Page 1

Chapter 1
The Nature and History of International Law

John Dugard
International law may be defined as a normative body of rules and principles which are
binding upon states in their relations with one another. [1] Rules may broadly be divided
into general and particular rules. The rule that the high seas are open to the shipping of
all nations is a general rule of international law binding on all states. A particular rule of
international law is created by a treaty establishing a relationship between two or a few
states only. For instance, the extradition agreement between South Africa and Malawi
providing for the reciprocal return of fugitives from justice binds these two states only.
Principles reflect the values upon which the international order is based and emphasise
that international law is a normative system.
Early international law concerned itself with states only. Now there are other actors
on the international stage. Since 1949, it has been accepted that international
organisations, such as the United Nations and its specialised agencies, enjoy
international legal personality. The recognition of the legal personality of international
organisations was the result of an advisory opinion of the International Court of Justice
in response to the question whether the United Nations could sue Israel for the death of
Count Bernadotte of Sweden, a UN mediator assassinated while on duty in Palestine.
The International Court held that the United Nations had the necessary legal personality
to bring an action against a state in such circumstances. The Court stated
[t]hat is not the same thing as saying that it is a state, which it certainly is not, or that its legal
personality and rights and duties are the same as those of a state. Still less is it the same thing
as saying that it is ‘a super-state’ . . . . What it does mean is that it is a subject of international
law and capable of possessing international rights and duties, and that it has capacity to
maintain its rights by bringing international claims. [2]
Since World War II numerous treaties have been signed extending the protection of
international law to individuals. These human
Page 2

rights treaties impose obligations of varying kinds upon signatory states to afford
protection to their own citizens. In this way, millions of people in the modern world
have become the beneficiaries of international law. [3] Individuals benefit from the
protection of international law and participate in its processes. They may also be
prosecuted for international crimes. They cannot, however, be described as full subjects
of international law.
Multinational corporations are busily engaged in international transactions with
states. Consequently, new rules of law have been developed to cover these
relationships. Although rules of international law may sometimes govern the
relationship between states and corporations, these corporations fail to qualify as
international subjects.
Other entities play an increasing role in international affairs. Non-governmental
organisations (NGOs), such as the International Committee of the Red Cross [4] and
Amnesty International, national liberation movements, and indigenous peoples
participate in the activities of the international community and appeal to international
law to advance their interests.
Although entities other than states participate in the contemporary international
legal order, it is essential to recall that states and inter-governmental organisations are
the main actors in the international community, the only entities with true international
personality and the principal creators of rules of international law.
Much has been written about the sovereignty of states. Unfortunately, the concept of
sovereignty conjures up images of supremacy and absolute power which international
law seeks to limit in the interests of the community of states. For this reason, it is more
accurate, and wiser, to speak of the ‘independence’ of states rather than of their
‘sovereignty’.
Public international law, the subject of this study, must be distinguished from private
international law. Public international law governs the relations between states. It
comprises a body of rules and principles which seek to regulate relations between
states. Private international law concerns the relations between individuals whose legal
relations are governed by the laws of different states. If a South African man marries a
German woman in India, the question will arise as to which system of law governs the
validity of their marriage and their property relations. The system of law that selects or
chooses
Page 3

the appropriate law is known as private international law, or conflict of laws. [5]

Differences between International Law and Municipal Law

International law is generally studied by lawyers with some knowledge of their own
national or municipal system of law. They see the existence of a law-making body, an
executive or law-enforcement body, and a compulsory system of courts as essential
features of any legal system. When they discover that international law lacks developed
institutions of this kind, they inevitably question the very existence of international
law. [6] At the outset, therefore, it is necessary to address the question whether
international law does have a law-making body, an executive power and a judicial
system.

1 Legislature
There is no central legislative body in international law with the power to enact rules
binding upon all states. The General Assembly of the United Nations is only empowered
to adopt recommendations that are not binding upon member states. Although the
Security Council may take decisions in terms of article 25 of the UN Charter binding on
all member states of the United Nations, action of this kind is limited to situations
determined by the Security Council to threaten international peace and security; and in
practice the Security Council is seriously restrained from making such determinations
by the veto power vested in each of the five permanent member states of the United
Nations (China, France, the United Kingdom, Russia and the United States of America).
The United Nations therefore cannot be described as an international legislature. [7]
The rules of international law are to be found in agreements between states, known
as treaties, and in international custom. These rules are not imposed from above by
any central law-making body.
Page 4
Instead they are created by the consent of states. Whereas municipal law operates
vertically, with rules imposed from above, international law is a horizontal system in
which lawmaker and subject are the same legal persona.

2 Executive authority
There is no central executive authority with a police force at its disposal to enforce the
rules of international law. Again, the United Nations comes closest to being such an
executive body, but it falls short of the domestic model on closer analysis. The United
Nations is not a world government: it lacks the power to direct states to comply with
the law, and it lacks a permanent police force to punish violators of the law. Where a
state’s conduct threatens international peace, the Security Council may direct it to
comply with its obligations under international law, but during the Cold War period of
1946–1990 this was a rare occurrence, as the veto power was generally exercised to
prevent any action being taken against a state for non-compliance. When the Cold War
came to an end there was a brief respite, reflected in the Security Council authorisation
of collective action by states in the Gulf War of 1991 to force Iraq to withdraw from
Kuwait. In more recent times the exercise of the veto power by the United States,
Russia or China has in most instances prevented collective action of this kind.
The United Nations is able to raise forces to police certain situations—as in the case
of the United Nations Transitional Assistance Group in Namibia (UNTAG), established in
1989/90 to supervise elections in Namibia; the United Nations Protection Force in
Yugoslavia (UNPROFOR), created in 1992 to oversee the cease-fire and control disputed
areas in the former Yugoslavia; and the United Nations Organization Mission in the
Democratic Republic of the Congo (MONUC), established to keep the peace in the Great
Lakes region. Alternatively, it may authorise member states to take action on its
behalf—as in the case of the Gulf War of 1991 and the civil war in Libya in 2011. But
there is no permanent force at the disposal of the Security Council that may be sent to
restore peace and order.
Before the Charter system came into existence in 1945, states frequently sought to
enforce the rules of international law by means of self-help. An aggrieved state would
take the law into its own hands and punish a transgressor state. The Charter of the
United Nations in article 2(4) now prohibits the use of force against states, except in
the exercise of the right of self-defence or under the
Page 5

authorisation of the United Nations. The result is that international law has lost one of
its instruments of law enforcement. [8]

3 International courts
Today there are a number of international courts. The best known is the International
Court of Justice in The Hague, which may be used to settle disputes between all states
in the world. In addition, there are a number of regional, specialised courts, such as the
European Court of Human Rights, which has jurisdiction over disputes arising out of the
European Convention on Human Rights. International law therefore does have a judicial
system capable of ruling on disputes between states. But there is an important
difference between international courts and domestic courts: international courts have
jurisdiction only over those states that have consented to their jurisdiction.
The International Court of Justice (known as the Permanent Court of International
Justice from 1920 to 1945) was created in 1920, but compared with any domestic
tribunal it has heard very few cases. Moreover, disputes which are referred to the Court
by consent are often minor disputes that do not threaten relations between states. For
instance, in the Kasikili/Sedudu Island Case [9] Botswana and Namibia referred a
dispute over the boundary between the two countries around Kasikili/Sedudu Island
and the legal status of the island—a matter which did not affect the vital interests of
either country—to the International Court of Justice. Sometimes disputes involving the
vital political interests of states are brought before the International Court of Justice on
the basis of consent conferred at an earlier time. In such cases, the Court is seldom
able to establish its authority over the parties to the dispute. For instance, when
Nicaragua brought a dispute against the United States in 1984, claiming that the United
States had mined its harbours and intervened militarily in Nicaragua, the United States
refused to appear before the Court—despite the fact that it had consented to disputes
between the two states being referred to the International Court of Justice in an earlier
treaty—and it repudiated the finding of the Court on the merits of the dispute. [10] The
International Court of Justice therefore cannot be seen to be a court suitable for
resolving serious political disputes between nations. Moreover, most states refuse to
consent to the jurisdiction of the International Court of Justice. Only a
Page 6

minority of states in the modern world are prepared to take full advantage of the
International Court of Justice for the purpose of settling legal disputes.
The International Court of Justice is also competent to give advisory opinions on
matters of concern to the United Nations at the request of the political organs of the
United Nations. These opinions carry considerable weight as statements of the law. For
example, the International Court gave four opinions on the legal status of South West
Africa/Namibia, which established the law to be applied by the international community
in its treatment of the dispute with South Africa over this territory. [11] More recently, it
rendered an opinion on many of the legal issues concerning the status of the Occupied
Palestinian Territory. [12] However significant such opinions might be, they remain
advisory and are not binding upon states.
International law therefore differs basically from municipal legal systems in that it
has no central legislature or executive authority, and its principal court, unlike domestic
courts, lacks compulsory jurisdiction over its subjects.
States are governed by individuals. Consequently, the punishment of state leaders
for crimes against the international order (particularly war crimes, genocide, crimes
against humanity and aggression) by international courts provides an effective means
for the enforcement of international law. After World War II, tribunals were established
by the victorious parties to try the Nazi and Japanese war leaders in Nuremberg and
Tokyo respectively. Subsequent attempts to create an international criminal court to
prosecute state leaders guilty of international crimes failed. However, in 1993/94 two
ad hoc tribunals, the International Criminal Tribunal for the Former Yugoslavia and the
International Criminal Tribunal for Rwanda, were created to try persons charged with
international crimes arising out of the ethnic conflicts in those territories. In July 1998,
a diplomatic conference in Rome gave its approval to the establishment of a permanent
international criminal court, the International Criminal Court, which came into being in
2002. These courts mark an important development in the enforcement of international
law.

The problem of sanctions

The absence of an external central authority with the power to enforce the rules of
international law in a regular and consistent manner, if necessary by means of armed
force, has prompted much of
Page 7

the cynicism about the entitlement of international law to be called ‘law’. Critics tend to
overlook the fact that domestic law enforcement is not always regular and consistent
and that compliance with the law is secured in a variety of ways, of which the threat or
use of force is but one. Moreover, international law does have a number of sanctions for
breach of a rule of law aimed at securing compliance with the law.
The Charter of the United Nations, in Chapter VII, empowers the Security Council to
direct its members either individually or collectively to use force against a delinquent
state whose violation of international law constitutes a threat to international peace.
There are only two clear precedents [13] for such action. In 1966, the Security Council
authorised the United Kingdom to use force on the high seas to prevent oil tankers from
reaching the port of Beira when their oil was destined for Rhodesia. [14] In 1990, the
Security Council authorised member states ‘to use all necessary means’ to secure the
withdrawal of Iraq’s forces from Kuwait, [15] after determining that Iraq’s invasion and
annexation of Kuwait was illegal. [16] This was the first occasion on which the veto
power was not allowed to obstruct the Security Council from taking action against an
aggressor.
Since the end of the Cold War in 1990, the United Nations has established
peacekeeping forces in many countries plagued by internal strife, such as Somalia,
Haiti, the former Yugoslavia and the Democratic Republic of the Congo. On occasion
these peacekeeping forces have engaged in the enforcement of the peace by military
means. Inevitably, this has resulted in the enforcement of rules of international law
relating to human rights, self-determination and humanitarian law. Although, strictly
speaking, these are not measures taken against a state for violation of international
law, it is in practice difficult to distinguish them from law-enforcement action.
Economic sanctions are another enforcement mechanism at the disposal of the
Security Council. A succession of resolutions calling on states to impose comprehensive
economic sanctions was adopted to compel Rhodesia to accept majority rule. [17] South
Africa too was subjected to a wide range of sanctions over apartheid. In 1977, the
Page 8

