London: Martinus Nijhoff (1989), p. 417.
Extraterritorial asylum under international law 127
ing that once an embassy has granted asylum, the territorial state is effectively
debarred from terminating the grant of refuge.58 It is due to the inviolability
of diplomatic premises that territorial states opposing refuge have often seen
no other option than to acquiesce in the situation and, in order to prevent
protracted stays and the political frictions flowing from it, have been willing
to grant safe conducts out of the country, also in situations where criminal
charges had been imposed on the fugitive.59
Under the Vienna Convention on Consular Relations, the inviolability of
consular premises is considerably more limited. Article 31 (2) prohibits the
receiving state to enter that part of the consular premises which is used
exclusively for the purpose of the work of the consular post.60 This could be
taken to mean that, if the receiving state has reason to believe that the con-
sulate is used for other purposes, it is permitted to enter the building and to
arrest persons charged with an offense.61 It follows that consular grants of
asylum are more susceptible to termination and that consulates are less appro-
priate locations for persons seeking asylum.62 Grahl-Madsen concludes that
the limited inviolability of consular premises makes a grant of refuge ‘pre-
carious at best’.63
Under general principles of treaty law a reasoning would be possible that
a grant of asylum interfering with local laws is a fundamental breach of the
VCDR or VCCR which relieves the receiving state from its own obligations under
these conventions vis-à-vis the sending state. Accordingly, the receiving state
would be able to legitimately assert a right to enter the premises in order to
ensure recover.64 The ICJ in the Tehran Hostage case made clear however that
the rules of diplomatic and consular law constitute a self-contained regime
which foresees the possible abuse of diplomatic privileges and immunities
and specifies the means at the disposal of the receiving state to counter any
such abuse.65 The two recourses mentioned by the Court are to declare mem-
bers of the diplomatic or consular staff persona non grata – which will obviously
not succeed in terminating a grant of asylum – and to break off diplomatic
relations altogether and call for the immediate closure of the offending
mission.66 Although effective, the latter option is notoriously drastic and will
normally not be considered politically opportune. In situations where premises
are used for the sole purpose of sheltering refugees, a third option might
further be to withdraw the diplomatic status from those premises.67
Apart from refuge in diplomatic and consular premises, there are other
situations in which extraterritorial grants of asylum can benefit from im-
munities. The most topical situation is refuge granted on board warships or
other public vessels. Under international maritime law, warships and govern-
ment ships operated for non-commercial purposes enjoy complete inviolabil-
ity.68 In the case of warships in the territorial sea of another state not comply-
ing with the laws of that state, the only remedy for the coastal state is to
require the vessel to leave the territorial waters.69 The result of the absolute
inviolability of warships is that the legal situation regarding asylum is similar
to that of refuge granted on diplomatic premises, the main difference being
that the warship may sail away with the refugee on board. Because the prob-
lem of requesting a safe-conduct out of the country is not present, a grant of
‘full-fledged’ asylum is better possible.
Other immunities than those of diplomatic and consular envoys or warships
will often depend on the particulars of bilateral or multilateral treaties, such
as Status of Forces Agreements. These immunities will not be discussed here.
67 The reasoning would be that premises solely used for other purposes than the diplomatic
mission cannot be defined as “premises of the mission” under Article 1 (i) VCDR. On this
option, see Denza (2008), p. 471.
68 Article 8(1) Convention on the High Seas, 29 April 1958, 6465 UNTS 450; Article 32 United
Nations Convention on the Law of the Sea (UNCLOS), 10 December 1982, 1833 UNTS 396.
Criminal jurisdiction may be exercised on board government ships operated for commercial
purposes, see Article 27 UNCLOS.
69 Article 23 Convention on the Territorial Sea and the Contiguous Zone, 29 April 1958, 516
UNTS 205; Article 30 UNCLOS.
Extraterritorial asylum under international law 129
violability, rather than a legal right to grant diplomatic asylum, which has
made the practice of diplomatic asylum a perpetuating phenomenon also
outside Latin America. Inviolability constitutes a legal obstacle to redress illegal
grants of asylum and hence gives rise to incentives on the part of the host state
wishing to terminate a grant of asylum to find alternative solutions. This,
together with the humanitarian and often passive nature of grants of diplom-
atic asylum, may explain why diplomatic asylum, also if constituting an affront
to local laws, is often tolerated and only rarely spurs bilateral frictions.
70 This section does not deal with other treaties containing a prohibition of refoulement, nor
with other provisions of the ECHR and ICCPR which may be construed as also prohibiting
refoulement.
71 Current legal textbooks on international refugee law contain specific reference to the (extra-
)territorial locus of the prohibition of refoulement: J.C. Hathaway, The Rights of Refugees under
International Law, Cambridge University Press (2005), p. 335-342; G.S. Goodwin-Gill and
J. McAdam, The Refugee in International Law, Oxford University Press (2007), p. 244-253;
K. Wouters, International Legal Standards for the Protection from Refoulement, Antwerp: Inter-
sentia (2009), p. 48-56, 203-216, 372-376, 435-438. For contributions dealing specifically with
the issue see, amongst many others, G. Noll, ‘Seeking Asylum at Embassies: A Right to
Entry under International Law?’, 17 IJRL (2005), p. 543-573; S. Legomsky, ‘The USA and
the Caribbean Interdiction Program’, 18 IJRL (2006), p. 677-695; and various contributions
in B. Ryan and V. Mitsilegas (eds), Extraterritorial Immigration Control. Legal Challenges,
Leiden/Boston: Martinus Nijhoff (2010).
72 U.S. Supreme Court 21 June 1993, Sale v. Haitian Centers Council, 509 US 155 (hereafter ‘Sale’);
House of Lords 9 December 2004, Regina v Immigration Officer at Prague Airport and another
ex parte European Roma Rights Centre and others, [2004] UKHL 55 (hereafter ‘Roma Rights’).
130 Chapter 4
73 Koh considered the Sale case ‘not lost in the legal but the political arena’ : H.H. Koh,
‘Reflections on Refoulement and Haitian Centers Council’, 35 Harvard International Law Journal
(1994), p. 20; Henkin referred to the judgment as an ‘eccentric, highly implausible interpreta-
tion of a treaty’: L. Henkin, ‘Notes from the President’, ASIL Newsletter, September-October
1993, p. 1. UNCHR, which had filed amicus curiae briefs in both Sale and Roma Rights,
considered the Supreme Court’s judgment ‘a setback to modern international refugee law’
and ‘a very unfortunate example’: UNHCR EXCOM, ‘UN High Commissioner for Refugees
Responds to US Supreme Court Decision in Sale v Haitian Centers Council’ (released 22
June 1993), excerpts published in 32 International Legal Materials (1993), p. 1215. UNHCR
has repeatedly affirmed that the prohibition of refoulement applies wherever a state operates
– including at the frontier, on the high seas or on the territory of another state: UNHCR,
‘Advisory Opinion on the Extraterritorial Application of Non-Refoulement Obligations under
the 1951 Convention relating to the Status of refugees and its 1967 Protocol’, Geneva, 26
January 2007, para 43; UNHCR, EXCOM, ‘Interception of Asylum-Seekers and Refugees:
The International Framework and Recommendations for a Comprehensive Approach’, UN
Doc. EC/50/SC/CRP.17, 9 June 2000, para 23. For further commentary see eg Legomsky
(2006), p. 686-691; G.S. Goodwin-Gill, ‘The Haitian Refoulement Case: A Comment’, 6 IJRL
(1994), p. 106-109.
74 The Inter-American Human Rights Commission considered Article 33 Refugee Convention
to apply to the Haitians interdicted on the high seas and found the United States Govern-
ment to have breached its treaty obligations in respect of Article 33: IACHR 13 March 1997,
The Haitian Centre for Human Rights et al. v. United States, Case 10.675, Report No. 51/96,
paras. 157-158. More recently, other treaty monitoring bodies have concluded that the
prohibition of refoulement (or: the wider duty not expose a person to ill-treatment) has no
territorial limitations, see the cases of Al-Saadoon and Mufdhi, Munaf v Romania and Marine
I, discussed in sections 4.3.2 and 4.3.3 below.
75 In this vein also ICJ 15 October 2008, Application of the International Convention on the
Elimination of All Forms of Racial Discrimination (Georgia v. Russian Federation) (Order), I.C.J.
Reports 2008, p. 642, para. 109, where the Court cumulatively observed that there was no
territorial restriction of general nature in the International Convention on the Elimination
of All Forms of Racial Discrimination nor a specific territorial limitation in the provisions
at issue. See also chapter 2.6.
Extraterritorial asylum under international law 131
76 The European Court of Human Rights employs the term expulsion however also in regard
of removals from a state’s territory raising issues under Article 3 ECHR regardless of
previous legal residence. See further infra n. 92 and accompanying text.
77 Hobbes defined the term exile as follows: ‘Exile (banishment) is when a man is for a crime
condemned to depart out of the dominion of the Commonwealth, or out of a certain part
thereof, and during a prefixed time, or for ever, not to return into it’. Thomas Hobbes,
Leviathan. The Matter, Form and Power of a Commonwealth Ecclesiastical and Civil (1651),
(translated A. Martinich, Broadview Press (2002)), Part II, Chapter 28, p. 235.
78 Articles 31 and 32 VCLT. On the ‘manifestly absurd or unreasonable’-test, see R.K. Gardiner,
Treaty Interpretation, Oxford University Press (2008), p. 329-330.
79 In this vein: Roma Rights, para. 64.
132 Chapter 4
The Refugee Convention does not contain a general provision outlining its
personal scope but sets out a hierarchy of attachments of the individual with
the state in delimiting the personal scope of application of the various rights
contained therein.80 Some rights accrue to all refugees ‘present within a state’s
territory’,81 other rights are reserved to those ‘lawfully within the state’82
and still others can only be invoked by refugees who have their ‘habitual
residence’ in a contracting state.83 This continuum of legal attachments reflects
the intention of the drafters of the Refugee Convention that not all refugees
should be able to claim all benefits contained in the Refugee Convention –
and in particular not those refugees who had ‘imposed themselves upon the
hospitality of [reception] countries’,84 and that some rights should only be
granted after the legal position of the refugee was regularised.85
While a majority of rights contained in the Refugee Convention specifically
refer to the required level of attachment of the refugee with the state, some
core rights, including the prohibition of refoulement laid down in Article 33(1),
are not equipped with any qualification as to the required legal or physical
relationship between the state and the refugee.86 Logically, this unqualified
nature gives rise to an assumption that these rights have a broader personal
scope than other Convention rights and do not necessarily depend on any
of the attachments mentioned under the other Convention rights. This as-
sumption finds support in the Convention’s drafting history, from which it
transpires that these rights were considered so central to refugee protection
that they had to be accorded to all refugees.87
In defining the territorial scope of the prohibition of refoulement laid down
in Article 33(1) Refugee Convention, we may depart from the understanding
that it applies to all refugees present on a state’s territory. At the other extreme
end of the territorial scale, it does not appear that Article 33(1) can be invoked
by persons still within their country of origin. Because Article 33(1) only
applies to refugees, and because only persons outside their country of national-
ity can be defined as refugees, a textual interpretation of the prohibition of
refoulement dictates that it cannot apply to persons who remain inside their
88 Article 1 (A)(2) Refugee Convention. According to the UNHCR Handbook: ‘It is a general
requirement for refugee status that an applicant who has a nationality be outside the country
of his nationality. There are no exceptions to this rule. International protection cannot come
into play as long as a person is within the territorial jurisdiction of his home country.’