Security Council imposed a mandatory arms embargo on South Africa as a punishment

for its discriminatory and repressive laws and practices, and its acts of aggression
against neighbouring states. [18] Repeated recommendations of the General Assembly in
support of wider economic sanctions prompted states, individually and collectively, to
isolate South Africa in the fields of trade, finance, sport and culture. Although the effect
of these sanctions was not immediate, they undoubtedly contributed to State President
FW de Klerk’s decision to abandon apartheid in February 1990.
The end of the Cold War witnessed an increase in the number of Security Council
resolutions imposing mandatory sanctions against states. Economic sanctions were
imposed on Iraq following the Gulf War in 1991; [19] Libya was subjected to a
mandatory arms embargo and to a prohibition on air traffic in 1992 because of its
failure to extradite two men suspected of placing a bomb on the ill-fated Pan Am flight
103, which exploded over Lockerbie; [20] the Federal Republic of Yugoslavia (Serbia and
Montenegro) was subjected to economic sanctions for its support to the Bosnian Serbs
in the Bosnian conflict and its repressive actions in Kosovo; [21] an arms embargo was
imposed on Libya in 2011; [22] and economic sanctions have been imposed on North
Korea in response to its nuclear and missile activities. [23]
Exclusion from membership in international organisations is another sanction that
was effectively employed against South Africa for violating its obligations under the UN
Charter. In the 1960s, South Africa was excluded from membership of a number of
specialised agencies of the United Nations—such as the International Labour
Organization (ILO) and the World Health Organization (WHO)—and in 1974, the
General Assembly of the United Nations excluded South Africa from participation in the
debates and work of the General Assembly. [24] In 1994, South Africa resumed its
membership of these bodies.
Non-recognition of a territorial adjustment must also be seen as a sanction. The
Security Council has directed states not to recognise the Turkish Republic of Northern
Cyprus as a state on the ground that its creation violated rules of international
law. [25] In addition, it has
Page 9

labelled Israel’s annexation of the Golan Heights and East Jerusalem illegal acts that
should be condemned by non-recognition. [26] During the apartheid era, Transkei,
Bophuthatswana, Venda and Ciskei (the ‘TBVC states’) were subjected to non-
recognition at the direction of the Security Council. [27] There is little doubt that non-
recognition has seriously obstructed the statehood of the Turkish Republic of Northern
Cyprus, and undermined Israel’s claims to the Golan Heights and East Jerusalem. It is
also clear that non-recognition of the ‘TBVC states’ played a major part in the
delegitimation of the apartheid state.
The punishment of individuals for international crimes, such as war crimes, piracy,
hijacking and crimes against humanity, is a sanction of international law in the best
tradition of the municipal criminal offence and provides evidence to even the most ill-
informed observer that international law is sometimes enforced by punishment.
International law is not, therefore, without sanctions. However, it must be conceded
that sanctions of this kind lack the comprehensiveness, regularity and consistency
associated with sanctions in domestic law.

Is International Law really law?

John Austin’s claim that law is the command of a political superior to a political inferior
backed by the threat of a sanction is largely discredited today, as is his conclusion that
international law fails to qualify as law. His principal 20th century disciple, HLA Hart,
has rejected the command theory of law and accepted that international law qualifies as
a species of ‘law’. [28] Few serious jurists insist on effective sanctions as a requirement
for the existence of international law. [29]
The cynic may remain unconvinced and ask what—if not a legislature, executive,
compulsory court system and sanctions—are the essential requirements of a system of
law? The most satisfactory response to this question is that of Sir Frederick Pollock;
namely, that a legal system requires the existence of a political community, and the
recognition by its members of settled rules binding upon them. [30]
Judged by these standards, international law certainly qualifies as a system of law.
First, there is a political community, namely the
Page 10
community of modern states, ‘over 193’ in number. Although there may be serious
political, economic and cultural divisions within this community, it is probably no less
divided than many heterogeneous societies. Secondly, there is a body of rules and
principles that comprise the international legal order. Thirdly, the members of the
international community recognise these rules and principles as binding upon them.
This is not to deny that international law is sometimes violated in the most brutal
manner and that such violations are sometimes allowed to go unpunished, particularly
when committed by a major power. Violations of this kind, however, are the exception
and should not be allowed to breed a general cynicism of the role of law in international
society. [31] International law is observed and honoured every day in diplomatic and
consular relations, trade, air and sea traffic, communications, extradition, etc. [32] On
the whole, it is probably observed as much as municipal law is in some societies.
Moreover, its binding quality is recognised by international and municipal courts
(including those of South Africa) [33] and by statesmen and diplomats in their daily
inter-state dealings. No national leader has repudiated the existence of international
law since Adolf Hitler. Others may have violated international law, but they have been
careful to raise some legal pretext for their action. Saddam Hussein, for example,
justified his invasion of Kuwait on 2 August 1990 on the ground that, historically,
Kuwait was a part of Iraq as they had both belonged to the Ottoman Empire before
World War I. However unfounded such an argument may have been in law and
fact [34] it was at least an acknowledgement of the existence of international law.
States acting through their governments recognise and comply with international law
for a wide range of reasons. These include an interest, either selfish or altruistic, in the
maintenance of peace and good order; an acceptance of the legitimacy of rules of
international law; [35]
Page 11

reputation both at home and abroad; [36] anticipated reciprocal treatment; the
realisation of the need for co-existence; [37] and fear of economic, political, cultural and
sports isolation. Until the Gulf War of 1991, fear of United-Nations-authorised military
sanctions probably contributed very little to the decision on the part of states to comply
with international law; now, this precedent may act as an additional and cogent reason
for compliance.
By and large, states comply with international law for reasons unrelated to the threat
of a sanction. This means that international law is not binding because it is enforced but
that it is enforced because it is already binding. [38] The basis of the international legal
obligation must therefore be found in some source other than the prospect of
enforcement. The debate over the identity of this source is a theme that runs like a
thread through the history of international law. Essentially this debate, as in municipal
law, has taken the form of an exchange over many centuries between naturalists and
positivists.

The history of International Law (including theories of

International Law) [39]
The roots of international law are to be found in the ancient histories of the Egyptians,
Jews, Greeks and Romans, and in the jurisprudence and inter-state practices of the
Muslim world. However, it was not until the emergence of the modern state in the wake
of the Renaissance and Reformation that the need arose for a law to regulate relations
between states. This was confirmed by the Peace of Westphalia of 1648, which brought
to an end the Thirty Years War and the authority of the Holy Roman Emperor. Before
this time, forces such as feudalism and the Holy Roman Empire, institutionalising the
power of the Catholic Church, had obstructed the growth of
Page 12

territorial units with strong centralised governments—ie sovereign, independent

states—and obviated the need for a law between nations. International law as we know
it today is therefore less than five hundred years old.
In both Roman and medieval times the scholarly jurist had played a key role in the
development of the law. It was therefore predictable that jurists would be responsible,
initially, for the formulation of the principles of law that were to govern states. Neither
is it surprising that their main sources in this endeavour would be the ius gentium of
Roman law, which had exercised a powerful influence on the law of Western Europe and
the precepts of natural law that had been absorbed into Canon law under the
inspiration of philosopher-theologians such as St Thomas Aquinas.
Early writers on international law came from Spain (Vittoria and Suarez) and Italy
(Gentilis, who fled to England to avoid religious persecution). Power then shifted from
Spain to the Netherlands, which occupied a key position in the economic and
intellectual life of Europe throughout the 17th and early 18th centuries. It was perhaps,
therefore, predictable that the Netherlands should produce the ‘father’ of international
law: Grotius, or Hugo de Groot (1583–1645). [40]
Grotius is acclaimed as the ‘father of international law’ largely in recognition of his
monumental treatise De Jure Belli ac Pacis Libri Tres. Its commitment to an
international legal order, transcending the interests of the nation state, gives it a
universal appeal that is lacking in Grotius’s earlier works on international law. Mare
Liberum, published in 1609, and the larger work of which it formed a part, De Jure
Praedae, [41] is chauvinist in outlook and preoccupied with the advocacy of the maritime
interests of the Netherlands. [42]
De Jure Belli ac Pacis was first published in 1625, in the midst of the Thirty Years’
War (1618–1648). This work is the starting point of modern natural law as, until its
appearance, natural law was so closely coupled with divine law that the two were
treated as being almost synonymous. Grotius, himself a devout Christian and
recognised theologian, took the extraordinary step—at least for his
Page 13

time—of emancipating natural law from theology. To Grotius ‘[t]he law of nature is a
dictate of right reason’, [43] which impels man to seek a peaceful and organised society,
and whose validity would hold ‘even if we should concede that which cannot be
conceded without the utmost wickedness that there is no God, or that the affairs of
men are of no concern to Him’. [44]
Drawing on the ‘dictate of right reason’, as exemplified in the ius gentium and the
writings of scholars, Grotius sought to construct a just international legal order whose
principal aim was the restraint of war. As the first comprehensive and systematic
treatment of international law integrated within the structure of a general system of law
and jurisprudence, it had immediate success. Today it is still regarded as the bible of
international law.
Another Dutch jurist, Cornelius van Bynkershoek (1673–1743), had a profound effect
on the development of international law. A distinguished judge, with great experience in
maritime and commercial practice, he wrote a number of works on special topics [45] in
which he stressed the importance of consent, in the form of custom or treaty, as the
basis of international law. Consequently, he was one of the early positivists for whom
consent, rather than the principles of natural law, explained the international legal
obligation.
Other scholars of this ‘classical period’ of international law contributed substantially
to the growth of international law. Of these, Zouche (1590–1660), Professor of Civil
Law at Oxford; Pufendorf (1632–1694), Professor at Heidelberg; and Vattel (1714–
1769), a Swiss who served in the diplomatic service of Saxony, are the best known.
The Dutch contribution to the development of international law through Grotius and
Van Bynkershoek was significant. Not only did they shape the new international legal
order, they also set the scene for the naturalist versus positivist debate that continues
today.
International law of this time was in reality an international law of Christian Europe.
On the positive side, international law drew on the values and intellect of the scholars
of the Renaissance and Enlightenment. On the other hand, there is no doubt that some
of the rules of the new legal order were simply rationalisations of the expansionist
colonial and imperialist policies of the European powers and their commitment to the
slave trade. [46] As late as 1884 the European
Page 14

colonial powers divided Africa among themselves in the Berlin Conference under the
guise of international law. Relics of colonialism’s rules persist to this day in international
law, particularly in the law governing statehood and the acquisition of territory and in
colonial boundaries which have become the frontiers of the modern states of Africa.
It was only in the late 18th century that non-European and later non-Christian
nations were admitted to the ‘club’ of Christian European nations whose relations were
governed by the international legal order fathered by Grotius. The admission of the
United States and independent South American republics to the European ‘club’ was
followed by the acceptance of Turkey, Japan, China, Persia (Iran) and Siam (Thailand).
The creation of the League of Nations in 1920 saw the extension of membership in
the ‘club’ to other nations, including South Africa. However, the League of Nations was
largely Eurocentric in its membership and its concerns, as indeed was the United
Nations in its early years. Since the decolonisation of European colonies in Africa, Asia,
the Pacific and the Caribbean, the position has changed dramatically. Today, the
community of nations is genuinely ‘international’ and numbers over 193 states, of
which nearly all are members of the United Nations. This diversity in the composition of
the international community has inevitably given rise to tensions within the
international legal order. For some 40 years, the Soviet Union propagated a communist
theory of international law, [47] which showed a distrust of bourgeois institutions (such
as the International Court of Justice) and some of the rules of ‘Western’ international
law—notably, the rules requiring compensation for the expropriation of foreign
investment. Developing states are hostile to those rules of international law most
closely associated with colonialism [48] and stress the importance of self-determination.
Although different approaches of this kind create difficulties, they should not be
exaggerated, as in many respects they resemble the problems faced by multicultural
national societies.
Jurisprudentially, the debate between naturalists and positivists remains unresolved.
While natural law was pre-eminent during the 17th and 18th centuries, it was replaced
in popularity by positivism in the 19th and early 20th centuries. At the national level,
positivism took the form of the command theory of law, while in international
Page 15

law, positivists proclaimed consent as its basis. Indeed, under the influence of the
German philosopher Hegel, the idea developed that a state was bound only by those
rules to which it had clearly consented, and thereby voluntarily restricted its own
sovereignty.
The logic of positivism was taken to its extreme form in Nazi Germany. National law
became the expression of the will of the Führer and international law was completely
subordinated to the will of the state, represented by the Führer. European statesmen
educated in the positivist belief that the manner in which a sovereign state treats its
own nationals is essentially its own business were legally restrained from intervening to
halt the holocaust. This experience starkly illustrated the dangers of legal positivism,
and after the war, this philosophy of law was discredited. [49] The trials of Nazi war
criminals by the Nuremberg International Tribunal and by post-war German municipal
courts testified to the existence of certain basic principles of justice superior to Nazi
law. Moreover, the recognition in the Charter of the United Nations of the need for
states to promote and respect human rights heralded an abandonment of the notion of
the absolute sovereignty of a state over its own nationals. The powerful international
human rights movement, which plays a major role in modern international law, is
largely inspired by the idealism of natural law.
Today, new theories of law seek to explain the nature of international law and to
shape its evolution. In the United States, a sociological movement emerged that draws
heavily on the notions of social engineering advocated by Roscoe Pound and the realist
movement. This policy-oriented school, associated with the names of Myres McDougal,
Harold Lasswell and Michael Reisman of the Yale Law School, [50] sees law as a
comprehensive process of authoritative decision-making by a wide range of actors and
not as a defined set of rules. The promotion of values, such as power, wealth,
enlightenment, skill, well-being, affection, respect and rectitude, plays an important
part in this process. More recently, members of the critical legal studies movement
have analysed traditional international law
Page 16

and highlighted the extent to which international politics dominates the formulation of
both rules and theory. [51]
International law has paid little attention to the needs of women and international
law institutions have been dominated by men. Only four women have been elected to
the International Court of Justice, [52] and women were first elected to the International
Law Commission in 2001. This has prompted a feminist appraisal of international
law, [53] which has already had an impact on the substance of international law—for
instance, in the recognition of sexual crimes in time of war and in the composition of
international bodies.
Most theories of international law seek to advance the international legal order by
explaining or analysing the nature of international law. However, in the United States, a
neo-conservative school has sought to question the value of international law by
arguing that international law is subordinate to the national interest. [54] This school,
which gained prominence during the presidency of George W Bush and seems to
explain the ‘America first’ policy of Donald Trump, has undermined the commitment of
the United States to its international obligations. Significantly, this legal theory ‘finds
virtually no echo among legal scholars outside the United States’. [55]
None of these schools of jurisprudence on its own can satisfactorily explain the
nature of international law and the source of the international legal obligation. While
notions of justice and the values of legal idealism associated with natural law form the
foundation of much of contemporary international law, [56] particularly the promotion of
human rights and the right of self-determination, it cannot be denied that for many
states consent remains the basis of their participation in the international community.
Consent on its own, however, fails to provide an explanation for the rules and principles
that comprise international law. Third World states, for instance, have at no time
expressly consented to the rules that shaped international law before they attained
independence. Indeed, as consent becomes more difficult to obtain for the creation of
new rules
Page 17

of law, consensus in the form of majority decision-making is increasingly adopted.