UNHCR, Handbook on Procedures and Criteria for Determining Refugee Status under
the 1951 Convention and the 1967 Protocol relating to the Status of Refugees, January 1992,
UN Doc. HCR/IP/4/Eng/REV.1, para. 88. Also Roma Rights, para. 64.
89 See, more extensively, chapter 8.1.
90 Sale at 181-182.
91 Roma Rights, esp. paras. 17, 64, 70. At first glance, the potential applicability of Article 33(1)
Refugee Convention to the situation at Prague Airport case appears problematic, because
the Roma asylum-seekers were of Czech nationality and still within their own country and
could therefore not be properly defined as refugees. It was nonetheless submitted by the
appellants that the actions of the immigration officers violated the principle of ‘good faith’
in implementing the Refugee Convention; that the acts had defeated the ‘object and purpose’
of the prohibition of refoulement; and that the principle of non-refoulement had established
itself as a rule of customary international law which applied regardless of the place where
the State would undertake to expose refugees to persecution or other forms of ill-treatment.
See, for an extensive exposé of these arguments the letter amicus curiae filed by UNHCR,
‘R (ex parte European Roma Rights Centre et al) v. Immigration Officer at Prague Airport
and another (UNHCR intervening)’, reprinted in 17 IJRL (2005), p. 432-440.
134 Chapter 4
The House of Lords also accepted that the word ‘refouler’ may have a
different dictionary definition than the word return, but reasoned that the
putting of the French word ‘refouler’ between brackets in the English text was
done to clarify the meaning of the word ‘refouler’ rather than ‘return’. Accord-
ing to Lord Bingham of Cornhill, the requirement of Article 31(4) VCLT that
a special meaning is to be given to a term if it is established that the parties
so intended was pertinent, because ‘the parties have made plain that “refouler”,
whatever its wider dictionary definition, is in this context to be understood
as meaning “return”.’96 And because the term ‘return’ was considered only
applicable to measures imposed on refugees within the territory but not yet
resident there, Article 33(1) not only has no bearing on refugees outside a
state’s territory, but neither on refugees who are rejected at the border.97
The explanation of the Supreme Court for the inclusion of the term ‘refouler’
in the English text is more plausible than the one of the House of Lords. If
the intention of the Convention drafters had indeed been to make clear that
the French word ‘refouler’ was to have no different meaning than the English
word return, one should have expected the drafters to clarify the French
version of the treaty by inserting the English word return between brackets,
rather than the other way around. It indeed appears from the negotiating
history that the drafters were primarily occupied with finding an appropriate
translation for the French word ‘refouler’. This term had also featured in the
1933 Refugee Convention, of which only the French language version was
authentic, and which had been translated in the unauthentic English version
as ‘non-admittance at the frontier’, and with reference to the French word
‘refouler’ between brackets.98 In discussing the meaning of the term ‘return’
during one of the last sessions of the conference of plenipotentiaries in 1951,
the United Kingdom delegate had remarked that the Style Committee had
considered that the word return was the nearest equivalent in English to the
French word ‘refoulement’, from which the delegate deduced that the word
‘return’ had no wider meaning than the French term ‘refouler’. Upon this
remark, the president had proposed to insert the French word ‘refouler’ in
brackets in the English text, as had also been done in the 1933 Convention.99
In line with Article 31(4) VCLT, we may hence assume that the parenthetical
reference in Article 33(1) to the French verb ‘refouler’ suggests that return has
96 Roma Rights, paras. 17-18, per lord Bingham of Cornhill; Lord Hope of Craighead appears
to share this interpretation, see para. 70.
97 Ibid. Both Law Lords derived this narrow interpretation of the word return from early
legal commentaries to the Refugee Convention.
98 Convention relating to the International Status of Refugees, 28 October 1933, 159 LNTS
200, Article 3.
99 Statements of Mr. Hoare from the United Kingdom and Mr. Larsen (President), UN Doc.
A/CONF.2/SR.35, p. 21-22.
136 Chapter 4
‘The word “refoulement” is used in Belgium and France to describe a more informal
way of removing a person from the territory and also to describe non-admittance
at the frontier. It may be applied to persons seeking admission, persons illegally
present in a country, and persons admitted temporarily or conditionally, in the
latter case, however, only if the conditions of their stay have been violated.’101
There are two reasons to treat the delegates remarks with caution. First,
the declarations made by the Swiss and Dutch appear to have been instigated
by a fear that Contracting States would be compelled to allow mass migrations
across their frontiers.110 Indeed, the Dutch delegate had only wished ‘to have
it placed on the record that the Conference was in agreement with the inter-
pretation that the possibility of mass migration across frontiers or of attempted
mass migrations was not covered by article 33’, to which no objections were
made.111 Accordingly, the only interpretation placed on the record (but not
voted upon112) was that Article 33(1) was considered not to cover mass migra-
tions – not that return only applies to those already within a country but not
resident there.
Secondly, the rather isolated comments of the two delegates stand in sharp
contrast with the views on this particular issue taken by the Ad Hoc Commit-
tee on Statelessness and Related Problems, which prepared the draft text
forwarded for adoption to the Conference of Plenipotentiaries. The Ad Hoc
Committee had extensively debated the provision laying down the prohibition
of refoulement and had achieved consensus on the substance of the obligation.
In discussing differences in state practice as regards deportation and non-
admission, the US delegate had stated that the Convention ought to apply to
persons who asked to enter the territory of the contracting parties and that
country, he may be refused permission to proceed farther inland, but he must be allowed
to stay in the bit of the territory which is situated between the actual frontier line and the
control post, because any other course of action would mean a violation of Article 33 (1).’
In an effort to explain why these results – which Grahl-Madsen admits may seem strange
from a logical point of view – are nonetheless not devoid of merit, Grahl-Madsen continues
by giving an even more bewildering statement: ‘It must be remembered that the Refugee
Convention to a certain extent is a result of the pressure by humanitarian interested persons
on Governments, and that public opinion is apt to concern itself much more with the
individual who has set foot on the nation’s territory and thus is within the power of the
national authorities, than with people only seen as shadows or moving figures “at the other
side of the fence.” The latter have not materialized as human beings [sic], and it is much
easier to shed responsibility for a mass of unknown people than for the individual whose
fate one has to decide.’
110 Mr. Zutter, the Swiss delegate had remarked that Switzerland would only be willing to
accept the provision on non-refoulement if the other delegates accepted his interpretation
that the word return applied solely to refugees who had already entered a country but
were not yet resident there and that ‘[a]ccording to that interpretation, States were not
compelled to allow large groups of persons claiming refugee status to cross its frontiers.’
Statement of Mr. Zutter from Switzerland, UN Doc. A/CONF.2/SR.16, p. 6 (11 July 1951).
The Dutch delegate similarly communicated: ‘article 28 [current Article 33 – author] would
not have involved any obligations in the possible case of mass migrations across frontiers
or attempted mass migrations.’ And: ‘The Netherlands could not accept any legal obligations
in respect of large groups of refugees seeking access to its territory.’ Statements of Baron
van Boetzelaer of the Netherlands, A/CONF.2/SR.35 (25 July 1951), p. 21-22.
111 Ibid, Statement of Baron van Boetzelaer of the Netherlands.
112 The dissent in Sale observed that the fragments of the negotiating history referred to by
the majority ‘were never voted on or adopted, probably represent a minority view, and
in any event do not address the issue in this case’, Sale at 198.
Extraterritorial asylum under international law 139
113 Statement of Mr. Henkin of the United States, UN Doc. E/AC.32/SR.20, 1 February 1950,
paras. 54-56.
114 Statement of Mr. Robinson of Israel, ibid, para. 60.
115 Statement of Mr. Cuvelier of Belgium, UN Doc. E/AC.32/SR.21, 2 February 1950, para. 15.
116 Statement of Sir Leslie Brass of the United Kingdom, ibid, para. 16.
117 Statement of Mr. Leslie Chance of Canada, ibid, para. 26.
118 Roma Rights, para 17.
119 It remains unclear why no weight was accorded to those earlier proceedings. In its letter
amicus curiae to the Supreme Court in Sale, UNHCR did refer to statements made in the
Ad Hoc Committee, Office of the United Nations High Commissioner for Refugees, ‘The
Haitian Interdiction Case 1993, Brief amicus curiae’, re-printed in 6 IJRL (1994), p. 100.
140 Chapter 4
120 See the dissent in Sale, at 197. Also see H.H. Koh, ‘The “Haiti Paradigm” in United States
Human Rights Policy’, 103 Yale Law Journal (1994), p. 2417; Hathaway (2005), p. 318.
121 This also transpires from the drafting sessions of the Ad Hoc Committee on Statelessness
and Related Problems. The Canadian chairman had gathered that there was agreement
among the delegates on the principle that ‘refugees fleeing from persecution on account
of their race, religion, nationality or political opinions should not be pushed back into the
arms of their persecutors’, Statement of Mr. Chance of Canada, UN Doc. E/AC.32/SR.21,
2 February 1950, para. 26. In commenting upon the Sale judgment, Louis Henkin, former
US delegate in the Ad Hoc Committee, had found it ‘incredible that states that had agreed
not to force any human being back into the hands of their oppressors intended to leave
themselves free to reach out beyond their territory to seize a refugee and return him to
the country from which he sought to escape’. Henkin (1993), p. 1.
122 Sale at 183.
123 Roma Rights, paras. 18, 21, 26, 64
124 Ibid, para. 63.
Extraterritorial asylum under international law 141
Following the VCLT rules on treaty interpretation and the considerations above,
the most plausible approach for interpreting the territorial scope of Article
33(1) Refugee Convention would be to conceive the term ‘return’ in accordance
with the French term ‘refouler’, which translates as defensive or exclusive acts
and which is not necessarily restricted to conduct undertaken within the state’s
territory (Article 31(4) VCLT). That Article 33(1) includes rejection at the border
and outside the border is in line with the object and purpose of the provision
(Article 31(1) VCLT). Because the meaning of the terms ‘refouler’ (and hence,
‘return’) and the object and purpose of the provision shed sufficient light on
the question of territorial scope, it is not necessary to have recourse to the
preparatory work of the treaty (Article 32 VCLT), which in any event does not
yield a clear answer to the anticipated territorial scope of the provision. This
interpretation finds further contextual support in the phrase not to return a
refugee ‘in any manner whatsoever’ laid down in Article 33(1) Refugee Con-
vention, which reflects the notion that the form or manner of the act amounting
to expulsion or return is immaterial and that it may cover a great variety of
measures by which refugees are expelled, refused admittance or removed.125
It does however not follow that Article 33(1) Refugee Convention can apply
to all excluding acts vis-à-vis refugees undertaken outside the state’s territory.
Firstly, as mentioned above, the provision does not cover persons still within
their country of nationality or habitual residence. Secondly, it follows from
the words ‘to the frontier of territories’ that only acts of return (or other defens-
ive acts) taking place outside the territory where persecution is feared can come
within its ambit. This not only affirms the finding that the Refugee Convention
is anyhow not applicable to persons still within their country of origin, but
also means that Article 33(1), which can also protect against persecution in
125 E. Lauterpacht and D. Betlehem, ‘The scope and content of the principle of non-refoulement:
Opinion’, in: E. Feller c.s. (eds), Refugee Protection in International Law. UNHCR’s Global
Consultations on International Protection, Cambridge University Press (2003), p. 112. Hathaway
(2005), p. 317-322. Note however that the formula ‘in any manner whatsoever’ does not
detract from the requirement that the act must still qualify as an act of ‘expulsion’ or ‘return
(‘refouler’)’. It is unclear, for example, whether the act of extradition, in form and manner
very similar to the act of expulsion, comes within the ambit of Article 33 Refugee Conven-
tion. Lauterpacht and Betlehem infer from the wordings ‘in any manner whatsoever’ that
Article 33 also covers extradition: Lauterpacht and Betlehem (2003), p. 112-113. Goodwin-Gill
and McAdam note that although the preparatory work indicates that Article 33(1) did not
prejudice extradition, state practice indicates that non-refoulement also protects refugees from
being extradited. This conclusion is however mainly based on developments under the
ECHR, CAT and ICCPR; Goodwin-Gill and McAdam (2007) p. 257-262. Under the ECHR,
the word expulsion is expressly understood not to cover extradition: ‘With the exception
of extradition, any measure compelling the alien’s departure from the territory where he
was lawfully resident constitutes “expulsion” for the purposes of Article 1 of Protocol No.
7’; ECtHR 12 February 2009, Nolan and K. v Russia, no. 2512/04, para. 112.
142 Chapter 4
The Convention Against Torture does not contain a general clause setting out
the personal or territorial scope of the treaty, but several provisions in the CAT
limit their application to ‘any territory under a State Party’s jurisdiction’.127
These terms have been interpreted by the Committee Against Torture much
in line with the approaches of the ECtHR and HRC set out in Chapter 2 of this
book. In response to the position recently taken up by the US government that
Article 2 CAT (the obligation to prevent torture), applies only to US territory
and not to the detention facilities located in Guantánamo Bay, Cuba, the CAT
considered that
126 This follows from the words ‘the frontiers of territories where (…).’ See E. Lauterpacht and
D. Betlehem, ‘The scope and content of the principle of non-refoulement: Opinion’, in: E.
Feller c.s. (eds), Refugee Protection in International Law. UNHCR’s Global Consultations on
International Protection, Cambridge University Press (2003), p. 122.
127 This wording is found in Articles 2(1), 5(2), 11, 12, 13 and 16 CAT.
128 Conclusions and recommendations of the Committee against Torture, United States of
America, CAT/C/USA/CO/2, 25 July 2006, para. 15.
129 ComAT, General Comment no. 2, CAT/C/GC/2, 24 January 2008, para 16: ‘The Committee
has recognized that “any territory” includes all areas where the State party exercises, directly
or indirectly, in whole or in part, de jure or de facto effective control, in accordance with
international law. The reference to “any territory” in article 2, like that in articles 5, 11,
12, 13 and 16, refers to prohibited acts committed not only on board a ship or aircraft
registered by a State party, but also during military occupation or peacekeeping operations
and in such places as embassies, military bases, detention facilities, or other areas over
which a State exercises factual or effective control.’
Extraterritorial asylum under international law 143
extent, this interpretation corresponds with the treaty’s preparatory work. The
original obligation to prevent torture had prevented torture from being
practiced ‘within its [a State Party’s – author] jurisdiction’,130 but these words
were replaced with the words ‘any territory under its jurisdiction’, under the
reasoning that nationals (who for some purposes may be considered to fall
under the state’s legislative jurisdiction) living in another country should not
be able to rely on such protection.131 It was nonetheless underlined that ‘any
territory under its jurisdiction’ could also refer to acts taking place in such
foreign locations as ships, aircrafts and occupied territories.132 Similar to the
discussions in the preparatory stages of the ECRH and ICCPR therefore, the
preparatory work of the CAT suggests that the drafters were well aware of
the limits to a state’s capacity to provide human rights protection abroad, but
that they also agreed that when states would embark upon extraterritorial
adventures, the CAT would not automatically be void of meaning.133
Article 3 CAT explicitly prohibits refoulement. The provision was partly
modeled after Article 33 Refugee Convention and is equipped with similar
terminology.134 It prohibits states to ‘expel, return (“refouler”) or extradite
a person to another State where there are substantial grounds for believing
that he would be in danger of being subjected to torture’. Article 3 CAT does
not contain a reference to the jurisdiction or territory of the State Party, but
the references in other Convention provisions to territories under a state’s
jurisdiction support a contextual understanding that Article 3 CAT may also
apply extraterritorially. This was confirmed by the Committee in its view in
the Marine I case, concerning the Spanish rescue and processing of a group
of migrants whose ship had been in distress close to the shores of Senegal and
who were subsequently brought to an abandoned fish processing plant in
Mauritania. The Committee considered that its interpretation of the concept
of jurisdiction as reflected in General Comment No. 2 is applicable in respect
not only of Article 2, but of all provisions of the Convention, and that ‘such
jurisdiction must also include situations where a State Party exercises, directly
or indirectly, de facto or de jure control over persons in detention’.135 It
observed that Spain maintained control over the persons on board the Marine I
from the time the vessel was rescued and throughout the identification and
repatriation process that took place at the plant in Mauritania and that, by
virtue of that control, the alleged victims were subject to Spanish jurisdiction
for the purpose of the complaints regarding possible onward removal of the
migrants to the conflict in Kashmir in violation of Article 3.
The extraterritorial application of Article 3 CAT was also addressed by the
Committee Against Torture in the context of the transfer of detainees held
in custody by United Kingdom military forces in Afghanistan and Iraq to the
Iraqi and Afghan authorities. Regarding these transfers, the Committee
observed that ‘the Convention protections extend to all territories under the
jurisdiction of a State Party and considers that this principle includes all areas
under the de facto effective control of the State Party’s authorities’. Specifically
referring to the prohibition of refoulement, the Committee recommended that
‘the State Party should apply articles 2 and/or 3, as appropriate, to transfers
of a detainee within a State Party’s custody to the custody whether de facto
or de jure of any other State.’136
It was concluded above that the wording of Article 33 Refugee Convention
makes it difficult to bring persons under its ambit who are still within the
country from which the threat with persecution emanates. This limitation is
also present under Article 3 CAT, which speaks of ‘expel, return (“refouler”)
or extradite a person to another State where (…)’.137 A literal reading would
implicate that Article 3 CAT can only apply to situations where a person is
outside the country from which the threat stems: one can simply not transfer
someone to a place where he already is. This was also the United Kingdom’s
position in response to the Committee’s recommendations on the transfer of
detainees held captive in Iraq and Afghanistan.138 Nowak and McArthur
endorse the Committee’s approach however by holding that ‘[t]aking into
account the purpose of the absolute prohibition of refoulement, the term ‘an-
other’ State should in fact be interpreted as referring to any transfer of a person
from one State jurisdiction to another’.139 It is indeed notable that, different
from Article 33 Refugee Convention, Article 3 CAT does not speak of return
‘to the frontiers of territories’ but of return ‘to another State’. One could
accordingly contend that because the word state not necessarily refers to a
territorial entity but may also refer to all the organs making up a state, Ar-
ticle 3 CAT can be applicable to all situations where a person is transferred
136 Conclusions and recommendations of the Committee against Torture, United Kingdom,
CAT/C/CR/33/3, 10 December 2004, paras. 4(b) and 5(e).
137 Emphasis added.
138 The UK government argued that ‘[a]lthough a detainee may be physically transferred from
UK to Iraqi custody, there is no question of expulsion, return (refoulement) or extradition
to another State, as referred to in Article 3, all of which include an element of moving a
person from the territory of one State to that of another’, Comments by the Government
of the United Kingdom of Great Britain and Northern Ireland to the conclusions and
recommendations of the Committee Against Torture, CAT/C/GBR/CO/4/Add.1, 8 June
2006, para. 14.
139 Nowak and McArthur (2008), p. 199.
Extraterritorial asylum under international law 145
from the organ of one state to another, regardless of any territorial considera-
tions. This suggestion remains problematic nonetheless, because the placement
of the adverb ‘where’ immediately after the term ‘State’ in Article 3 CAT would
appear to indicate that, in the context of Article 3 CAT, the term state has a
geographical rather than a functional meaning (otherwise the adverb should
have been ‘which’). To wit, one could maintain that departure from the strict
literal meaning of Article 3 CAT is justified by the purpose of the absolute
prohibition of torture and the fact that the drafters of the CAT had probably
not contemplated situations in which refoulement would take place in an
extraterritorial setting. Construing Article 3 CAT in this way appears moreover
better possible than under the Refugee Convention, because the CAT was not
designed as an instrument to address the problem of persons having fled their
country of origin, but to maximize the effectiveness of the struggle against
torture throughout the world.140
The difference in approach of, on the one hand, the House of Lords and the
US Supreme Court in interpreting the extraterritorial applicability of Article 33
Refugee Convention, and, on the other hand, the Committee Against Torture
in interpreting Article 3 CAT, is remarkable to say the least, given the similarity
in wording of both provisions. This difference also points to the broader
jurisprudential development in which human rights treaty monitoring bodies
have construed the body of international human rights as having extraterri-
torial implications. On a general level, one may indeed contrast the pre-
sumption against extraterritoriality propounded by the House of Lords and
the US Supreme Court in regard of Article 33(1) Refugee Convention not only
with the position taken by the Committee Against Torture, but also with the
considerably more lenient position of the ECtHR and HRC in matters involving
extraterritorial acts of states as described in chapter 2. This difference in
approaches may be explained from the explicit reference in the ECRH and ICCPR
to the term jurisdiction, which is generally accepted to be not necessarily
confined to a state’s territory; and from the fact that Article 33 Refugee Conven-
tion speaks specifically of ‘expel’ and ‘return’, terms which are traditionally
used only in relation to the deportation of persons from a state’s territory. But
it is perhaps also appropriate to acknowledge that, in interpreting treaty terms,
human rights monitoring bodies have tended to attribute primary importance
to a teleological interpretation focused on the object and purpose of the treaty.
Meron and others have observed that this ‘teleological bias’ of human rights
140 See the CAT’s preamble (‘Desiring to make more effective the struggle against torture and
other cruel, inhuman or degrading treatment or punishment throughout the world’).