Sociological and critical theories of law have a useful role to play. International law—
like municipal law—has no autonomy as a discipline and must be seen as an integral,
and important, part of the international political process. Traditionalists ignore this truth
at their peril.
The sources of international law are too diverse and the functions it seeks to achieve
are too complex to permit a single explanation of its nature and existence. There is a
measure of truth in each of the theories advanced to provide an understanding of the
nature of international law. Rules and principles of law constitute the flesh, blood,
organs and bones of international law. Like physicians, we must study the parts of this
body. But the mind of international law is more difficult to comprehend.

The fragmentation of International Law

Today, international law accommodates many branches, each with its own special rules.
There are special legal regimes governing, for instance, the sea, trade, environment,
human rights, humanitarian law and foreign investment. In addition, many of these
branches have their own special courts to pronounce on these special rules. Thus, there
is an International Tribunal for the Law of the Sea to adjudicate on maritime disputes,
an International Criminal Court to judge issues of humanitarian law, several regional
human rights courts to consider violations of human rights law and a Dispute
Settlement Body of the World Trade Organization to decide matters of trade. The
proliferation of tribunals and the growth of autonomous branches of the law have led to
fears that international law is in a process of fragmenting into separate structures, so-
called self-contained regimes that threaten the universality of international law. This
matter was subjected to careful examination by the United Nations International Law
Commission, [57] which concluded that the emergence of special treaty regimes, and the
proliferation of tribunals, had not seriously undermined the unity of international law.
The present treatise bears this out. While there are undoubtedly special treaty regimes,
the general rules of international law remain intact and provide a framework within
which the special treaty regimes operate. Moreover, the problem of different rulings
from different courts is much exaggerated. Tribunals, in large measure, defer to the
jurisprudence of the International Court of Justice.

[1] See G Simpson (ed) The Nature of International Law (2001).

[2] Reparation for Injuries Suffered in the Service of the United Nations 1949 ICJ Report 174 at 179.
[3] See Chapter 15.
[4] G Abraham ‘Yes, . . . but does it have personality? The International Committee of the Red Cross
and sovereign immunity’ (2007) 124 SALJ 499.
[5] See, on this system of law in South Africa, C Forsyth Private International Law: The Modern Roman-
Dutch Law including the Jurisdiction of the High Courts 5 ed (2012).
[6] English legal philosopher John Austin preferred to describe international law as ‘positive morality’
for this reason: see Province of Jurisprudence Determined (1954) 133, 201. Criticisms of this kind were
prevalent in Grotius’s days. In the opening passages of his Prolegomena to De Jure Belli ac Pacis (1625)
Grotius writes that such a work on international law is ‘all the more necessary because in our day, as in
former times, there is no lack of men who view this branch of law with contempt as having no reality
outside an empty name’: para 3 (translation by FW Kelsey in J Brown Scott (ed) The Classics of
International Law (1925)).
[7] See Chapter 23. Cf P Szasz ‘The Security Council starts legislating’ (2002) 96 AJIL 901.
[8] International law does, however, recognise the lawfulness of non-forcible self-help measures—
'countermeasures’—taken by an injured state against a wrongdoing state to compel compliance with
international law. See Chapter 13.
[9] 1999 ICJ Reports 1045.
[10] See, generally on this subject, ‘Appraisals of the ICJ’s decision: Nicaragua v United States (Merits)’
(1987) 81 AJIL 77–183.
[11] See Chapter 22.
[12] Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory 2004 ICJ
Reports 136; (2004) 43 ILM 1009.
[13] On 27 June 1950, the Security Council recommended to states that they assist South Korea to
repel an armed attack by North Korea (SCOR, 5th year, Resolutions and Decisions 5). The absence of the
Soviet Union from the Security Council when this resolution was ‘passed’ raises questions about the
validity of this precedent. See Chapter 23.
[14] Resolution 221 (1966).
[15] Resolution 678 (1990).
[16] Resolution 662 (1990).
[17] See, in particular, resolutions 232 (1966) and 235 (1968).
[18] Resolution 418 (1977). See Chapter 23.
[19] Resolutions 661, 665, 678 (1990).
[20] Resolution 748 (1993).
[21] Resolutions 757 (1992), 820 (1993), 942 (1994).
[22] Resolutions 1970, 1973 (2011).
[23] See, for example, resolution 2375 (2017).
[24] See Chapter 23.
[25] J Dugard Recognition and the United Nations (1987) 108–11.
[26] Dugard (n 25) 111–15.
[27] Ibid 98–108.
[28] The Concept of Law (1961) ch 10.
[29] R Higgins Problems and Process: International Law and How We Use It (1994) 13–16.
[30] First Book of Jurisprudence (1929) 28. See, too, R Jennings and A Watts (eds) Oppenheim’s
International Law 9 ed (1992) vol 1 at 9–13.
[31] Both Hans Morgenthau (eg, in ‘Positivism, functionalism, and international law’ (1940)
34 AJIL 260) and George Kennan (American Diplomacy 1900–1950 (1951) 959) base much of their
scepticism about international law on its inability to play a meaningful role in power relations between the
major states.
[32] According to Henkin, ‘almost all nations observe almost all principles of international law and
almost all of their obligations almost all of the time’: How Nations Behave—Law and Foreign Policy 2 ed
(1979) 47.
[33] See Chapter 4.
[34] RV Pillai and M Kumar ‘The political and legal status of Kuwait’ (1962) 11 ICLQ 108.
[35] TM Franck The Power of Legitimacy among Nations (1990); TM Franck Fairness in International
Law and Institutions (1995).
[36] Both domestic and international public opinion may induce compliance with international law.
South Africa abandoned its policy of apartheid in 1990, in part as a result of hostile international opinion.
The United States was compelled to withdraw from Vietnam in 1974, in part as a result of domestic
opposition to the war.
[37] G Tunkin ‘Co-existence and international law’ (1958–III) 95 Recueil des Cours 1.
[38] See G Fitzmaurice ‘The foundations of the authority of international law and the problem of
enforcement’ (1956) 19 Modern LR 1.
[39] See A Nussbaum A Concise History of the Law of Nations (1947); JW Verzijl International Law in
Historical Perspective 10 vols (1968–1979); M Koskenniemi The Gentle Civilizer of Nations: The Rise and
Fall of International Law 1870–1960 (2002); D Bederman International Law in Antiquity (2001); B
Fassbender and A Peters (eds) The Oxford Handbook of the History of International Law (2012); A
Orakhelashvili (ed) Research Handbook on the Theory and History of International Law (2011); A Orford
and F Hoffman (eds) The Oxford Handbook of the Theory of International Law (2016); H Strydom and C
Gevers ‘Defining Epochs’ in H Strydom (ed) International Law (2016).
[40] J Dugard ‘Grotius, the jurist and international lawyer: Four hundred years on’ (1983)
100 SALJ 213; E Kahn ‘Hugo Grotius, 10 April 1583–29 August 1645: A sketch of his life and his writings
on Roman-Dutch law’ (1983) 100 SALJ 192; P van Warmelo ‘Hugo Grotius 10th April 1583–28th August
1645’ (1983) 9 SAYIL 1.
[41] De Jure Praedae was not published during Grotius’s lifetime. It was discovered in 1864 and
published in 1868.
[42] F de Pauw in Grotius and the Law of the Sea (1965) accuses Grotius of adapting the ‘divine rules
of the law of nature to the economic interests of the Dutch East India Company’ (at 61).
[43] De Jure Belli ac Pacis 1.1.10.1 (translation by F W Kelsey in J Brown Scott (ed) The Classics of
International Law (1925)).
[44] Idem Prolegomena para 11.
[45] De Domino Maris (1702), De Foro Legatorum (1721) and Quaestionum Juris Publici (1737). See
the review of the last treatise in (2010) 23 Leiden Journal of International Law 269–71.
[46] See A Anghie Imperialism, Sovereignty and the Making of International Law (2005).
[47] G Tunkin Theory of International Law (1974); K Grzybowski Soviet Public International
Law (1970); H Chiu ‘Communist China’s attitude towards international law’ (1966) 60 AJIL 245.
[48] See RP Anand ‘Attitude of the Asian-African states towards certain problems of international law’
(1966) 15 ICLQ 55; TO Elias New Horizons in International Law (1980).
[49] For an illuminating exchange of ideas on this subject, see HLA Hart ‘Positivism and the separation
of law and morals’ (1958) 71 Harvard LR 593 and LL Fuller ‘Positivism and fidelity to law—a reply to
Professor Hart’ (1958) 71 Harvard LR 630.
[50] See, for example, McDougal, Lasswell and Reisman ‘Theories about international law: Prologue to
a configurative jurisprudence’ (1968) 8 Virginia Journal of International Law 188; Higgins (n 29) ch 1.
[51] See D Kennedy International Legal Structures (1987); M Koskenniemi From Apology to
Utopia (1989); P Allott Eunomia (1990); S Marks The Riddle of All Constitutions: International Law,
Democracy and the Critique of Ideology (2000).
[52] Rosalyn Higgins was President of the Court from 2006 to 2009.
[53] See H Charlesworth and C Chinkin The Boundaries of International Law: A Feminist
Analysis (2000); H Charlesworth, C Chinkin and S Wright ‘Feminist approaches to international law’
(1991) 85 AJIL 613.
[54] JL Goldsmith and E Posner The Limits of International Law (2005); M Glennon Limits of Law,
Prerogatives of Power: Interventionism after Kosovo (2001).
[55] TM Franck ‘The power of legitimacy and the legitimacy of power: International law in an age of
power disequilibrium’ (2006) 100 AJIL 88, 90.
[56] TM Franck Fairness in International Law and Institutions (1995).
[57] Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of
International Law, Report of the study group of the International Law Commission (finalised by M
Koskenniemi) A/CN.4/L682 (2006).
 Page 18