146 Chapter 4
courts has on occasion resulted in the ordinary meaning of treaty terms being
overridden and the legislative history or preparatory work ignored.141
In the context of extraterritorial state activity, the Human Rights Com-
mittee’s interpretation of the terms ‘within its territory and subject to its
jurisdiction’ laid down in Article 2(1) ICCPR is perhaps the most salient example
of this development. While it has been (tacitly) acknowledged by the Human
Rights Committee – and explicitly in legal discourse142 – that a literal reading
of this provision can only imply that ‘territory’ and ‘jurisdiction’ are cumulative
requirements for attracting a state’s obligations under the ICCPR, the Human
Rights Committee, in a course later adopted by the International Court of
Justice,143 has discarded this ordinary grammatical meaning by concluding
that ‘it would be unconscionable to so interpret the responsibility under
article 2 of the Covenant as to permit a State Party to perpetrate violations
of the Covenant on the territory of another state, which violations it could
not perpetrate on its own territory.’144 One may continue to oppose this
approach as being contra legem or as constituting an affront to the mainstream
of international treaty law.145 Nonetheless, the essentially convergent views
of the different human rights treaty bodies and the confirmation thereof by
ICJ, allow us to presently depart from the understanding that human rights
treaties, including the ICCPR, bind states with regard to persons in foreign
territories who can be considered to fall within the jurisdiction of that state.
Because the rights and freedoms defined in human rights treaties benefit
anyone within the jurisdiction of a Contracting State, it may be assumed that
extraterritorial acts of refoulement can bring potential victims within the ambit
141 T. Meron, The Humanization of International Law, Leiden/Boston: Martinus Nijhoff (2006),
p. 193.
142 See esp. Noll (2000), p. 440; Noll (2005), p. 557.
143 ICJ 9 July 2004, Legal Consequences of the Construction of a Wall in the Occupied Palestinian
Territory, Advisory Opinion, ICJ Reports 2004, paras. 108-111; ICJ 19 December 2005, Armed
Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda), para. 216.
144 HRC 29 July 1981, Delia Saldias de Lopez v. Uruguay, no. 52/1979, para. 12.3. In justifying
this departure from the text of Article 2(1) ICCPR, the Committee referred to Article 5 (1)
ICCPR, which stipulates that: ‘[n]othing in the present Covenant may be interpreted as
implying for any state, group or person any right to engage in any activity or perform any
act aimed at the destruction of any of the rights and freedoms recognized herein or at their
limitation to a greater extent than is provided for in the present Covenant’. Article 5 ICCPR,
a provision otherwise rarely invoked, is reminiscent of the good faith argument brought
forward by the applicants in the Roma Rights case. Tomuschat however, found Article 5
ICCPR to be incorrect as a basis for affirming the applicability of the Covenant outside
a State’s territory and instead suggests that construing the words “within its territory”
pursuant to their strict literal meaning must be rejected because that would lead to ‘utterly
absurd results’. See Individual opinion appended to the Committee’s views of Mr. C.
Tomuschat.
145 Noll (2005), p. 558-564, who, while acknowledging the HRC’s divergent position on the
matter, maintains that ‘[t]he expansion of Article 2(1) ICCPR (…) is hard to justify in
dogmatic terms’, and that the Committee’s interpretation ‘runs counter to the ordinary
meaning of its terms (an ‘and’ not being synonymous to an ‘or’)’, at p. 563.
Extraterritorial asylum under international law 147
‘In interpreting the Convention regard must be had to its special character as a
treaty for the collective enforcement of human rights and fundamental freedoms
(…). Thus, the object and purpose of the Convention as an instrument for the
protection of individual human beings require that its provisions be interpreted
and applied so as to make its safeguards practical and effective (…). In addition,
any interpretation of the rights and freedoms guaranteed has to be consistent with
Accordingly, with references to the object and purpose of both the Convention
as a whole and Article 3, a prohibition of refoulement is inserted in the Euro-
pean Convention. This is notwithstanding the fact that neither the text nor
the drafting history indicates that Article 3 was envisaged to apply to such
situations, notwithstanding that it concerns acts taking place outside a
country’s territory, notwithstanding that it concerns acts committed by or
under the responsibility of another state, and notwithstanding that it concerns
conduct which is yet (and may indeed prove never) to materialize.
In Soering, the key criterion for engaging a State’s responsibility was
formulated as there ‘having been shown substantial grounds for believing that
the person concerned, if extradited, faces a real risk of being subjected to
torture or to inhuman or degrading treatment or punishment in the requesting
country.’149 On a more general note, the Court added that the extraditing
Contracting State’s liability is incurred ‘by reason of its having taken action
which has as a direct consequence the exposure of an individual to proscribed
ill-treatment.’150 These considerations put the prohibition of refoulement under
Article 3 ECRH much in line with the Court’s doctrine on positive obligations,
and in particular the duty to protect, requiring states to take measures designed
to ensure that individuals within their jurisdiction are not subjected to treat-
ment contrary to Articles 2 and 3. Under this doctrine, states are required to
undertake reasonable steps to prevent ill-treatment of which the authorities
had or ought to have had knowledge, regardless of whether the ill-treatment
stems from private persons, from state organs or from naturally occurring
148 ECtHR 7 July 1989, Soering v the United Kingdom, no. 14038/88, paras. 87-88.
149 Ibid, para. 91.
150 Ibid.
Extraterritorial asylum under international law 149
illnesses.151 These categories are also covered by the protection against refoule-
ment.152 And similar to cases concerning expulsion or extradition (or foreign
cases), the existence of a ‘real risk’ of treatment contrary to the Convention
has in domestic situations been considered decisive in enlivening a duty on
the side of the state to undertake protective measures.153
Accordingly, the case law of the European Court is highly supportive of
a general rule that, whenever it is known or when it ought to have been known
that an individual within the jurisdiction of a Contracting State is exposed
to a real risk of ill-treatment, it is incumbent on that state to take steps to
prevent that risk from materializing. Decisive, in this regard, for engaging a
state’s duty to protect is not the form or manner in which the risk materializes,
nor the manner in which the state can negate this risk (be it through providing
police protection,154 through providing necessary medical services,155 by
removing abused children from parental care,156 through not releasing a
dangerous criminal from prison,157 or, indeed, by not expelling a person)
but whether the state has taken reasonable and appropriate preventive
measures to remove the risk of ill-treatment, or at the least to alleviate that
risk to such a level that it is no longer ‘real’ or ‘immediate’.
So construed, the prohibition of refoulement under the ECRH and ICCPR is
essentially an obligation to shield a person from harm. This protective duty
is well apt to apply regardless of territorial considerations, provided that a
person is within the jurisdiction of a Contracting State. In the context of
Article 2 ECRH, the right to life, the ECtHR has confirmed that states may also
incur protective duties in respect of persons outside their national bound-
aries.158 The ECtHR has affirmed that the same rationale applies to situations
where a state transfers or hands over a person in and to a potentially maltreat-
ing state. In the case of Al-Saadoon and Mufdhi, further discussed in section
4.5 below, the ECtHR found the physical transfer of two Iraqi detainees held
in a British military facility in Iraq to the custody of the Iraqi authorities to
have breached the United Kingdom’s obligations under Articles 2 and 3 of
151 ECtHR 28 October 1998, Osman v the United Kingdom, no. 23452/94, para. 115. ECtHR 10
May 2001, Z. a.o. v the United Kingdom, no. 29392/95, para. 73; ECtHR 4 May 2001, Kelly
a.o. v the United Kingdom, no. 30054/96, paras. 94-95; ECtHR 9 June 1998, L.C.B. v the United
Kingdom, no. 23413/94, para. 38.
152 ECtHR 29 April 1997, H.L.R. v France, no. 24573/94 (non-state organs); ECtHR 27 May 2008,
N. v the United Kingdom, no. 26565/05 (natural illness).
153 L.C.B. v the United Kingdom, para. 38; ECtHR 10 October 2000, Akkoc v Turkey, nos. 22947/93
and 22948/93, para. 81; ECtHR 28 March 2000, Mahmut Kaya v Turkey, no. 22535/93, para.
89; Osman v the United Kingdom, para. 116.
154 Osman v the United Kingdom, para. 115-121.
155 L.C.B. v United Kingdom, paras. 36-41.
156 Z. a.o. v the United Kingdom, para. 74.
157 ECtHR 24 October 2002, Mastromatteo v Italy, no. 37703/97, para. 69.
158 ECtHR 11 January 2001, Xhavara a.o. v Italy, appl. 39473/98. ECtHR 24 June 2008, Isaak v
Turkey, appl. 44587/98.
150 Chapter 4
the Convention and Article 1 of Protocol No. 13 because there were substantial
grounds for believing that the applicants would face a real risk of being
sentenced to death and executed.159 A similar reasoning, although no finding
of a violation of the ICCPR, was adopted by the Human Rights Committee in
the case of Munaf v Romania, concerning the handover of an Iraqi-American
dual national from the Romanian embassy in Baghdad to the multinational
forces in Iraq who was subsequently sentenced to death. The Human Rights
Committee held its previous jurisprudence in refoulement-cases to also apply
to the circumstances of this case, but considered that the Romanian authorities
could not have known that the complainant would face criminal charges.160
To conclude, the prohibition of refoulement established under the ECRH and
ICCPR articulates the essential protective duty of a State Party to not expose
a person within its jurisdiction to a real risk of ill-treatment. If established that
a person can be considered to be within the state’s jurisdiction, the state
becomes bound to comply with this protective duty, regardless of whether
the person is on the territory of that state, in his country of origin or in a third
state from which the threat with ill-treatment stems. This is not to exclude
however, that the limited practical and/or legal capabilities a state in certain
foreign situations may have, can inform (or displace) the substance (or material
scope) of a state’s protective duties.161
159 ECtHR 2 March 2010, Al-Saadoon and Mufdhi v the United Kingdom, no. 61498/08.
160 HRC 21 August 2009, Mohammad Munaf v Romania, no. 1539/2006, paras. 14.2, 14.5.
161 See, in general, chapter 2.5.2; and specifically with respect to situations where human rights
obligations may conflict with obligations vis-à-vis the host state, section 4.5.
162 Article 12(3) African Charter on Human and Peoples’ Rights: ‘Every individual shall have
the right, when persecuted, to seek and obtain asylum in other countries in accordance
with the law of those countries and international conventions.’
163 OJ 2007 C303/01. See further chapter 5.3.3.
Extraterritorial asylum under international law 151
Although lacking a clear basis in treaty law, legal scholars have employed
the right to seek asylum as informing a variety of rights associated with the
institution of asylum, albeit in divergent manner. Some perceive the right to
seek asylum as covering the entire range of rights associated with receiving
a proper status determination in and from the desired state of refuge,164 while
others connect the right to persons who try to flee from persecution but who
are subjected to deterrent mechanisms preventing them from reaching and
successfully claiming asylum in a safe country.165 Notably, most authors
derive from the right to seek asylum obligations of destination countries vis-à-
vis asylum-seekers instead of what could perhaps be its most obvious and
immediate significance: a right to escape from persecution – which is first and
foremost exercisable vis-à-vis the country which one attempts to flee.
It transpires from the drafting history of Article 14 UDHR that the right to
seek asylum was perceived as a right to escape persecution and that this right
did not prejudice the right of the petitioned state to deny asylum. The Drafting
Committee, responsible for preparing the text of the UDHR, had recommended
to include a provision stipulating that “Everyone has the right to escape
persecution on grounds of political or other beliefs or on grounds of racial
prejudice by taking refuge on the territory of any State willing to grant him
asylum”, a right which could be read as exercisable by both the individual
and the state granting asylum vis-à-vis the country of persecution but which
falls short of endowing the individual with a right to be granted asylum.166
Later discussions in the drafting sessions predominantly concerned the question
of whether the provision on asylum should not explicitly recognize a right
of the individual to be granted asylum by another state. A majority of re-
presentatives in the Working Group of the Commission on Human Rights was
in favor of including such a right and provisionally agreed upon the formula
that “Everyone has the right to seek and be granted, in other countries, asylum
from persecution.”167 During the final negotiations In ECOSOC however,
164 D. Stevens, ‘The Asylum and Immigration Act 1996: Erosion of the Right to Seek Asylum’,
61 The Modern Law Review (1998), p. 207-222; K. Coffey, ‘The Due Process Right To Seek
Asylum in the United States: The Immigration Dilemma and Constitutional Controversy’,
19 Yale Law and Policy Review (2001), p. 255-291; E. Ferris, ‘Internal Displacement and the
Right to Seek Asylum’, 27 Refugee Survey Quarterly (2008), p. 76-92; H. O’Nions, ‘The Erosion
of the Right to Seek Asylum’, 2 Web Journal of Current Legal Issues (2006)
165 M. Kjaerum, ‘Article 14’, in: A. Eide et al. (eds), The Universal Declaration of Human Rights.
A Commentary, Oslo: Scandinavian University Press and Oxford University Press (1992),
p. 221-224; B. Frelick, ‘‘Preventive Protection’ and the Right to Seek Asylum: A Preliminary
Look at Bosnia and Croatia’, 4 IJRL (1992), p. 439-454; T. Gammeltoft-Hansen and H.
Gammeltoft-Hansen, ‘The Right to Seek – Revisited. On the UN Human Rights Declaration
Article 14 and Access to Asylum Procedures in the EU’, 10 EJML (2008), p. 439-459.
166 UN Doc. E/CN.4/21, Annex F. For commentaries to the drafting of Article 14 UDHR see
A. Grahl-Madsen, The Status of Refugees in International Law, vol. II, Leiden: Sijthoff (1972),
p. 100-102; H. Lauterpacht, International Law and Human Rights, Hamden: Archon Books
(1968 reprint), p. 421-422; Gammeltoft-Hansen and Gammeltoft-Hansen (2008), p. 442-446.
167 UN Doc. E/CN.4/SR.57, p. 11.
152 Chapter 4
The right to seek asylum is taken here as giving expression to the right of
persons fearing persecution to make use of their right to leave a country for
the specific purpose of trying to obtain asylum. Although the right to seek
asylum and the more generally applicable right to leave a country are only
scarcely touched upon in international refugee law, they would appear to have
special significance in the context of extraterritorial migration controls, precisely
because those controls may prevent persons from approaching a destination
country in order to apply for asylum. In the absence of self-standing legal
168 UN Doc. A/C.3/253, UN Doc. A/C.3/SR.121, see in particular statements by Mrs. Corbet
from the United Kingdom; cf. UN Doc. E/800.
169 UN Doc. A/C.3/244, Statement of Mr. Pavlov from the Soviet Union.
170 Lauterpacht (1968), p. 422 at n. 72.
171 Ibid; Grahl-Madsen (1972), p. 101.
172 Ibid. In this vein also Kjaerum, in: Eide (1992), p. 224-225: ‘The right to seek asylum has,
in fact, one of its basis in the ‘right of emigration’’, referring to Article 13(2) UDHR, Article
12(2) ICCPR and Protocol No. 4 to the ECHR. Contra, Gammeltoft-Hansen and Gammeltoft-
Hansen (2008) p. 446, who, after a lengthy review of the drafting history of Article 14 UDHR
posit that the right to seek asylum should be perceived as a procedural right: the right to
be allowed access to an asylum procedure, which should be guaranteed by the receiving
state.
Extraterritorial asylum under international law 153
provisions giving substance to the right to seek asylum, this section describes
the contents of the right to seek primarily by analogy to the conditions under
which the right to leave can be asserted. Particular attention is paid to the
following questions: under what circumstances can refusing leave to migrants
in general, and asylum-seekers in particular, be considered in breach of the
right to leave? What is the relationship between measures of entry control and
the right to leave? Can the right to leave only be invoked vis-à-vis the terri-
torial state or also against a state employing extraterritorial border measures?
173 The origins of the right to leave can be traced back to the Magna Carta of 1215 which
stipulated that ‘All merchants may safely and securely go away from England, come to
England, stay in and go through England, (…)’ and that ‘Every one shall henceforth be
permitted, saving our fealty, to leave our kingdom and to return in safety and security,
by land or by water (…)’, Articles 41 and 42. Grotius also connected the right to leave to
the idea of free trade. He defended the axiom that ‘[e]very nation is free to travel to every
other nation, and to trade with it’, by observing that nature had not supplied every place
with all the necessaries of life and that, given the need for mutual exchange of resources
and services, ‘[n]ature has given to all peoples a right of access to all other peoples’, Hugo
Grotius, The Freedom of the Seas, originally published 1608, trans. by R. Van Deman Magoffin,
New York: Oxford University Press (1916), Ch. 1. Vattel recognised the right of a citizen
‘to quit his country’, which in some cases he regarded as absolute: ‘[t]here are cases in which
a citizen has an absolute right to renounce his country, and abandon it entirely – a right
founded on reasons derived from the very nature of the social compact.’ Vattel distinguishes
the right to quit a country from the right of emigration, which he also sees as a natural
right with which the state may not interfere: ‘[i]f the sovereign attempts to molest those
who have a right to emigrate, he does them an injury; and the injured individuals may
lawfully implore the protection of the power who is willing to receive them.’ E. de Vattel,
The Law of Nations, originally published 1758, trans. J. Chitty, Philadelphia: Johnson (1867),
Bk. 1, Ch. XIX, paras. 223-226. For modern appraisals of the right to leave, see eg R. Higgins,
‘The Right in International Law of an Individual to Enter, Stay in and Leave a Country’,
49 International Affairs (1973), p. 342; S. Juss, ‘Free Movement and the World Order’, 16 IJRL
(2004), p. 292; H. Hannum, The Right to Leave and Return in International Law and Practice,
Dordrecht/Boston/Lancaster: Martinus Nijhoff (1987), p. 4; C. Harvey and R.P. Barnidge,
‘Human Rights, Free Movement, and the Right to Leave in International Law’, 19 IJRL (2007).
154 Chapter 4
The right to leave is pronounced in the exact same terms in Article 12 (2) ICCPR
and Article 2(2) Protocol 4 ECRH: Everyone shall be free to leave any country,
including his own.175 The HRC and ECtHR have interpreted the right to leave
broadly and have affirmed that the right to leave is a self-standing norm, the
enjoyment of which does not depend on the purposes of travel.176 Apart from
travel bans or border-police measures preventing persons from leaving a
country,177 the HRC and ECtHR have accepted that that the confiscation, refusal
to issue or refusal to renew a passport can also come within the ambit of the
right to leave. According to the ECtHR ‘[a] measure by means of which an
individual is dispossessed of an identity document such as, for example, a
passport, undoubtedly amounts to an interference with the exercise of liberty
of movement’.178 The Human Rights Committee has explained that ‘[s]ince
international travel usually requires appropriate documents, in particular a
passport, the right to leave a country must include the right to obtain the
necessary travel documents.’179 Although passports are indeed the sine qua
non of the right to leave,180 analogous considerations apply to all identity
174 Refugee Convention, Article 28; Convention on Stateless Persons, Article 28; International
Convention on the Protection of the Rights of All Migrant Workers and Members of Their
Families, Article 8.
175 This is not a coincidence. While the original draft of Article 2 (2) Fourth Protocol to the
ECHR spoke of the right ‘to leave any State’, the Committee of Experts decided to substitute
the word ‘State’ for ‘country’, by referring to the text of Article 12 (2) ICCPR. The difference
appears marginal, although it was considered that the term ‘country’ could also apply to
regions which could not be designated as states. Council of Europe Committee of Experts,
‘Explanatory reports on the Second to Fifth Protocols to the European Convention for the
Protection of Human Rights and Fundamental Freedoms’, Doc. H (71) 11, Strasbourg (1971),
para. 10.
176 HRC, General Comment No. 27: Article 12 (Freedom of Movement), 2 November 1999,
CCPR/C/21/Rev.1/Add.9, para. 8. ECtHR 13 November 2003, Napijalo v Croatia, no. 66485/
01, para. 73.
177 Eg ECtHR 23 May 2006, Riener v Bulgaria, appl. 46343/99, para. 110 (travel ban); ECtHR
17 July 2003, Luordo v Italy, appl. 32190/96, para. 92 (travel ban). With regard to the former
East-German border-policing regime, see ECtHR 22 March 2001, Streletz, Kessler and Krenz
v Germany, appls. 34044/96, 35532/97 and 44801/98, paras. 98-101.
178 Napijalo v Croatia, para. 69. Also see ECtHR 31 October 2006, Földes and Földesné Hajlik v
Hungary, no. 41463/02, para. 33; ECtHR 22 May 2001, Baumann v France, no. 33592/96, para.
62; ECtHR 21 December 2006, Bartik v Russia, no. 55565/00, para. 36.
179 HRC, General Comment 27, para. 9.
180 Harvey and Barnidge (2007), p. 7
Extraterritorial asylum under international law 155
and travel documents necessary for exercising the right to leave, such as exit
visa.181
Other infringements of the right to leave may consist of the imposition
of various legal and bureaucratic barriers, such as exceedingly high fees for
travel documents,182 the obligation to describe precisely the envisaged travel
route,183 the requirement to be in the possession of a return ticket,184 or
such far-reaching measures as a prohibition on women to leave without the
consent of their husband.185 It has also been considered that to make depart-
ure of a mentally deranged offender conditional on the receiving country
placing that offender in a mental hospital may fall under Article 2 (2) Protocol
No. 4 ECRH.186
The expression any country in Article 12 (2) ICCPR and Article 2(2) Protocol 4
ECRH is important and implies, firstly, that the right to leave is applicable to
nationals and aliens alike, which also follows from the word ‘everyone’.187
Secondly, and of particular importance for this study, the expression has been
interpreted as obliging states not only to secure the right to leave from their
own territories, but also from that of territories of other states. In Peltonen v
Finland, the European Commission of Human Rights held the right to leave
to be applicable to a situation in which a Finnish national had already left
Finland for Sweden and when the Finnish authorities consecutively took
measures which prevented him from leaving Sweden.188 Similarly, the Human
Rights Committee has repeatedly accepted that that to refuse a passport to
a national living abroad can impede a person from leaving that other country
and therefore come within the ambit of the right to leave.189 In Peltonen v
Finland, the European Commission did not address the question whether
181 On exit visas, see HRC 26 April 2005, Concluding observations on Uzbekistan, CCPR/CO/
83/UZB, para. 19; HRC 24 April 2001, Concluding observations on Syrian Arab Republic,
CCPR/CO/71/SYR, para. 21; HRC 18 November 1996, Concluding observations on Gabon,
CCPR/C/79/Add.71, para. 16.