Chapter 2
South Africa and International Law: A
Historical Introduction

John Dugard
International law is the product of the European state system that came into being in
the 16th century. [1] Until the 19th century it was in reality a European law of nations.
This system of law, rooted in convictions of European superiority, imperialism and
colonialism, accorded little recognition to the political organisms of Africa, whatever
their level of sophistication. [2] Indeed, territories occupied by non-European peoples
not constituting a social or political aggregation were treated as belonging to no one—
as terra nullius. [3] The history of international law in southern Africa therefore begins
with the first European settlement. This is an unashamedly Eurocentric view, but
international law, until the last century, was unashamedly Eurocentric.
The Cape of Good Hope was settled by the Dutch in 1652 when Jan van Riebeeck
took possession of the Cape as a refreshment station for the Dutch East India
Company, the colonising instrument of the Republic of the United Netherlands. [4] Thus
began an association between South Africa and Roman-Dutch law that continues to this
day. [5] Between the 16th and 18th centuries, the Netherlands was a major maritime
and mercantile power whose lawyers were busily involved in international affairs. It is
not surprising therefore that the Netherlands was in the forefront of the development of
international law. [6] Several Roman-Dutch jurists wrote on international law topics and
two of them—Grotius and Van Bynkershoek—achieved great
Page 19

prominence in this field. [7] Roman-Dutch law and international law consequently
evolved together in the Netherlands, and jurists made little attempt to separate the two
legal orders. [8]
As a colony, first of the Netherlands (1652–1795, 1803–1806) and then of Britain
(1795–1803, 1806–1910), the Cape of Good Hope was not an independent actor in the
international community. Acts governed by international law were undertaken on its
behalf by the colonial power. Treaties were extended to the Cape, [9] territory was
annexed [10] and agreements were signed with native chiefs on a wide range of
subjects—including boundaries, sovereignty over conquered lands, the extradition of
fugitives and the protection of tribes. [11] Whether these agreements with native chiefs
were seen as international agreements at the time is highly doubtful, but today there is
support for the view that they were in some cases governed by principles of
international law. [12]
The annexation of Natal by Britain in 1843 added another South African colony to the
Empire, whose international affairs were conducted along the same lines as those of the
Cape of Good Hope. The creation of the Boer Republics introduced a new dimension to
the state system of southern Africa. Although Britain recognised the South African
Republic (Transvaal) and the Orange Free State as independent states in 1852 and
1854 respectively, it later imposed such severe restraints on the treaty-making power
of the Transvaal, after its annexation of that territory from 1877 to 1881, that there
were serious doubts about the independence of the South African Republic. While
Britain took the view that it was not an independent state, many writers contended that
the Pretoria Convention of 1881 and the
Page 20

London Convention of 1884 did not deprive the Transvaal of its independent
statehood. [13]
Despite doubts about the independence of the South African Republic, the Anglo-
Boer War (1899–1902) was viewed by all states—including the United Kingdom—as a
war between sovereign independent states. [14] Until the formal annexation of the
Orange Free State and Transvaal, in May and October 1900 respectively, the Boer
forces were recognised by Britain as lawful belligerents, entitled to be treated as
prisoners of war on their capture. [15] Both parties to the conflict claimed to act in
accordance with the customs and usages of the law of war, and municipal courts judged
many disputes according to the customary laws of war. [16] The annexation of the Boer
Republics in 1900 occurred before Britain had established military control over the rural
areas of the republics, which gave rise to the claim that Britain’s annexation had been
premature and contrary to international law. [17] For two years a bitter guerrilla war was
waged. Although Britain treated the Boer guerrillas as lawful belligerents, their families
were herded into concentration camps and their properties devastated. Serious
questions were raised about the legality of the British action under international
humanitarian law, then in its infancy.
In 1910 the Union of South Africa was formed. In matters of foreign policy, the
Imperial Government continued to act on behalf of the Union. [18] Constitutionally, the
South African Parliament remained subject to the authority of Westminster in terms of
the Colonial Laws
Page 21

Validity Act. [19] Consequently, there was no suggestion that South Africa or the other
dominions were independent states. [20] In World War I they fought loyally and
effectively as part of the Imperial Forces. This contribution did not go unnoticed by the
international community and the dominions were substantially rewarded after the War.
South Africa, Australia and New Zealand were given responsibility for the mandated
territories of South West Africa, New Guinea, Samoa, and Nauru respectively, for whose
administration they were directly accountable to the League of Nations. [21] Moreover,
South Africa, together with Australia, Canada, New Zealand and India, were admitted to
original membership of the League of Nations despite the fact that constitutionally they
remained subordinate to the British Crown. Although their precise status was much
debated, [22] it is clear that they obtained ‘for all League purposes a definite position as,
for these matters at least, states of international law’. [23]
The uncertain international status of the dominions did not continue for long. In
1926 an Imperial Conference resolved that Britain and the dominions were ‘autonomous
communities within the British Empire, equal in status, in no way subordinate one to
another in any aspect of their domestic or external affairs though united by a common
allegiance to the Crown, and freely associated as members of the British
Commonwealth of Nations’. [24] This resolution was given legislative endorsement in
1931 when the British Parliament passed the Statute of Westminster, [25] which
repealed the Colonial Laws Validity Act and provided that in future, no Act of the British
Parliament would extend to a dominion without the latter’s consent. Thus, by 1931 it
was clear beyond all doubt that South Africa was a sovereign independent state, a full
subject of international law. Domestically, this new status was confirmed by the Status
of Union Act [26] and the Royal Executive Functions and Seals Act. [27]
Page 22

The role played by South Africa in the League of Nations [28] was very different from
that which it was later to play in the United Nations from 1946 to 1994. Although the
League was largely concerned with European affairs, South Africa made a positive
contribution to the maintenance of international peace and gave its full support to the
League’s collective measures. In fact, South Africa was one of the foremost advocates
of collective sanctions against Italy following Italy’s invasion of Ethiopia. In recognition
of its contribution to the League, a South African—Mr Charles te Water—was elected
President of the Assembly of the League of Nations in 1933, and in 1939, in the twilight
of the League, South Africa was elected a member of the Council. Despite its close
relationship with Britain during the inter-war period, South Africa pursued an
independent line in the League of Nations and on occasions seemed to use its position
in the League as a means of asserting its independence from Britain.
In 1945 the United Nations was established and in the following year the League of
Nations was dissolved. South Africa played a prominent part in the creation of the
United Nations. General Smuts, the South African Prime Minister, enjoyed considerable
international prestige [29] and was in part responsible for the drafting of the preamble of
the United Nations Charter, which reaffirms ‘faith in fundamental human rights, in the
dignity and worth of the human person, in the equal rights of men and women and of
nations large and small’. [30] From the outset, however, South Africa was compelled to
play a defensive role as a result of its racial policies. [31] Initially South Africa was
attacked for its discriminatory treatment of Indians. Later, after the National Party
came to power on the platform of apartheid, South Africa became a symbol of racial
injustice in a world committed to racial equality and decolonisation. South Africa’s
protests that her racial policies were a domestic issue that fell outside the jurisdiction of
the United Nations at first received the support of many Western states, but after the
police shooting of peaceful demonstrators at Sharpeville in 1960, this support
disappeared and apartheid was regarded as a matter of international concern, as a
Page 23

violation of the clauses in the United Nations Charter promoting human rights, [32] and,
later, as a crime against humanity. In the ensuing 30 years, South Africa became a
pariah state against which a wide range of United Nations sponsored measures were
taken, including a mandatory arms embargo in 1977 and exclusion from participation in
the General Assembly of the United Nations in 1974. [33]
South Africa’s contribution to the development of international law during this period
was enormous, although unintended. New rules of treaty and customary law to promote
human rights, racial equality and decolonisation evolved as a result of international
opposition to apartheid. [34] The prohibition on interference in the domestic affairs of
states, enshrined in article 2(7) of the Charter, was substantially weakened by a
succession of resolutions condemning apartheid. [35] Notions of statehood were
reassessed as a result of the United Nations’ refusal to recognise Transkei,
Bophuthatswana, Venda and Ciskei as independent states. [36] Humanitarian law was
rewritten to confer prisoner-of-war status on combatants belonging to the African
National Congress (ANC) and other national liberation movements. [37] In addition,
South Africa’s six appearances before the International Court of Justice [38] over South
West Africa/Namibia enabled the Court to formulate new rules of law on the status of
international territories, the powers of the United Nations, human rights and self-
determination. While apartheid undermined and discredited the law of South Africa, it
succeeded, perversely, in injecting notions of racial equality, self-determination and
respect for human rights into an international legal order that in 1945 had few
developed rules on these subjects.
While South Africa’s negative contribution to international law during the apartheid
years was substantial, its positive contribution was minimal. In part the blame for this
may be laid at the door of the United Nations, which sought to isolate South Africa by
excluding it from international organisations and law-making conferences. But in large
measure South Africa itself opted for exclusion and isolation by refusing to accept the
primary values of the post-World War II legal
Page 24

order—racial equality, respect for human rights and the advancement of self-
determination.
Law featured prominently in South African foreign policy before 1994. The men who
guided and shaped South Africa’s foreign policies—JC Smuts, JMB Hertzog, Eric Louw,
Hilgard Muller and RF (’Pik’) Botha—were all lawyers. Smuts alone used his talents to
advance the international order. [39] At both the Paris Peace Conference of 1919, [40] and
the San Francisco Conference of 1945, he worked vigorously to promote his vision of a
world in which the rule of law would govern the affairs of states. [41] Louw, Muller and
Botha were less concerned with the advancement of a just world order than they were
with the sovereign right of South Africa to pursue its racial policies. While international
law was for Smuts a sword with which to fashion a new world, for his successors it was
a shield, a means of protecting South Africa against the encroaching values of the latter
half of the 20th century. [42]
In 1994, South Africa’s position in the world changed dramatically following the
abandonment of apartheid and the holding of democratic elections, which saw the
establishment of a Government of National Unity under the presidency of Nelson
Mandela of the African National Congress. South Africa resumed its seat in the General
Assembly and was restored to full membership in specialised agencies of the United
Nations, from which it had been excluded during the apartheid years. It was readmitted
to the Commonwealth of Nations, from which it had withdrawn in 1961, and it became
a member of the Organization of African Unity (OAU), the Non-Aligned Movement and
the Southern African Development Community (SADC). United Nations sanctions
against South Africa were lifted. Diplomatic ties were established with states that had
refused
Page 25

to have anything to do with South Africa during the apartheid era and new embassies
were established abroad, particularly in Africa and Asia. South Africa’s identification
with the values of the international community was evidenced by its signing of the
principal international human rights conventions.
Since its return to the international community, South Africa has played a leading
role in the United Nations, the African Union, the Non-Aligned Movement and the
Commonwealth. From 2007 to 2009 it held a non-permanent seat on the Security
Council of the United Nations and was again elected as a non-permanent member of
the Security Council for the periods 2011–2013 and 2018–2019. In 2011 it became a
member of BRICS (Brazil, Russia, India, China and South Africa). Within Africa, South
Africa has acted as peace broker in the Democratic Republic of the Congo and Burundi.
Since it became a democracy, South Africa has therefore progressed from a pariah
state into a leader in Africa and the world.
Initially, under President Mandela, South Africa pursued a universalist human rights
foreign policy, [43] but under Presidents Mbeki and Zuma the interests of Africa and the
African Union have been prioritised. [44] This is reflected in the decision to withdraw
from membership of the International Criminal Court rather than hand over President
al-Bashir of Sudan to the Court for prosecution—a decision that was later reversed. [45]
******
International law comes before national courts in disputes between individuals or
between the individual and the state. South African courts have probably entertained
more international-law-based arguments than most domestic courts due to the
country’s vibrant commercial life and conflict-ridden political history.
Early cases demonstrated a clear commitment to international law. In 1894, in CC
Maynard et alii v The Field Cornet of Pretoria, [46] Chief Justice Kotze of the Transvaal
Republic declared that municipal law
must be interpreted in such a way as not to conflict with the principles of international law. . . .
It follows from [this], as put by Sir Henry Maine, ‘that the state which disclaims the authority
of international law places herself outside the circle of civilized nations’. It is only by a strict
adherence to these
Page 26

recognized principles that our young state can hope to acquire and maintain the respect of all
civilized communities, and so preserve its own national independence.
In the aftermath of the Anglo-Boer War, courts frequently invoked international law to
settle claims relating to the confiscation of property. [47] Later disputes, particularly
those relating to South Africa’s administration of South West Africa, [48] were likewise
settled by recourse to international law.
In the apartheid era, wide divisions appeared in the South African population over
attitudes towards international law. Whites saw international legal norms as irrelevant,
unfair, and characterised by double standards in their implementation, while blacks
turned to international law both as a source of redress from the injustices of apartheid
and as a standard by which to measure the legitimacy of South African law. [49] The
conflict between the two legal orders was highlighted in many court cases. Leaders of
the South West Africa Peoples Organization (SWAPO), charged with statutory treason
after the revocation of the Mandate for South West Africa by the United Nations,
claimed that a South African court had lost its competence to try them under
international law. [50] Both Nobel Prize Winner Archbishop Desmond Tutu and the leader
of the Pan-Africanist Congress, Robert Sobukwe, invoked international law norms on
freedom of movement to challenge administrative restrictions placed upon their right to
travel abroad. [51] Laws zoning South Africa’s cities along racial lines were challenged on
the ground that they violated the Charter of the United Nations’ provisions on human
rights. [52] Members of the military wings of both SWAPO and the ANC challenged the
jurisdiction of the courts to try them on the ground that they were entitled to prisoner-
of-war status under contemporary humanitarian law. [53] The leaders of an abortive
army coup in Bophuthatswana who were charged with treason argued that the charge
was incompetent on the ground that treason was a crime against a ‘state’, and
Bophuthatswana failed to qualify as a ‘state’ under international law. [54] International
law arguments of this kind
 Page 27

did not receive sympathetic consideration before the courts of the apartheid period. On
the contrary, judgments of this time showed a hostility to international law resembling
that of the executive. [55]
South Africa’s new constitutional order gives full recognition to the place of
international law in national law [56] and requires courts to interpret all
legislation, [57] and particularly the Bill of Rights, [58] to accord with international law.
This has led to a renaissance of international law in the jurisprudence of its courts. This
will be apparent in the chapters that follow.