182 HRC 19 November 1997, Concluding observations on Iraq, CCPR/C/79/Add.84, para. 14.
183 HRC, General Comment 27, para. 17.
184 Ibid.
185 HRC 1 April 1997, Concluding observations on Lebanon, CCPR/C/79/Add.78, para. 18.
186 EComHR 4 October 1989, I.H. v Austria, appl. 10533/83, par. 11. For a situation of com-
pulsory care preventing a person from leaving his country see also EComHR 13 October
1993, Nordblad v Sweden, appl. 19076/91.
187 To this effect, see eg HRC Concluding Observations on Lebanon, 1 April 1997, CCPR/C/79/
Add.78, para. 22; and HRC 18 November 1996, Concluding observations on Gabon, CCPR/
C/79/Add.71, para. 16, concerning the confiscation of passports of foreign workers and
exit visa requirements imposed on foreign workers, respectively.
188 EComHR 20 February 1995, Peltonen v Finland, no. 19583/92.
189 HRC 15 November 2004, Loubna El Ghar v Libya, no. 1107/2002, para. 7.3.; HRC 29 July
1994, Peltonen v Finland, no. 492/1992, para. 8.4; HRC 23 March 1982, Vidal Martins v
Uruguay, no. R.13/57, para. 7; HRC 31 March 1983, Montero v Uruguay, no. 106/1981, para.
9.4; HRC 31 March 1983, Lichtensztejn v Uruguay, no. 77/1980, para. 8.3; HRC 22 July 1983,
Nunez v Uruguay, no. 108/1981, para. 9.3.
156 Chapter 4
Peltonen actually was ‘within the jurisdiction’ of Finland for the purposes of
Article 1 ECRH. The Human Rights Committee, on the other hand, has explicitly
accepted that nationals living abroad who are refused a passport, come within
the jurisdiction of the refusing state.
In the case of Baumann v France, the ECtHR considered that ‘the right to
leave implies a right to leave for such country of the person’s choice to which
he may be admitted’.190 Although this reasoning could be taken as to imply
that a right to leave only exists in so far as another country is willing to accept
a person,191 the better interpretation is that the Court indicates that the right
to leave can also be interfered with in situations where a person may be able
to leave for one particular country, but is prohibited from going to another.192
This is also apparent from the earlier Commission decision in the case of
Peltonen v Finland – from which the quote in Baumann was taken – in which
the denial to issue a passport to mr. Peltonen was considered to constitute
an interference with the right to leave, even though the refusal did not prevent
him from leaving Sweden for another Nordic country. It is from a similar
rationale that the Human Rights Committee has frequently stressed that ad
hoc travel documents such as laissez-passers or safe conducts out of the country
are no adequate substitutes for a passport, since they only allow for travel
to one particular destination.193 According to the Human Rights Committee:
‘the right of the individual to determine the State of destination is part of the
legal guarantee’.194
In line with the interpretation that an interference of the right to leave does
not depend on there being another country willing to grant entry, the ECtHR
has on multiple occasions stated that measures making it impossible for
persons to travel abroad must be considered as automatically restricting the
right to leave, also when the person concerned has no inclination to travel
abroad. Thus, in Napijalo v Croatia, the Court held that to deny the use of an
identity document to the applicant which, ‘had he wished’, would have per-
mitted him to leave the country, restricted his right to liberty of movement.195
And in Luordo v Italy, the Court found an order to stay in a place of residence
190 Baumann v France, para. 61, emphasis added. The formulation was repeated in Földes and
Földesné Hajlik v Hungary, para. 32; Napijalo v Coatia, para. 68; Bartik v Russia, para. 36.
191 G.S. Goodwin-Gill and J. McAdam, The Refugee in International Law, Oxford University Press
(2007), p. 381; V. Moreno Lax, ‘Must EU Borders have Doors for Refugees? On the Com-
patibility of Schengen Visas and Carriers’ Sanctions with EU Member States’ Obligations
to Provide International Protection to Refugees’, 10 EJML (2008), p. 352.
192 On the relationship between the individual right to leave and the right of the state to control
entry, see section 4.4.3 below. It does not transpire from the case of Baumann or from later
judgments of the ECtHR that the question of whether another country is in fact prepared
to allow entry is material for applicability of the right to leave.
193 Loubna El Ghar v Libya, para. 7.2; Nunez v Uruguay, para. 9.2; Lichtensztejn v Uruguay, para.
8.2; Vidal Martins v Uruguay, paras. 6.2, 9; Montero v Uruguay, paras. 9.2, 10.
194 HRC, General Comment 27, para. 8.
195 Napijalo v Croatia, para. 73. Also see Bartik v Russia, para. 36.
Extraterritorial asylum under international law 157
It was posited above that a state may interfere with the right to leave also if
no other state is in fact willing to grant entry. The question may be posed,
conversely, whether a state, by refusing entry into its territory, may also
interfere with a person’s right to leave the territory of another state. It is,
indeed, sometimes argued that measures of immigration control imposed by
countries of destination have a negative impact, or may even nullify, a person’s
right to leave his country of residence. Juss, stressing that 20th century re-
strictions on migration are a departure from the historical norm of free move-
ment, posits that immigration barriers render the right to leave practically
meaningless.200 Nafziger argues that the right to leave imposes an implicit
obligation on territories not to entirely deny entry to foreign nationals.201
And Dummett goes so far as to state that an individual authorized to leave
his country but not accepted by any other country would see his right to
emigration violated.202 Although it is certainly true that the practical meaning
of the right to leave depends on a corresponding right to be allowed entry
into another country, one must, for a number of reasons, be careful in con-
struing immigration restrictions as infringements of the right to leave. Such
pronouncements risk neglecting that the rights to enter and to leave are firmly
set apart in human rights law and subject to different principles of international
law.
First, one must not confuse the right to leave with a right to emigrate. The
latter is not a right in international law. Although the HRC states that the
freedom to leave covers ‘departure for permanent emigration’, this pronounce-
ment must be understood from the HRC’s insistence that the right to leave is
to be secured regardless the purpose of travel, and that it may not, for example,
be made dependent on whether a person intends to return or not.203 Concept-
ually, the difference between a right to emigrate and the right to leave is that
the former encompasses not only the activity of leaving but also that of (per-
manent) settlement in another country. It is clear that the possibility of obtain-
ing permission to settle in another country is a totally different matter than
the act of leaving a country. It may well be that persons who will not be able
to enter a country for purposes of settlement, are able to enter that country
for other purposes, such as business or family visits.204
Secondly, rights can exist without possibilities – and vice versa. While the
possibility to leave a country can very well be dependent on another country
allowing entry (unless one wishes to sail the Seven Seas), the right to leave
is not in principle affected by the unwillingness of other countries to allow
entry. To be sure, a person could still endeavor to make lawful use of his right
to leave in order to try to gain illegal entry. Vattel, therefore, after positing
that there is not only a right of citizens ‘to quit’ their country but also a ‘right
to emigrate’, rightfully submits that this is not a ‘full’ right ‘but imperfect with
respect to each particular country’, because the citizen ‘must ask permission
of the chief of the place; and, if it is refused, it is his duty to submit’.205 Such
interpretation does not render the right to leave nugatory. While one may
perceive the right to leave as an ‘imperfect’, ‘half’ or ‘dormant’ right when
no other country is willing to open its doors, its practical consequences get
in full swing once another country is willing to grant entry.
Thirdly and most fundamentally, the problem with construing entry
restrictions as potential infringements of the right to leave is that this would
in effect transform the right to leave into a qualified right of entry into another
state which would then only be subject to the restrictions permitted under
Articles 12 (3) ICCPR and 2(3) Protocol No. 4 ECRH.206 Such interpretation
would seriously transgress upon the axiom that states have, subject to their
treaty obligations, exclusive control over the admittance of aliens into their
territory. To account for the latter problem, Nafziger has submitted that the
right to leave would seem to require states, taken together, to respect the right
by not totally barring entry.207 This view corresponds with the one of Vattel,
when he speaks of the right as perfect in the general view but imperfect with
respect to individual countries. The reasoning would be, accordingly, that the
international community as a whole is under the duty to complete the right
to leave by allowing entry into at least one its constituents. As Nafziger admits
however, construing the right to leave as imposing a corresponding duty of
entry on the international community at large would have practical meaning
only if that duty is made more concrete and specific, by negotiating and
formulating agreements giving expression to the notion that a state has a
qualified duty to admit aliens.208 And as long as such agreements do not
exist, it remains problematic to construe the right to leave as more than a right
engaging the responsibility of individual states, in which the starting point
is that each state is primarily to guarantee this right to those within their own
territories.209
The question remains whether this reasoning – i.e. entry controls should not
be construed as coming within the scope of the right to leave – should also
apply to pre-frontier border control arrangements, which sort their practical
or legal effect already within the territory of the country of departure. There
is as of yet only scarce legal authority addressing the issue. Although, as noted
above, the HRC has accepted that carrier sanctions and other pre-frontier
arrangements do attract a state’s duties under the right to leave, the ECtHR,
in Xhavara v Italy, considered the ramming of an Albanian migrant boat by
an Italian coast guard vessel in international waters not to raise an issue under
the right to leave because the aim of the Italian operation was not to prevent
Albanians to depart from their country, but rather to prevent their entry into
Italy:
‘La Cour relève que les mesures mises en cause par les requérants ne visaient pas
à les priver du droit de quitter l’Albanie, mais à les empêcher d’entrer sur le territoire
italien. Le second paragraphe de l’article 2 du Protocole n° 4 ne trouve donc pas
à s’appliquer en l’espèce.’210
210 ECtHR 11 January 2001, Xhavara v Italy, no. 39473/98, emphasis added.
Extraterritorial asylum under international law 161
crossed the border of the territorial sea, a person’s right to leave his country
is deprived of any meaningful effect and the preventing of departure consti-
tutes an essential element of the enforcement activity. As is explained in the
section below, it does not follow from this reasoning that such controls are
necessarily in violation of the right to leave, but rather that the interference
must find justification in the particular circumstances of the case.
Different from the prohibition of refoulement under Articles 3 ECRH and 7 ICCPR,
the right to leave is not absolute. Although external migration controls restrict-
ing persons from leaving another country may well serve a legitimate purpose
and can hence find justification in international law, the cardinal implication
of the finding that such measures may interfere with the right to leave is that
they should comply with the requirements of Articles 2(3) Protocol No. 4 ECRH
and 12(3) ICCPR.211 This has profound consequences for the manner in which
such controls must be conducted: they are exported from the realm of a state’s
discretionary powers regarding entry and admittance, and placed in the regime
of human rights scrutiny in which it must be assessed whether a person’s right
to leave is affected, whether the restrictions are taken in pursuit of a legitimate
aim, whether they are in accordance with the law, and whether they can be
considered necessary and proportionate. A general framework for assessing
the legitimacy of restrictions under Article 12 ICCPR is set out in General
Comment 27 of the HRC. Although more scarce than under other qualified
rights, the available case law of the ECtHR on permitted restriction under Article
2(3) Protocol No. 4 ECRH indicates that the test to be applied is similar to that
under other provisions.212
What follows from these requirements, in particular, is that restrictions
which are applied outside a legal and procedural framework and implemented
not on an individual but general basis are problematic. The quality of law-
doctrine requires the law not only to establish the grounds for restricting the
right to leave but also to protect against arbitrary interferences, which implies
that the law ‘must indicate the scope of any such discretion conferred on the
211 It is said, in the context of the ECHR, that the right to leave a country does not have a very
broad effective scope, because practically all conceivable motives on the part of the author-
ities to refuse a person this right can be brought under the permitted restrictions: Van Dijk,
F. van Hoof, A. van Rijn et al (eds), Theory and Practice of the European Convention on Human
Rights, Antwerpen/Oxford: Intersentia (2006), p. 942. This argument is not very persuasive,
as the permitted restrictions under Articles 2(3) Protocol No. 4 ECHR and 12(3) ICCPR
do essentially not differ from those listed under other qualified human rights.