[1] See Chapter 1.

[2] Clearly, there were highly developed political structures in Africa at this time. See AJGM
Sanders International Jurisprudence in the African Context (1979) part 2; TO Elias Africa and the
Development of International Law (1972) ch 1.
[3] See Chapter 8.
[4] For a description of the government of the Republic of the United Netherlands and its relationship
with the Dutch East India Company, see HR Hahlo and E Kahn The South African Legal System and its
Background (1968) 531–41.
[5] Ibid; JW Wessels History of the Roman-Dutch Law (1908).
[6] A Nussbaum A Concise History of the Law of Nations (1947) 88, 92, 139.
[7] See Chapter 1.
[8] See Chapter 4.
[9] For example, extradition agreements were extended to the Cape by Order-in-Council in terms of the
British Extradition Act of 1870 (33 & 34 Vict c 52). For a full account of the manner in which treaties were
made on behalf of the Cape of Good Hope, see RP Schaffer (née Balkin) A Critical Analysis of the Treaty-
making Powers of the Union of South Africa and the Republic of South Africa (PhD thesis, University of
the Witwatersrand, 1978) 4–8.
[10] For example, Walvis Bay was annexed by the Crown in 1878 and incorporated into the Cape
Colony by Act 35 of 1884. See L Berat Walvis Bay: Decolonization and International Law (1990) 37.
[11] For references to these agreements, see JA Kalley South Africa’s Treaties in Theory and Practice
1806–1998 (2001). See, in particular, Treaties entered into by Governors of the Colony of the Cape of
Good Hope and other British Authorities with Native Chieftains between 1803 and 1854 (1857).
[12] See MN Shaw Title to Territory in Africa: International Legal Issues (1986) 36–45; MF Lindley The
Acquisition and Government of Backward Territory in International Law (1926) 46; TW Bennett ‘Aboriginal
title in South Africa’ (1993) 9 SAJHR 443.
[13] The 1884 Convention stipulated that the ‘South African Republic will conclude no treaty or
engagement with any state or nation other than the Orange Free State . . . until the same has been
approved by her Majesty’. It was this provision, which seriously limited the independence of the SAR, that
gave rise to the debate over its status. See R Jennings and A Watts (eds) Oppenheim’s International
Law 9 ed (1992) vol 1 at 391. T Baty International Law in South Africa (1900) provides an excellent
contemporary account of the debate. Baty favoured the view that the Transvaal was a ‘semi-sovereign’
state (45–68). See S Hofmeyr Die Boere-Republieke en die Volkereg (1933); DJ Pieterse ‘Transvaal en
Britse susereiniteit’ in Archives Year Book for South African History (1940) I ch 6. This issue was
discussed in R v Christian 1924 AD 101 at 108, 125–6. See further Chapter 5.
[14] Baty (n 13) 45–68; JM Spaight War Rights on Land (1911) 14–15.
[15] J Dugard ‘The treatment of rebels in conflicts of a disputed character: The Anglo-Boer War and the
“ANC-Boer War” compared’ in AJM Delissen and GJ Tanya (eds) Humanitarian Law of Armed Conflict:
Challenges Ahead: Essays in Honour of Frits Kalshoven (1991) 447 at 448–50.
[16] See Chapter 25.
[17] Oppenheim (n 13) at 700; Baty (n 13) 92–3.
[18] MM Lewis ‘The international status of the British self-governing dominions’ (1922–1923)
3 BYIL 21.
[19] 28 & 29 Vict c 63.
[20] Oppenheim (n 13) 257.
[21] For a consideration of some of the constitutional implications of the conferment of the mandate for
South West Africa upon South Africa, see J Dugard ‘South West Africa and the supremacy of the South
African Parliament’ (1969) 86 SALJ 194.
[22] M Friedlander ‘The admission of states to the League of Nations’ (1928) 9 BYIL 84 at 85.
[23] A Berriedale Keith The Constitutional Law of the British Dominions (1933) 47.
[24] See further on this conference G Carpenter Introduction to South African Constitutional
Law (1987) 205.
[25] 22 Geo v c 4.
[26] Act 69 of 1934.
[27] Act 70 of 1934.
[28] S Pienaar South Africa and International Relations between the Two World Wars: The League of
Nations Dimension (1987); N Diederichs Die Volkebond (1933).
[29] J Barber and J Barratt South Africa’s Foreign Policy: The Search for Status and Security 1945–
1988 (1990).
[30] WK Hancock Smuts: 2 The Fields of Force 1919–1950 (1968) 428–33. For a more nuanced
assessment of Smuts’ contribution to the drafting of the preamble, see C Heyns and W Gravett ‘“To save
succeeding generations from the scourge of war”: Jan Smuts and the ideological foundations of the
United Nations’ (2017) 39 Human Rights Quarterly 574, 589–94. See, too, C Heyns ‘The Preamble of the
United Nations’ (1995) 7 African Journal of International and Comparative Law 329.
[31] See Chapters 15 and 23.
[32] Articles 55 and 56 of the Charter. See further Chapter 15.
[33] See Chapter 23.
[34] See, in particular, the International Convention on the Suppression and Punishment of the Crime
of Apartheid of 1973. See Chapter 9.
[35] See Chapter 15.
[36] See Chapter 5.
[37] See Chapter 25.
[38] See Chapter 22.
[39] The attitude of Hertzog towards the advancement of an international order was ambivalent. It has
been suggested that his support for South Africa’s participation in world politics and the League of
Nations was motivated mainly by his desire to assert South Africa’s independence from Britain (Pienaar (n
28) 7).
[40] In 1918, Smuts published a pamphlet entitled The League of Nations: A Practical Suggestion,
which served as a foundation for debates at the Paris Peace Conference, particularly on the mandates
system.
[41] WK Hancock Smuts: 1 The Sanguine Years 1870–1919 (1962) ch 21; WK Hancock Smuts: 2 The
Fields of Force 1919–1950 (1968) 428–33. In recent years, much has been written about Smuts’
contribution to international peace and human rights. For a critical assessment, see M Mazower No
Enchanted Palace: The End of Empire and the Ideological Origins of the United Nations (2009) 28–65. For
a more favourable assessment, see Heyns and Gravett (n 30).
[42] South Africa’s attitude towards international law and the United Nations during this period is well
described by JC Heunis United Nations versus South Africa: A Legal Assessment of United Nations and
United Nations Related Activities in Respect of South Africa (1986).
[43] In 1995 President Mandela sought the expulsion of Nigeria from the Commonwealth following the
execution of Sara Wiwo and eight other environmental activists. This action was much criticised by
African states on the ground that it failed to support African solidarity.
[44] For an annual record of South Africa’s foreign policy from a legal perspective, see the annual
report in the South African Yearbook of International Law.
[45] Further on this decision, see Chapters 4 and 10.
[46] (1894) 1 SAR 214 at 223. See, too, the judgment of Jorissen J 232.
[47] See Chapter 25.
[48] See R v Christian 1924 AD 101.
 [49] J Dugard ‘The conflict between international law and South African law: Another divisive factor in
South African society’ (1986) 2 SAJHR 1.
[50] S v Tuhadeleni and Others 1969 (1) SA 153 (A).
[51] Tutu v Minister of Internal Affairs 1982 (4) SA 571 (T); Sobukwe v Minister of Justice 1972 (1) SA
693 (A).
[52] S v Adams; S v Werner 1981 (1) SA 187 (A).
[53] S v Sagarius 1983 (1) SA 833 (SWA); S v Petane 1988 (3) SA 51 (C). See further Chapter 24.
[54] S v Banda 1989 (4) SA 519 (B). See further Chapter 5.
[55] J Dugard ‘The South African judiciary and international law in the apartheid era’ (1998)
14 SAJHR 110.
[56] See Chapter 4.
[57] Section 233 of Constitution of the Republic of South Africa, 1996.
[58] Section 39(1) of Constitution of the Republic of South Africa, 1996.

Page 28

Chapter 3
Sources of International Law

John Dugard and Dire Tladi

Introduction

The sources of international law [1] are described in article 38(1) of the Statute of the
International Court of Justice as:
(a)
international conventions (treaties), whether general or particular;
(b)
international custom, as evidence of a general practice accepted as law
(customary international law);
(c)
the general principles of law recognised by civilised nations; and
(d)
judicial decisions and the teachings of the most highly qualified publicists, as
subsidiary means for the determination of rules of law.
Article 38 was first drafted in 1920 for the Statute of the Permanent Court of
International Justice. [2] It no longer accurately reflects all the materials and forms of
state practice that comprise today’s sources of international law. Despite this, every
effort is made to bring new developments in respect of sources of law within the
categories of sources recognised in article 38. Inevitably, this, at times, leads to the
expansion of these sources beyond those originally contemplated in 1920.
There is no rule in international law providing for a general hierarchy of sources,
particularly between treaty rule and rules of customary international law. [3] Yet, as a
matter of practice, there is greater reliance on treaties and customary international
law. [4] The
Page 29
empirical superiority of these two sources, both of which are founded on the consent of
states, emphasises the consensual basis of international law. [5] Modern international
law has seen important developments in the hierarchy of norms. Whereas in classical
international law all norms and rules enjoyed equal ranking, today, certain norms,
known as peremptory norms of general international law (jus cogens), enjoy a higher
status in the normative hierarchy of sources. Obligations under the UN Charter also
enjoy normative superiority over other rules of international law. [6]

Treaties or conventions

A treaty is an agreement between states or between states and international

organisations governed by international law. The rules relating to the capacity to enter
into treaties, the procedure to be followed for entering into treaties, the interpretation
of treaties, and the termination of treaties are governed by the Vienna Convention on
the Law of Treaties of 1969 and the Vienna Convention on the Law of Treaties between
States and International Organisations and between International Organisations of
1986. [7] These multilateral conventions define treaties as being written instruments.
Many of the provisions of the 1969 Vienna Convention are generally accepted as
reflecting rules of customary international law. [8]
Today treaties govern a wide area of international life. Treaties, which may be
multilateral in the sense that they bind many states, or bilateral and binding on two
states only, are divided broadly into three categories.

1 Contractual

These are treaties of a contractual nature between states governing matters such as
trade, extradition, air and landing rights, and mutual defence. Here, two or more states
‘contract’ with each other to establish a particular legal relationship.
Page 30

2 Legislative

A number of treaties have been entered into between states that codify existing rules of
customary international law or that create new rules of law. Although these treaties are
called legislative or law-making, they are not binding upon non-signatory states. The
basic rule governing treaties is pacta tertiis nec nocent nec prosunt, ie treaties do not
confer obligations or benefits upon non-signatory states. [9]

3 Constitutional

International organisations such as the United Nations are created by multilateral

treaties. The Charter of the United Nations is a treaty to which all member states are
party and which serves as the constitution of the United Nations. Similarly, the African
Union Constitutive Act and the Southern African Development Community (SADC)
Treaty serve as the constitution of the African Union and the SADC respectively. The
Rome Statute of the International Criminal Court is, likewise, a multilateral treaty that
creates a constitution for this court.
Customary International Law

Custom plays an important role in undeveloped societies without institutions for law-
making or adjudication. [10] When a society develops and establishes a legislature and
an effective judicial system, customary rules may be codified, replaced by statute, or
formulated with greater precision by judicial decision with the result that statute and
judicial decision become the primary sources of law. The international system as yet
has no legislature or compulsory judicial system. Consequently, customary international
law occupies a significant role in the international legal order.
Page 31
While states give their express consent to be bound by a rule when they enter into a
treaty, the ‘consent of states’ to a customary rule is inferred from their conduct.
Inevitably, the question whether a state has consented to a rule by its conduct will raise
difficult questions of proof. It is hardly surprising, therefore, that disputes over the
existence of customary rules feature prominently in international litigation.
There are two main requirements for the existence of a rule of customary
international law: settled practice of states (usus) and the acceptance of an obligation
to be bound (opinio juris sive necessitatis).

1 Elements of customary international law

(a) Settled practice (usus)

In the context of customary international law, practice means the practice of
states. [11] The practice of other entities, including that of international organisations
and non-governmental organisations, while not irrelevant, does not generally constitute
practice for the purposes of establishing customary international law. The practice of
other entities is not irrelevant in that it may provide context and may serve as a
catalyst for state practice, but it cannot be a substitute for the practice of states in the
formation of customary international law.
The practice of one, two or a few states is not a sufficient basis for the establishment
of a rule of customary international law. The practice has to be general and widespread.
Evidence of state practice is to be found in a variety of materials, including treaties,
the decisions of national courts, national legislation, diplomatic correspondence, policy
statements by government officers, opinions of national law advisers, reports of the
International Law Commission (ILC) and comments by states on these reports, and
resolutions of international organisations, in particular the political organs of the United
Nations. There is thus a wide range of materials that can qualify as evidence of state
practice. There is, however, a view that only ‘real action’ or ‘action on the ground’ can
qualify as state practice for the purposes of establishing rules of customary
international law. In holding that no customary rule had come into existence in S v
Petane, Conradie J stated:
One must . . . look for state practice at what states have done on the ground in the harsh
climate of a tempestuous world, and not at what their representatives profess in the
ideologically overheated environment of the United Nations where indignation appears
frequently to be a surrogate for action. [12]
Page 32
According to Conradie J, it is necessary to consider the action or practice of states, and
not their promises or rhetoric, as ‘customary international law is founded on practice,
not on preaching’. [13] Here Conradie J echoed the warning of Mr Justice JT van Wyk, ad
hoc judge in the South West Africa legal proceedings of 1960–1966, that practice and
not political statements are required as evidence of custom. [14] This view, which seeks
to exclude political statements by states as well as resolutions of the political organs of
the United Nations from the ambit of state practice, is not consistent with the generally
held view on the formation of customary international law, which accepts a variety of
materials as practice.
Some states provide easy access to their practice by publishing official reports on
this subject. [15] In other countries, digests are published by private institutions.
Although there is currently no comprehensive record of South Africa’s state practice,
the South African Yearbook of International Law annually publishes a survey of events
from the Office of the Chief State Law Adviser for International Law, which, in some
instances, provides examples of South Africa’s state practice. The Yearbook also
publishes a list of treaties that have been signed and/or ratified by South Africa, which
qualifies as state practice. The United Nations has recently produced a useful
compendium of resources where state practice can be found. [16]
Where states actively demonstrate their support for a particular rule, no problem of
proof arises. In many cases, however, there will be no clear evidence of this kind. It
may, under certain circumstances, be possible to infer consent or acquiescence to a
practice from the inaction of some states. [17] According to the ILC only ‘deliberate
abstention from acting’ may count as practice. [18]
According to the International Court of Justice, a practice must constitute ‘constant
and uniform usage’ before it will qualify as custom. This principle was enunciated in
the Asylum Case, [19] which involved the question whether the practice of granting
asylum to
Page 33
political refugees in embassies in Latin American countries amounted to a customary
rule. After examining state practice in this regard, the Court concluded:
The facts brought to the knowledge of the Court disclose so much uncertainty and
contradiction, so much fluctuation and discrepancy in the exercise of diplomatic asylum and in
the official views expressed on various occasions, there has been so much inconsistency in the
rapid succession of conventions on asylum, ratified by some states and rejected by others, and
the practice has been so much influenced by considerations of political expediency in the
various cases, that it is not possible to discern in all this any constant and uniform usage,
accepted as law, with regard to the alleged rule of unilateral and definitive qualification of the
offence. [20]
Although, as a doctrinal matter, the International Court of Justice has insisted on
‘constant and uniform usage’ [21] or ‘widespread’ acceptance of a rule as a prerequisite
for usus, [22] there have been cases in which the Court has not been rigorous in its
application of the rule. In the Arrest Warrant case, [23] for example, it was prepared to
dispense with evidence of state practice. Here, it held that a Minister of Foreign Affairs
is entitled to immunity from the criminal jurisdiction of a foreign court on the ground
that the nature and function of his office require such immunity. [24] The Court made no
attempt to provide evidence of ‘constant and uniform’ or ‘widespread’ practice in
support of such a rule. [25] Incidentally, when considering whether there was an
exception to the immunity rule in cases of international crimes, the Court did apply a
more rigorous search for state practice and, consequently, found that there was no
exception. [26] The Court was similarly not rigorous in the search for state practice when
it found that there was a customary international law rule obliging states to perform an
environmental impact assessment for activities
Page 34
with transboundary impacts. [27] These instances when the Court, for whatever reason,
departs from the rigorous search for practice should not detract from the requirement
of widespread practice for the establishment of a customary rule. The judgment of the
Supreme Court of Appeal in Minister of Justice v SALC provides an excellent example of
the search for state practice. [28]
In most cases, some passage of time is required for a practice to crystallise into a
customary rule. In some cases, however, where little practice is needed to establish a
rule, it may come into existence very rapidly. When the General Assembly unanimously
approved a resolution in 1963 [29] declaring the legal principles governing activities in
outer space, there was widespread agreement that a new rule of customary law had
been created. In addition to general customary international law, international law also
recognises the possibility of regional customary international law, applicable only to
certain regions. The Asylum Case provides an example of a possible regional customary
international law since the case concerned a possible rule of customary international
law applicable in Latin America. Although the Court did not find that a rule of regional
customary international law existed in that case, it accepted the possibility but held that
the practice of granting asylum was not sufficiently ‘constant and uniform’ to form the
basis of a rule of customary international law.
There also exists the possibility for local custom, applicable between two
states. [30] The formation of a local customary international law rule, applicable to South
Africa and Lesotho alone, was considered in Nkondo v Minister of Police, [31] in which the
Court was asked to find that custom permitted passengers, who were on board an
aircraft en route to Lesotho and were forced to land in South Africa in distress due to
bad weather, to exercise a right of transit across South African territory with full
immunity from arrest for
Page 35
crimes against the safety of the state. [32] Relying on the Asylum Case, [33] Smuts J held
that ‘[t]he fact that on only four occasions passengers were allowed passage across
South African territory without having to comply with immigration formalities’ could not
serve as evidence of a customary rule. [34] Although the Court does not find that there is
sufficient practice in that case, the possibility of a local customary international law rule
applicable between two states is accepted.
For a practice to give rise to a rule of customary international law, it must receive
‘general’ or ‘widespread’ acceptance. [35] Universal acceptance is not
necessary. [36] Although Rumpff CJ suggested that universal acceptance was required
in Nduli v Minister of Justice, [37] subsequent South African decisions have questioned
the correctness of this view and indicated that general acceptance is sufficient. [38] This
raises the question whether a dissenting state will be bound by the general acceptance
of a rule by other states. One view is that article 38(1)(b) of the Statute of the
International Court of Justice ‘does not exclude the possibility of a few dissidents for the
purpose of the creation of a customary international law’. [39] Another view, approved
by Conradie J in S v Petane, [40] is that if a state persistently objects to a particular
practice while the law is still in the process of development, it cannot be bound by any
customary rule that may emerge from such a practice. There is both judicial [41] and
academic opinion [42] supporting the view that a ‘persistent objector’ is not bound in
such a
Page 36
case. This question has been the subject of debate in the context of apartheid and
international law as it is clear that states refused to accept South Africa’s persistent
objection to treating apartheid as a violation of customary law. [43] This is best
explained on the ground that the prohibition on apartheid is a peremptory norm, a
norm of jus cogens, to which the normal rules relating to persistent objection are
inapplicable.

(b) Opinio juris

A settled practice on its own is insufficient to create a customary rule. In addition, there
must be a sense of obligation, a feeling on the part of states that they are bound by the
rule in question—that the general practice is, in the language of article 38(1)(b),
‘accepted as law’. This requirement, sometimes described as the psychological element
in the formation of customary international law, was emphasised by the International
Court of Justice in the North Sea Continental Shelf Cases between West Germany on
the one hand, and the Netherlands and Denmark on the other. [44] Here, the Court
found that a provision requiring the continental shelf to be divided in accordance with
the principle of equidistance contained in the 1958 Geneva Convention on the
Continental Shelf had not become a customary rule as, despite some practice in favour
of the application of the equidistance principle, there was ‘no evidence that [states] so
acted because they felt legally compelled to draw [continental-shelf boundaries] in this
way by reason of a rule of customary law obliging them to do so’. [45] Elaborating on
this requirement, the Court stated:
Not only must the acts concerned amount to a settled practice, but they must also be such, or
be carried out in such a way, as to be evidence of a belief that this practice is rendered
obligatory by the existence of a rule of law requiring it. The need for such a belief, ie the
existence of a subjective element, is implicit in the very notion of the opinio juris sive
necessitatis. The states concerned must therefore feel that they are conforming to what
amounts to a legal obligation. The frequency, or even habitual character of the acts is not in
itself enough. [46]
Proof of opinio juris is difficult to produce. Consequently, it is argued by some jurists
that opinio juris will be presumed when there
Page 37
is evidence of a general practice in support of a particular rule. [47] The judgments of
the International Court of Justice in the North Sea Continental Shelf Cases and
the Nicaragua case, [48] however, do not endorse such a presumption. Similarly, the ILC
has reconfirmed the continued applicability of both requirements for the establishment
of a rule of customary international law.
Evidence of opinio juris can be found in the same materials that are used for state
practice. The difference is that for opinio juris the search is for evidence that states feel
bound or entitled to act in a particular way, not just for political or policy reasons but
because of the existence of a rule of customary international law. To give a concrete
example, the consistent and unanimous adoption of an annual resolution may well
constitute practice regarding a rule contained in the resolution. The resolution itself,
however, cannot establish a rule of customary international law unless it can be shown,
whether by reference to the content of the resolution or by some other acts, that the
states adopting the resolution believe the rule in question to be one of customary
international law.
As with practice, the question of the ‘silent majority’—those states that do not
express an opinion as to the legal status of a rule—arises. In these instances, silence
can, in certain circumstances, be taken to imply the acceptance of a rule. However, this
inference can only be made when those silent states ‘were in a position to react and the
circumstances called for some reaction’. [49]

2 Resolutions of international organisations

Resolutions of international organisations, in particular political organs of the United