212 See, in particular, Bartik v Russia, Luordo v Italy, Napijalo v Croatia, Földes and Földesné Hajlik
v Hungary, Riener v Bulgaria.
Extraterritorial asylum under international law 163
competent authorities and the manner of its exercise with sufficient clarity.’213
It follows from the requirements of necessity and proportionality that re-
strictions must be assessed in the light of the individual circumstances of the
case, implying that restrictive measures must always take account of the
particular situation of each individual subjected to the measure.214 Further,
the right to an effective remedy requires allegations of violations of the right
to be subject to the possibility of thorough and effective scrutiny by the re-
sponsible authorities.215
In view of current practices of external immigration control, several more
specific remarks are in order. Firstly, as regards the requirement of ‘legitimate
aim’, it can be observed that the aim of immigration controls conducted in
foreign countries which may prevent foreigners from leaving that country not
necessarily corresponds with that of border checks and border surveillance
conducted along the state’s own territorial border. These latter, ‘regular’
controls are normally conducted as a measure of immigration enforcement,
to prevent illegal entry and to prevent persons from circumventing border
checks.216 Pre-border controls, on the other hand, have been described as
primarily aiming at reducing the potential burden posed by ‘failed’ migrants
altogether: by preventing persons who are unlikely to have a right of entry
from presenting themselves at the border, the risk of having to incur admin-
istrative, financial and social costs as a result of processing asylum-seekers
and not being able to enforce the removal of failed asylum-seekers or other
categories of migrants is minimized.217 If it can be established that this is
indeed the aim of a particular measure of pre-border control, the question rises
under what legitimate aim the measure must fall. Notably, the aim of ‘eco-
nomic well-being’ of the country, which is frequently referred to by the Euro-
pean Court as justification of measures of immigration control interfering with
213 ECtHR 2 August 1984, Malone v the United Kingdom, no. 8691/79, paras. 66-68; ECtHR 24
March 1988, Olsson v Sweden (No. 1), appl. 10465/83, para. 61; ECtHR 20 June 2002, Al-Nashif
v Bulgaria, 50963/99, para. 119 ; ECtHR 24 April 2008, C.G. a.o. v Bulgaria, no. 1365/07, para.
39. For the application of these requirements to restrictions on the right to leave, see in
particular Riener v Bulgaria, paras. 112-113.
214 The ECtHR has derived a duty on the side of the authorities under Article 2 of Protocol
No. 4 ‘to take appropriate care that any interference with the right to leave one’s country
should be justified and proportionate throughout its duration, in the individual circum-
stances of the case’: Riener v Bulgaria, para. 128. Similarly, the Human Rights Committee
speaks of the application of restrictions which, ‘in any individual case’, must be based on
clear legal grounds and meet the test of necessity and proportionality, HRC, General
Comment No. 27, para. 16.
215 Article 13 ECHR; Article 2(3) ICCPR.
216 This is also the manner in which border controls are defined in the Schengen Borders Code,
see Articles 2 (9)-(11) EC Regulation 562/2006.
217 In this vein: Roma Rights, para. 2, where the posting of British immigration officers at the
airport of Prague was explained from the background of there being an ‘administrative,
financial and indeed social burden borne as a result of failed asylum-seekers’.
164 Chapter 4
218 For instance ECtHR 21 June 1988, Berrehab v the Netherlands, no. 10730/84, para. 26; ECtHR
31 January 2006, Rodrigues da Silva and Hoogkamer v the Netherlands, no. 50435/99, para. 44.
219 The question whether limitations for economic reasons can be brought under the permitted
restrictions of Article 2(3) Protocol No. 4 ECHR and Article 12(3) ICCPR has received
considerable attention, although mainly in the context of limitations imposed by countries
of departure for purposes of preventing ‘brain drain’. Van Dijk, van Hoof and van Rijn
(2006), at p. 944, conclude, in the context of Article 2(3) Protocol No. 4 ECHR, that the
freedom to leave may not be restricted on purely economic grounds. Hannum (1987), at
p. 40, in respect of Article 12(3) ICCPR, concludes that ‘most limitations imposed on the
right to leave on economic grounds must be judged in the context of good faith – or lack
thereof – of the government concerned. If the limitation of a particular right is necessary
to deal with a demonstrable socio-economic problem that threatens public order in a country
– particularly when the limitation is proportional, temporary , and determined with ade-
quate notice to those affected by it – it may well fall within the narrow range of limitations
permitted under article 12(3) of the Covenant.’
220 See extensively chapter 6.
221 P. Minderhoud and S. Scholten, ‘Regulating Immigration Control: Carrier Sanctions in the
Netherlands’ 10 EJML (2008), p. 131, 137-138; Gammeltoft-Hansen and Gammeltoft-Hansen
(2008), p. 450-451.
222 Roma Rights, paras. 5, 77-86.
Extraterritorial asylum under international law 165
Although the material scope of the right to leave is limited on account of its
qualified nature, the personal scope is wider than that of duties inherent to
the prohibition of refoulement as discussed in section 4.3. The right to leave
not only benefits persons who are seeking – and entitled to – international
protection, but can be invoked by anyone who wishes to leave a country, for
whatever (or indeed: no) purpose he may have in mind.
223 Földes and Földesné Hajlik v Hungary, para. 32; Riener v Bulgaria, para. 109.
224 HRC, General Comment 27, para. 14.
225 Ibid.
226 Streletz, Kessler and Krenz v Germany, para. 100.
227 G. Cornelisse, ‘European Vessels, African Territorial Waters and ‘Illegal Emigrants’: The
Right to Leave and the Principle of (Il)legality in a Global Regime of Mobility’, available
at <www.libertysecurity.org>.
166 Chapter 4
228 See, in this connection, Moreno Lax (2008), p. 356; who argues that ‘the aggregate right
to leave to seek asylum imposes a stricter principle of proportionality’.
229 Ibid.
230 This reasoning corresponds to the requirement under Article 12(3) ICCPR that restrictions
of the right to leave may not impair other rights of the Covenant.
Extraterritorial asylum under international law 167
The previous sections 4.3 and 4.4 concluded that the prohibition of refoulement
and the right to leave any country, including his own, inform the duties of
states who are confronted with asylum-seekers in another state. But it was
also concluded in section 4.2 that the conduct of the sending state must as
a rule respect the sovereignty of the host state and that the sending state has
a right to grant extraterritorial protection only in so far as that right does not
intervene in matters which are essentially within the domestic jurisdiction of
the host state. Scenarios may arise in which the sending state is approached
by a person requesting asylum or another form of protection but where the
host state opposes to the grant of protection. These scenarios are not merely
hypothetical. Persons subjected to pre-entry clearances at a foreign airport who
wish to exit that country without valid identity papers, for example, will
normally offend the laws of that country and immigration officers of a sending
state conducting pre-clearances must respect the local laws in force. But they
may also be confronted with valid individual claims for asylum. This may
confront the sending state with a conflict of norms, one stemming from being
party to a human rights treaty, the other stemming from the principle of
territorial sovereignty and the derivative rule of non-intervention. How should
a state reconcile these opposing norms?
This question should be addressed in conjunction with the deliberations
on the different meaning of the term jurisdiction in general international law
and human rights law as set out in Chapter 2. It would follow from that
analysis that essentially two approaches to the question would be conceivable.
First, it could be argued, in line with the ordinary meaning of the term ‘juris-
diction’ in general international law, that the requirement of ‘jurisdiction’ under
a human rights treaty may avoid a situation of norm conflict from arising,
by considering the person in question not to fall under the jurisdiction of the
sending state. The reasoning would be that even though a person may
(initially) be within the ‘jurisdiction’ or ‘effective control’ of the sending state,
any act with regard to that person which constitutes an affront to the territorial
sovereignty of the host state is an act over which the sending state de jure lacks
‘control’ or ‘authority’, because this control and authority accrues to the host
state in its capacity as the sovereign power. It would follow from this reasoning
that the sending state is only obliged to secure the human rights of the person
concerned in so far as doing so will not encroach upon the host state’s sover-
eignty.231
The other conceivable approach would be to argue that the term ‘juris-
diction’ within human rights law has clearly distanced itself from the original
notion under international law and has attained sui generis standing. This
approach would not preclude the possibility of a state being obliged to ensure
human rights to persons who are within their ‘jurisdiction’ in the human rights
meaning of the term, also if they remain subject to the concurrent jurisdiction
of another state and if an act regarding that person would transgress upon
the sovereignty of the other state. In this approach, the mere existence of a
sufficiently close legal or physical relationship between the extraterritorial state
and the individual suffices to enliven the state’s human rights obligations vis-à-
vis the individual. The question whether this would conflict with obligations
stemming from other sources of international law is then not relevant for the
jurisdiction issue, but may alternatively be incorporated in defining the scope
of a state’s substantive human rights obligations.
Outside the asylum context, confirmation of the first proposition can be
found in the case of Gentilhomme, concerning complaints lodged against the
refusal of French state schools situated in Algeria to continue to enrol several
children with dual French and Algerian nationality. The refusal was in com-
pliance with an indication of the Algerian government to the French govern-
ment that French state schools in Algeria should close their doors for children
with Algerian nationality. The ECtHR, while expressly noting that ‘a State may
not actually exercise jurisdiction on the territory of another without the latter’s
consent, invitation or acquiescence’,232 considered that the French refusal
constituted an implementation of a decision imputable to Algeria, taken by
the sovereign on its own territory and therefore beyond the control of
France.233 Accordingly, it found that the children could not be said to fall
within French jurisdiction.