Nations, can play an important part in the formation of customary international law,
either as practice or as opinio juris. [50] Clearly a resolution of either the General
Assembly or the Security Council categorised as a recommendation is not binding on
states per se. However, an accumulation of resolutions, a repetition of
recommendations on a particular subject, may amount to
Page 38
evidence of collective practice on the part of states. Moreover, to the extent that the
content of a resolution specifies that a practice is a rule of customary international law,
such a resolution may also serve as opinio juris.
In some cases, it is accepted that a recommendation of the General Assembly may
become a customary rule with very little repetition. As shown above, the Declaration of
Legal Principles Governing the Activities of States in the Exploration and Use of Outer
Space of 1963 probably became a customary rule upon its adoption by the General
Assembly.
More controversial and politically contentious resolutions create problems. In the
1960–1966 legal proceedings between Ethiopia/Liberia and South Africa over the status
of South West Africa (Namibia) it was argued that a customary rule of non-
discrimination had been created by repeated General Assembly resolutions condemning
apartheid and practices of racial discrimination. South Africa vigorously opposed this on
the ground that it would amount to conferring legislative powers upon the General
Assembly. In his separate opinion in the South West Africa Cases, Second Phase, the
South African ad hoc judge, Mr Justice JT van Wyk, rejected this argument that
repeated resolutions may establish rules of customary international law, declaring that
it ‘involved the novel proposition that the organs of the United Nations possessed some
form of legislative competence whereby they could bind a dissenting
minority’. [51] Although there was support for Judge Van Wyk’s view in 1966, [52] today,
Judge Tanaka’s dissenting opinion [53] in favour of the existence of a customary rule of
non-discrimination generated by the political organs of the United Nations prevails.
The Declaration on the Granting of Independence to Colonial Countries and Peoples
of 1960 (General Assembly Resolution 1514 (XV)) provides another example of an
instance in which a highly controversial resolution has been accepted as evidence of a
customary rule. In this resolution the General Assembly asserted the right of all peoples
to self-determination and called on colonial powers to take immediate steps to grant
independence to their colonial peoples, thereby, in effect, outlawing colonialism.
Despite initial resistance to this resolution on the part of colonial powers, it became the
basis for the decolonisation of most colonies. The reiteration of this Declaration by large
majorities in the General Assembly and the fact of
 Page 39
decolonisation, which constitutes practice, provide evidence of a settled practice in
support of a customary rule outlawing colonialism.
During the apartheid era, South African courts took a conservative approach to the
generation of customary rules by resolutions of the General Assembly. In S v Petane,
involving the question whether Additional Protocol I of 1977, extending prisoner-of-war
status to members of national liberation movements, reflected a customary rule as it
has been endorsed by resolutions of the General Assembly, Conradie J stated:
[I]t is doubtful whether resolutions passed by the United Nations General Assembly qualify as
state practice at all. . . . They may constitute opinio juris which, if expressed with respect to a
rule sufficiently delineated through usus, may create a customary rule of international law
. . . . But, if there is no preceding usus, such a declaration cannot give birth to a customary
rule, unless, of course, the declaration itself is treated as usus at the same time. However, it
takes too wide a stretching of the concept of usus to arrive at the latter conclusion. [54]
Conradie J accordingly held that there was insufficient evidence of usus, in the form of
‘material, concrete and/or specific acts of states’, to justify a customary rule on this
subject. He warned that ‘United Nations resolutions cannot be said to be evidence of
state practice if they relate, not to what the resolving states take it upon themselves to
do, but to what they prescribe for others’. [55]
In reaching this conclusion, Conradie J cast doubt on the validity of the view that
certain provisions of the Universal Declaration of Human Rights, adopted by the General
Assembly in 1948, had become rules of customary international law. Here, he reasoned
that it was difficult to discern a customary rule in abstract statements in support of the
Universal Declaration and the incorporation of its provisions into national constitutions
when they were contradicted by state practice constituting a flagrant violation of its
provisions. [56] A similar approach was adopted by Cooper J in S v Rudman. [57]
This cautious judicial approach may be contrasted with that of the United States
Second Circuit Court of Appeals in Filartiga v Pena-Irala, [58] in which the Court relied on
the Universal Declaration of Human Rights and other resolutions of the General
Assembly to substantiate the creation of a customary rule of international law
prohibiting state torture. Here the Court stated:
Page 40
This prohibition has become part of customary international law, as evidenced and defined by
the Universal Declaration of Human Rights . . . which states, in plainest of terms, ‘no one shall
be subjected to torture’. The General Assembly has declared that the Charter precepts
embodied in this Universal Declaration ‘constitute basic principles of international law’. GA Res
2625 (XXV) (1970) . . . . These UN declarations are significant because they specify with great
precision the obligations of member nations under the Charter. [59]
S v Petane, as with many apartheid-era cases, reflected South Africa’s aversion and
suspicion of international law. This has now given way to what has been termed an
‘international law-friendly’ constitutional dispensation. [60] As exemplified by the
Constitutional Court judgment in S v Makwanyane, the South African judiciary now
adopts a more open posture towards international law and resolutions adopted by the
United Nations, amongst others. [61] This approach is consistent with the approach of
the International Court of Justice in the Nicaragua case [62] in which the Court was
required to decide whether the prohibition on the use of force contained in article 2(4)
of the United Nations Charter had acquired the additional status of a customary rule.
Without any consideration of usus on this subject, the Court held that it
has however to be satisfied that there exists in customary international law an opinio juris as
to the binding character of such abstention [from the use of force]. This opinio juris may,
 though with all due caution, be deduced from, inter alia, the attitude of . . . states towards
certain General Assembly resolutions . . . . The effect of consent to the text of such resolutions
. . . may be understood as an acceptance of the validity of the rule or set of rules declared by
the resolution. [63]
A cautionary admonition on this subject has, however, been given by the International
Court of Justice in its advisory opinion on the Legality of the Threat or Use of Nuclear
Weapons: [64]
The Court notes that General Assembly resolutions, even if they are not binding, may
sometimes have normative value. They can, in certain circumstances, provide evidence
important for establishing the existence of a rule or the emergence of an opinio juris. To
establish whether this is true of a given General Assembly resolution, it is necessary to look at
its content and the conditions of its adoption; it is also necessary to see whether an opinio juris
Page 41
exists as to its normative character. Or a series of resolutions may show the gradual evolution
of the opinio juris required for the establishment of a new rule. . . . Examined in their totality,
the General Assembly resolutions put before the Court declare that the use of nuclear weapons
would be ‘a direct violation of the Charter of the United Nations’; and in certain formulations
that such use ‘should be prohibited’. The focus of these resolutions has sometimes shifted to
diverse related matters; however, several of the resolutions under consideration in the present
case have been adopted with substantial numbers of negative votes and abstentions; thus,
although those resolutions are a clear sign of deep concern regarding the problem of nuclear
weapons, they still fall short of establishing the existence of an opinio juris on the illegality of
the use of such weapons. [65]
The Security Council is not a legislative body. Its primary function is to maintain
international peace and to ensure, if necessary by means of binding resolutions, that
states comply with their obligations under the Charter. In recent years, however, the
Security Council has passed a number of binding resolutions requiring states to adopt
measures to combat terrorism that have a decidedly legislative character. [66] Whether
this heralds a new approach to its role remains to be seen.

3 Soft law [67]

Lawyers are accustomed to drawing a clear distinction between law and non-law—hence
the importance of rules for identifying the point at which a practice on the part of states
becomes a customary rule of law. Today it is suggested that there is ‘something’ in
between that merits the attention of lawyers: ‘soft law’. These are imprecise standards,
generated by declarations adopted by diplomatic conferences or resolutions of
international organisations, that are intended to serve as guidelines to states in their
conduct, but which lack the status of ‘law’. The Helsinki Final Act of the Conference on
Security and Co-operation in Europe of 1975 [68] is an example of a non-binding
declaration by states which has had a profound effect on the promotion of human rights
in Eastern Europe. Another such declaration is the Rio Declaration on the Environment
and Development of 1992, [69] which expounds a number of principles, such as the duty
on
Page 42
states to inform states of environmental hazards and the duty on states to carry out an
environmental impact assessment for activities likely to have an adverse impact on the
environment. The difficulty in securing the consent of states to multilateral treaties
governing environmental matters, on which there is a need for urgent action, has
prompted this recourse to non-binding standards. Indeed, today environmental
international ‘law’ is heavily dependent on ‘soft law’. [70] The passage of time and state
practice in support of such a standard may convert it into a customary rule, but until
this occurs it serves as a useful guide to state conduct without creating legal obligations
for states.

The relationship between Customary International Law and

Treaty Law

Customary international law rules may be embodied in treaty rules. In

the Nicaragua case, [71] for example, the United States argued that the Court did not
have jurisdiction since the case concerned the interpretation of article 2(4) of the UN
Charter relating to the prohibition on the use of force. In its declaration accepting the
jurisdiction of the Court, the United States had excluded the jurisdiction of the Court in
relation to ‘disputes arising under a multilateral treaty’. [72] The Court, however,
concluding that the reservation cannot bar the Court from exercising jurisdiction
because Nicaragua does not only rely on article 2(4) of the UN Charter but also on
customary international law, observed that:
Nicaragua invokes a number of principles of customary and general international law that,
according to the Application, have been violated by the United States. The Court cannot
dismiss the claim of Nicaragua under principles of customary and general international law,
simply because such principles have been enshrined in the texts of the conventions relied upon
by Nicaragua. The fact that the abovementioned principles, recognised as such, have been
codified or embodied in multilateral conventions does not mean that they cease to exist and to
apply as principles of customary law, even as regards countries that are parties to such
conventions. [73]
The reasoning by the Court points to the relationship between a treaty and rules and
customary international law having the same or
Page 43
similar content. In the North Sea Continental Shelf Cases the International Court of
Justice identified three ways that a treaty provision can result in or contribute to a rule
of customary international law having the same content. [74] First, a rule of customary
international law could be codified into a treaty after the crystallisation of the
customary international law rule. Secondly, the process of concluding the treaty could
have led to the crystallisation of a rule of customary international law that was
emerging. Finally, state practice subsequent to the adoption of the treaty—partly in
response to the treaty—led to the emergence of a new rule of customary international
law. It should be stressed, however, that in each case it remains necessary to show the
existence of a customary international law rule by reference to state practice and opinio
juris. Moreover, in relation to opinio juris, it is necessary to show that state practice is
not due to the application of the treaty but rather due to the belief that there exists a
‘mandatory rule of customary international law’. [75]
The emergence of a rule of customary international law reflected in a treaty
provision was also considered in S v Petane, [76] in which the Cape Provincial Division
considered the question as to whether a member of the ANC’s military wing, Umkhonto
we Sizwe, was entitled to prisoner-of-war status on the ground that Additional Protocol
1 of 1977 to the Geneva Conventions of 1949, which accords such status to members
of national liberation movements, had become part of customary international law and
was thus binding upon South Africa. (At this time South Africa was not a party to the
Protocol. It became a party in 1995.) The fact that some 60 states had signed the
Protocol, said the Court, did not amount to settled practice as there was no evidence
that the rule extending prisoner-of-war status to members of national liberation
movements had been endorsed by states in their practice. Moreover, the two states at
which this aspect of the Protocol was directed, namely Israel and South Africa, had
persistently refused to sign the Protocol or to accept its prescriptions in practice.

General principles of law recognised by civilised nations

General principles of law constitute a reserve store of legal principles upon which courts
and tribunals may draw when there are no rules of treaty or customary law applicable.
In such instances, courts turn to common principles of law found in municipal
systems—in so far
Page 44
as they are capable of application to relations between states—in order to fill the gaps
in international law. Thus, while international law does not, as a doctrinal matter,
recognise hierarchy of sources, general principles are by definition subordinate to treaty
and customary international law.
Although sparse use is made of this source, international courts have on occasion
invoked the principles of unjust enrichment, [77] reparation for breach of an
undertaking, [78] res judicata, [79] the limited liability of a
corporation, [80] estoppel [81] and nemo judex in re sua. [82] As international tribunals,
such as the International Court of Justice, comprise judges from different national
backgrounds, they are well qualified to draw on general principles of this kind.
An example of the manner in which an international court may draw on general
principles of law is provided by the separate opinion of Judge McNair in
the International Status of South West Africa Advisory Opinion. [83] In considering the
nature of the Mandate for South West Africa, Judge McNair stated that it was
permissible to have regard to the English system of trusts and to continental
institutions of mandate and tutelage in order to ascertain the principles to be applied to
the Mandate for South West Africa. He concluded that these institutions indicated that
the rights of the trustee were limited and that the trustee was under a legal obligation
to administer the property for the benefit of another. In particular, the trustee was not
permitted to absorb the trust property into his own estate. Applying these principles to
the Mandate for South West Africa, he held that South Africa was unable to alter the
status of South West Africa by absorbing the territory into South Africa without first
obtaining the consent of the United Nations.
Unlike treaties and custom, general principles of law do not have a consensual basis.
Consequently, natural lawyers have claimed that the existence of this source of law
confirms the natural law basis of the international legal order. In the 1966 South West
Africa Cases, Second Phase, Judge Tanaka of Japan stated:
Page 45
[I]t is undeniable that in article 38(1)(c) some natural-law elements are inherent. It extends
the concept of the source of international law beyond the limit of legal positivism according to
which, the states being bound only by their own will, international law is nothing but the law of
the consent and auto-limitation of the state. [84]
States which cling to a positivist approach to international law have difficulty in
accepting this source. In the 1966 South West Africa Cases, Judge ad hoc Van Wyk
rejected the argument that a norm of non-discrimination was binding on South Africa as
a general principle of law, stating that article 38(1)(c) of the Statute of the
International Court of Justice does not mean that by legislating on particular domestic
matters a majority of nations could compel a minority to introduce similar
legislation. [85]
Although the natural law basis of this source should not be exaggerated, its influence
is clear. This is further evidenced by the manner in which international courts have
invoked considerations of humanity [86] and equity [87] under the rubric of general
principles of law.

Judicial precedent

Judicial decisions are a subsidiary means for the determination of rules of law. Judicial
decisions in this context refers to both decisions of national and international courts.
Article 38(1)(d) of the Statute of the International Court of Justice contains two
caveats. First, judicial decisions are only a ‘subsidiary means for the determination of
the rules of law’. This means that judicial decisions, under article 38(1)(d) of the
Statute, do not themselves constitute rules of international law but are only a way of
identifying the rules of international law. Since judicial decisions are only a subsidiary
means for determining the rules of law, the authority of each decision ultimately
depends on the quality of the reasoning. [88]
The second caveat is that judicial decisions as a source of law are subject to article
59 of the Statute, which provides that: ‘The decision of the Court has no binding force
except between the parties and in respect of that particular case.’
This means that international law, like in the civil-law tradition, knows no doctrine
of stare decisis. In practice, however, there is a natural tendency for courts to follow
their own previous decisions or the decisions of other international tribunals. While the
International
Page 46
Court of Justice refers to its own previous decisions with caution, other international
tribunals and domestic courts charged with international law matters do not hesitate to
invoke previous decisions on international law from both international and domestic
tribunals.
In recent times—as a result of the increased workload of the International Court of
Justice, the proliferation of courts and arbitral tribunals, and the willingness of states to
submit to international adjudication—judicial decisions, particularly of international
courts, have come to play a more central role in international law. This is a natural
development in the evolution of international law from a primitive system in which
adjudication was rare to a sophisticated legal order in which courts play a central role.
It is important to draw a clear distinction between the use of judicial decisions under
article 38(1)(d), on the one hand, and, on the other, the reliance of judicial decisions as
a source of state practice. The former refers to decisions of both domestic and
international courts and serves a subsidiary purpose of determining the rules of
international law, while the latter refers only to decisions of domestic courts as organs
through which the state acts. For the former, the quality of the reasoning is important
for its authority while for the latter the quality of the reasoning plays a less important
role.