In the particular context of grants of protection which may run counter
to demands of the host state, a similar approach was followed by the England
and Wales Court of Appeal in the case of Al-Saadoon and Mufdhi, which was
later brought before the European Court of Human Rights. The case concerned
the lawfulness of the proposed transfer of two Iraqi nationals, who were
accused of the murder of two British soldiers, from British military facilities
in Iraq to Iraqi custody for trial by the Iraqi High Tribunal. The Iraqi Tribunal
had repeatedly requested their transfer and the United Kingdom was, under
the various agreements concluded with the interim government of Iraq, obliged
to comply with these requests. The Court of Appeal, quoting at length from
the ECtHR decision in Bankovic and recalling that the ECtHR, in interpreting the
term jurisdiction, had underlined that account had to be taken of ‘of any
relevant rules of international law applicable in the relations between the
232 ECtHR 14 May 2002, Gentilhomme v France, nos. 48205/99, 48207/99 and 48209/99, para.
20. The Court referred here to its earlier pronouncements in Banković, paras. 59-61.
233 Ibid.
Extraterritorial asylum under international law 169
parties’ (Article 31(3)(c) VCLT),234 held that for a person to come within
‘Article 1 jurisdiction’ a mere exercise of de facto power – in the meaning of
effective control or authority – is insufficient. Instead, Lord Justice Laws
formulated four ‘core propositions’ from which it would follow, amongst other
things, that the question of jurisdiction must be ascertained in harmony with
other applicable norms of international law and that it implies the possibility
of exercising sovereign legal authority.235 Observing that the British forces
in Basra were not, from at least May 2006 until 31 December 2008, entitled
to carry out any activities on Iraq’s territory in relation to criminal detainees
save as consented by Iraq and that the British forces no longer enjoyed a legal
power to detain any Iraqi from 1 January 2009 onwards, Lord Justice Laws
concluded that the United Kingdom ‘was not exercising any autonomous
power of its own as a sovereign state’ and that ‘the detention of the appellants
by the British forces at Basra did not constitute an exercise of Article 1 juris-
diction by the United Kingdom’.236
The England and Wales Court of Appeal had approached the matter
differently in the earlier case of B and others, on the traditional question of
diplomatic asylum. The case concerned the legality under the United Kingdom
Human Rights Act and the ECRH of the refusal of the British authorities to
comply with a request for asylum lodged by two minor Afghan brothers in
the British consulate in Melbourne, Australia, who submitted that their return
234 Court of Appeal (England and Wales) 21 January 2009, R (on the application of (1) Faisal
Attiyah Nassar Al-Saadoon (2) Khalaf Hussain Mufdhi) v. Secretary of State for Defence, [2009]
EWCA Civ 7, para. 25.
235 Ibid, para. 37: ‘It is not easy to identify precisely the scope of the Article 1 jurisdiction where
it is said to be exercised outside the territory of the impugned State Party, because the
learning makes it clear that its scope has no sharp edge; it has to be ascertained from a
combination of key ideas which are strategic rather than lexical. Drawing on the Bankovic
judgment and their Lordships’ opinions in Al-Skeini, I suggest that there are four core
propositions, though each needs some explanation. (1) It is an exceptional jurisdiction. (2)
It is to be ascertained in harmony with other applicable norms of international law. (3) It
reflects the regional nature of the Convention rights. (4) It reflects the indivisible nature of
the Convention rights. The first and second of these propositions imply (as perhaps does
the term jurisdiction itself) an exercise of sovereign legal authority, not merely de facto power,
by one State on the territory of another. That is of itself an exceptional state of affairs,
though well recognized in some instances such as that of an embassy. The power must
be given by law, since if it were given only by chance or strength its exercise would by
no means be harmonious with material norms of international law, but offensive to them;
and there would be no principled basis on which the power could be said to be limited,
and thus exceptional.’
236 Ibid, paras. 32-36, 40. The language used may be taken to suggest that the actions of the
British forces should not be attributed to the United Kingdom but to Iraq. The Court of
Appeal only dealt with the jurisdiction issue however; the High Court had already con-
cluded that, in view of the autonomous tasks performed by the Multi-National Forces, the
detention and possible transfer of the appellants were properly attributable to the United
Kingdom: High Court (England and Wales) 19 December 2008, Al-Saadoon & Anor, R (on
the application of) v Secretary of State for Defence, [2008] EWHC 3098 (Admin), para. 79.
170 Chapter 4
237 Court of Appeal (England and Wales) 18 October 2004, R (B and others) v Secretary of State
for Foreign and Commonwealth Affairs [2004] EWCA Civ 1344, [2005] QB 643.
238 Ibid, para. 66. In particular, the Court of Appeal relied on an analogy with the case of WM
v Denmark, which had also concerned a person seeking refuge at an embassy (the Danish
embassy in Eastern Berlin), and where the European Commission of Human Rights had
applied the test of whether the acts of the Danish ambassador constituted an ‘exercise of
authority’ over the person in question to an extent sufficient to bring him within the
jurisdiction of Denmark: EComHR 14 October 1992, W.M. v Denmark, no. 17392/90. Also
see chapter 2.5.2.
239 Ibid, paras. 85-89.
240 Ibid, para. 88. The Court of Appeal further did not exclude the possibility that a lesser level
of threatened harm could also justify an entitlement to grant diplomatic asylum, but the
law to provide insufficient guidance on the issue..
241 Ibid, paras. 93-94. This approach was followed in the first instance decision in Al-Saadoon
and Mufdhi: High Court 19 December 2008, Al-Saadoon & Anor, R (on the application of) v
Secretary of State for Defence, [2008] EWHC 3098 (Admin).
Extraterritorial asylum under international law 171
were arrested by British armed forces, that they were detained in premises
which were inviolable and subject to exclusive control and authority of the
Multi-National Forces and that, ‘given the total and exclusive de facto, and
subsequently also de jure, control exercised by the United Kingdom authorities’,
the individuals were within the United Kingdom’s jurisdiction.242 As to the
question whether the obligations vis-à-vis Iraq could nonetheless ‘modify or
displace’ the obligations under the ECRH, the Court found that this was a matter
to be considered in relation to the merits of the complaints.243 In its judgment
on the merits, the ECtHR refrained from according the principle of territorial
sovereignty overriding importance however. Instead of defining the scope of
a contracting state’s duties under the ECRH in accordance with that principle,
the Court referred to the principles set out in its earlier case law – which
concerned situations where the guaranteeing of human rights within a state’s
territory potentially conflicted with other international law obligations – in
holding that ‘a Contracting Party is responsible under Article 1 of the Conven-
tion for all acts and omissions of its organs regardless of whether the act or
omission in question is a consequence of domestic law or of the necessity to
comply with international legal obligations.’244 It made a reference to the
Soering case, in observing that the Court in that case had neither limited the
application of Article 3 ECRH on account of a conflicting obligation on the part
of the United Kingdom under the Extradition Treaty it had concluded with
the United States in 1972. It follows, according to the Court, that in principle
all acts and omissions attributable to the state are subject to the Court’s
scrutiny.245 Although the Court did not explicitly consider that this can also
imply that the European Convention may require a contracting state to act
in contravention of another state’s sovereignty, it found the United Kingdom
to have made insufficient attempts at procuring a guarantee that the detainees
would not be subjected to the death penalty and to have entered into an
arrangement with another state which conflicted with its obligations under
Articles 2 and 3 of the Convention and Article 1 of Protocol No. 13.246 And,
in respect of the complaint under Article 34 ECRH, it considered that the
242 ECtHR 30 June 2009, Al-Saadoon and Mufdhi v the United Kingdom, no. 61498/08 (admissib-
ility), paras. 86-88.
243 Ibid, para. 89.
244 ECtHR 2 March 2010, Al-Saadoon and Mufdhi v the United Kingdom, no. 61498/08 (merits),
para. 128, referring to ECtHR 30 June 2005, Bosphorus v Ireland, no. 45036/98, para. 153.
Also see ECtHR 18 February 2009, Andrejeva v Latvia, no. 55707/00, para. 56: ‘the fact that
the factual or legal situation complained of is partly attributable to another State is not
in itself decisive for the determination of the respondent State’s “jurisdiction”.’
245 Also see Bosphorus, para. 137. Although this may be different in situations where state
activity serves the effective fulfillment of the mandate of United Nations Security Council,
see ECtHR 2 May 2007, Behrami v France and Saramati v France and Norway, nos. 71412/01
and 78166/01, para. 149 and ECtHR 9 June 2009, Galić v the Netherlands, no. 22617, paras.
47-48.
246 Al-Saadoon and Mufdhi (merits), paras. 141-143.
172 Chapter 4
The case law above signifies that courts have developed divergent lines of
reasoning in reconciling the rule of non-intervention with human rights obliga-
tions. The different approach of the ECtHR in Gentilhomme and Al-Saadoon and
Mufdhi on the question whether respect for the territorial sovereignty of another
state is relevant for the jurisdiction issue underscores the conclusion in chapter
2 that the European Court’s interpretation of that term is not always consistent
and that the Court tends to confuse the ordinary meaning of the term juris-
diction under international (i.e. to allocate state competences) with the more
specific delimiting function it fulfils in human rights law. The most recent
approach of the ECtHR in Al-Saadoon and Mufdhi also affirms the conclusion
of Chapter 2 however that the ECtHR is distancing itself from the imperative
to interpret the term jurisdiction in conformity with its ‘ordinary meaning’,
leaving room for an interpretation that limits set by other norms of inter-
national law do not as such prevent the Convention from being applicable.
Other and potentially conflicting international obligations remain subject to
the scrutiny of the Court, which also implies that the Court leaves open the
possibility that extraterritorial human rights obligations may trump the prin-
ciple of respect for the territorial sovereignty of the host state.
What is notable in this respect is that the ECtHR in Al-Saadoon and Mufdhi
does not appear to distinguish as a matter of principle between state activity
carried out on its own territory and activity carried out within the territorial
sovereignty of another state. Its reference to the Soering case appears to indicate
that in establishing whether other international obligations can modify the
scope of a state’s obligations under the ECRH, the obstacle of the territorial
sovereignty of another state should not be addressed fundamentally different
from ordinary extradition obligations. This raises questions in view of the ICJ’s
emphasis in the Asylum Case on the fundamental distinction which exists
between situations involving extradition and situations of extraterritorial
asylum, with the latter constituting a potential intervention in the sovereign
matters of the other state. It should also be observed however that the Court
underlined the lack of genuine efforts on the part of the United Kingdom’s
authorities to ensure that a potential future transfer would not expose the
applicants to treatment contrary to the Convention. The Court hence avoids
an explicit pronouncement that human rights must prevail over territorial
sovereignty, and instead appears to argue that because the United Kingdom
248 The ECtHR appears to hint in this direction in para. 140 of the judgment.
174 Chapter 4
and/or human rights obligations in respect of the individual and the rule of
non-intervention on the other hand. As was shown in the last section of this
chapter, attempts undertaken in recent case law to reconcile the norms in
question have not been consistent, which is due not only to divergent inter-
pretation of the notion of ‘jurisdiction’ in human rights law, but also to the
fundamental status of both human rights and the notion of territorial sover-
eignty in international law.
Third, the right to seek asylum – understood as the right to leave a country
in order to escape persecution – remains crucial for asylum-seekers who are
subjected to measures of border enforcement in countries of origin or countries
of transit. Policies aimed at preventing persons from leaving another country
may well interfere with the right to leave. Although there can be legitimate
reasons for placing restrictions on that right, those restrictions must have a
basis in law, may not be applied arbitrarily and must be subject to meaningful
and independent review. These conditions constrain the liberty of states to
deter or prevent migrants from leaving another country and ordain that such
activity is grounded in norms of procedure and good administration. Although
persons fearing ill-treatment or persecution who are restricted in their right
to leave may also base a claim directly on one of the prohibitions of refoulement
(or the underlying prohibition of exposure to ill-treatment), the fact that the
right to leave may be invoked by anyone, regardless of protection entitlements,
implies that it engenders a general procedural framework for employing
measures of external migration control which have the potential to deprive
persons of the factual possibility to leave the country in question.