Text writings
The teachings of the most highly qualified publicists are also a subsidiary means for the
determination of law. The writings of jurists accordingly provide evidence of rules of law
in the same way as writings of Roman-Dutch jurists in South Africa. International
tribunals make less use of the writings of jurists today than in the past, as new sources
of law have to a large extent made it unnecessary to rely upon the views of jurists for
an exposition of the law.

Unilateral Acts of States

Sometimes the unilateral acts of states, particularly in the form of statements by

government officials, may create international obligations, and are thus a source of
law. [89] In a dispute between Denmark and Norway about the sovereignty over
Greenland, Denmark
Page 47
claimed that the Norwegian Foreign Minister, Mr Ihlen, had acknowledged Denmark’s
claim to sovereignty over Greenland in the course of a minuted conversation with the
Danish Minister accredited to Norway. The Permanent Court of International Justice
held that in the light of the ‘Ihlen Declaration’, [90] Norway was under an obligation to
refrain from contesting Danish sovereignty over Greenland. In a subsequent dispute
between Australia and France over the unlawfulness of French nuclear tests in the
atmosphere in the South Pacific Ocean, the International Court of Justice held it was
unnecessary for the Court to give a judgment in this matter because statements by the
French President at a news conference indicated that France would discontinue
atmospheric tests, thereby rendering the dispute moot. In so finding, the Court stated:
It is well recognized that declarations made by way of unilateral acts, concerning factual or
legal situations, may have the effect of creating legal obligations . . . . When it is the intention
of the State making the declaration that it should become bound according to its terms, that
intention confers on the declaration the character of a legal undertaking, the State being
thenceforth legally required to follow a course of conduct consistent with the declaration. An
undertaking of this kind, if given publicly and with intent to be bound, even though not made
within the context of international negotiations, is binding. [91]
Unilateral acts of this kind do not fall within any of the sources listed in article 38(1) of
the Statute of the International Court of Justice. Nevertheless, they do create legal
obligations.

Codification

The uncertainties of customary international law have prompted the international

community to resort to codification. Article 13(1) of the United Nations Charter
authorises the General Assembly to initiate studies and make recommendations to
encourage ‘the progressive development of international law and its codification’. To
carry out this task the General Assembly has established an International Law
Commission consisting of 34 persons of recognised competence in international law,
elected by the General Assembly and representing the main legal systems of the world.
To date, two South Africans have served on the International Law Commission. John
Dugard served on the ILC from 1997 to 2011 while Dire Tladi was first elected in 2011
and continues to serve on the ILC.
Although the constitution of the International Law Commission distinguishes between
the progressive development of international
 Page 48
law, ie the exposition of new principles of law, and the codification of law, ie the
formulation and systematisation of existing rules of law, it has been difficult to maintain
such a distinction in practice. In codifying existing areas of the law, the International
Law Commission has been required progressively to develop the law by expounding
new principles, with the result that the distinction between progressive development
and codification has become blurred.
The ILC prepares reports and draft conventions, which may later be submitted to
international conferences for approval. In this way, a number of significant multilateral
law-making conventions have come into existence. The Vienna Convention of the Law
of Treaties (1969), the Geneva Convention on Diplomatic Relations (1961), the Geneva
Convention on Consular Relations (1963), the four Geneva Conventions on the Law of
the Sea (1958), the Convention on the Prevention and Punishment of Crimes against
Internationally Protected Persons, including Diplomatic Agents (1973), the Convention
on the Non-Navigational Use of International Watercourses (1997) and the Convention
on Jurisdictional Immunities of States and their Property (2004) are examples of
conventions drafted by the ILC. [92] In fact, even the Rome Statute of the International
Criminal Court was negotiated on the basis of a draft Statute prepared by the
International Law Commission.
The International Law Commission’s products goes beyond draft conventions
converted into multilateral treaties: it includes draft codes, guidelines, restatements of
the law, draft articles that may be transformed into treaties and the reports of the
special rapporteurs. The Draft Articles on the Responsibility of States for Internationally
Wrongful Acts (2001) and the Draft Articles on Diplomatic Protection (2006) are
examples of instruments yet to be translated into treaty form which serve as an
authoritative statement of the law. [93] The
Page 49
work of the ILC is frequently relied on by international tribunals, [94] which suggests that
it qualifies as a source of law, [95] either in its own right or as evidence of custom or the
expression of views of ‘highly qualified publicists’. Sir Arthur Watts states:
To a considerable extent, debate about the Commission’s place in the scheme of things set out
in Article 38 of the Statute of the International Court of Justice is artificial. As already noted,
that scheme, even if it is about sources at all, is not necessarily exhaustive of the sources of
international law; also as noted, the Court has not hesitated to invoke the work of the
International Law Commission where it seemed appropriate to do so, without concerning itself
with the formal question of how that work might fall within Article 38 of the Court’s Statute. It
may be more fruitful to think of the Commission’s work, and particularly its final draft Articles,
in terms of material, rather than formal sources, of international law. [96]

Jus Cogens, obligations Erga Omnes and a system of higher

norms

Traditionally, international society is viewed as a horizontal system premised on the

sovereign equality of states, while international law is seen as a body of rules based on
consent and characterised by their neutrality. A necessary consequence of this is that
all legal norms are equal in status. [97] The horizontality of international law means that
general rules of international law can be modified or derogated from by states through
treaty and customary international law. Two states or more can decide, whether
through treaty or local custom, that they will not apply customary international law
rules on immunities in their relations with each other. By the same token, two or more
states can decide to modify the application of the Vienna Convention on Diplomatic
Relations. These general rules of international law that can be modified easily through
the agreement of states are referred to as jus dispositivum.
In contrast to jus dispositivum, there exists a category of norms of international law
which cannot be modified or derogated by the mere will of states. [98] This category of
norms of international law are
Page 50
referred to as peremptory norms of general international law (jus cogens).
The notion of jus cogens has a long history in international law, dating as far back as
Hugo de Groot. [99] However, its position in international law was cemented with its
inconclusion in the Vienna Convention on the Law of Treaties of 1969, which, in article
53, [100] provides:
A treaty is void, if, at the time of its conclusion, it conflicts with a peremptory norm of general
international law. For the purposes of the present Convention, a peremptory norm of general
international law is a norm accepted and recognized by the international community of states
as a whole as a norm from which no derogation is permitted and which can be modified only by
a subsequent norm of general international law having the same character.
There is general support for jus cogens as a doctrine and much state practice and
decisions of international courts in support of it. [101] In the prohibition against South
Africa, for example, the Constitutional Court recognised that the prohibition against
torture is a peremptory norm of international law. [102] Certain norms are now generally
recognised as peremptory. These include the prohibition of aggression, the prohibitions
against slavery, genocide, racial discrimination (including apartheid), torture and the
denial of self-determination. [103] The International Court of Justice, after studiously
avoiding any acknowledgement of peremptory norms, recognised that the prohibitions
on genocide and racial discrimination were peremptory norms
Page 51
in Democratic Republic of the Congo v Rwanda in 2006. [104] In that case, John Dugard,
sitting as a judge ad hoc, suggested that norms of jus cogens had an important role to
play in the judicial process:
The judicial decision is essentially an exercise in choice. Where authorities are divided, or
different general principles compete for priority, or different rules of interpretation lead to
different conclusions, or State practices conflict, the judge is required to make a choice. In
exercising this choice, the judge will be guided by principles (propositions that describe rights)
and policies (propositions that describe goals) in order to arrive at a coherent conclusion that
most effectively furthers the integrity of the international legal order.
Norms of jus cogens are a blend of principle and policy. On the one hand, they affirm the
high principles of international law, which recognize the most important rights of the
international order—such as the right to be free from aggression, genocide, torture and slavery
and the right to self-determination; while, on the other hand, they give legal form to the most
fundamental policies or goals of the international community—the prohibitions on aggression,
genocide, torture and slavery and the advancement of self-determination. This explains why
they enjoy a hierarchical superiority to other norms in the international legal order. The fact
that norms of jus cogens advance both principle and policy means that they must inevitably
play a dominant role in the process of judicial choice. [105]
While there are a core group of norms that are recognised as jus cogens, uncertainty
still remains about the method of identifying jus cogens norms and the consequences
that flow from the peremptory status of a norm. To address these and other questions,
the International Law Commission decided to appoint a Special Rapporteur to study the
topic in 2015. To date, the Special Rapporteur has produced three reports.
In its work thus far, the Drafting Committee of the ILC has adopted conclusions that
describe the general characteristics of jus cogens as norms that reflect and protect the
fundamental values of the international community, that are hierarchically superior to
other norms of international law and that are universally applicable. [106] The Drafting
Committee has also defined a two-step approach to the identification of jus
cogens norms. [107] First, it must be a norm of general international law. Secondly, it
must be recognised and accepted by the international community of states as a whole
as one
Page 52
from which no derogation is permitted. Whether a norm protects the fundamental
values of the international community, and is deemed to be universally applicable and
hierarchically superior to other norms, must be taken into account in assessing the
second element.
The question of the consequences of jus cogens has perhaps been the most divisive
and controversial. [108] International and regional courts have generally tended to adopt
a narrow approach to the consequences of jus cogens—sometimes unduly so. This is
perhaps most reflected in the current debate on whether there are exceptions to
immunity for jus cogens crimes. [109] In the Arrest Warrant case, the International Court
of Justice implicitly concluded that there was no such exception. [110] This conclusion
was made more explicit by the Court in Jurisdictional Immunities of the State, where
the Court drew a somewhat arbitrary distinction between immunities, which it said were
procedural in nature, and the more substantive norms of jus cogens: [111]
Assuming for this purpose that the rules of law of armed conflict which prohibit the murder of
civilians in occupied territory, the deportation of civilian inhabitants to slave labour and the
deportation of prisoners of war to slave labour are rules of jus cogens, there is no conflict
between those rules and the rules on State immunity. The two sets of rules address different
matters. The rules of State immunity are procedural in character and are confined to
determining whether or not the courts of one State may exercise jurisdiction in respect of
another State. They do not bear upon the question of whether or not the conduct in respect of
which the proceedings are brought was lawful or unlawful.
This line of reasoning has generally been followed in regional courts, in particular the
European Court of Human Rights. [112] Whatever the merits or demerits of the reasoning
of the Court in Jurisdictional Immunities of the State, this reasoning should be
restricted to immunity from civil proceedings and not extended to immunities in
criminal proceedings. [113] The Court itself cautioned that ‘the question of whether, and
if so to what extent, immunity
Page 53
might apply in criminal proceedings against an official of the State’ was not an issue in
this case. [114] In the South African context, the Supreme Court of Appeal in Minister of
Justice v SALC took the view that jus cogens norms did not establish exceptions to the
customary international law immunity of heads of state from foreign criminal
jurisdiction. [115] However, two cautionary notes should be made. First, the Court did
find that under South African domestic law there was indeed an exception created by
statute. Secondly, the conclusion that jus cogens did not establish an exception appears
limited to the immunity of heads of state, heads of government and ministers for
foreign affairs.
 A development related to jus cogens is the emergence of the concept of
obligations erga omnes, which developed in response to the South West Africa Cases of
1966, [116] in which the International Court of Justice denied legal standing to Ethiopia
and Liberia to enforce an obligation owed to the international community—namely the
obligation on the part of the South African government to ‘promote to the utmost the
material and moral well-being and social progress’ of the people of South West Africa
(Namibia). [117] In 1970, in Barcelona Traction, Light and Power Company
Limited, [118] the International Court went out of its way to repudiate its finding of 1966
in an obiter dictum, which indicated that a litigant state would no longer be required to
prove a national interest in the subject matter of its claim where an obligation of
concern to all states—an obligation erga omnes—was involved. Here, the Court stated:
[A]n essential distinction should be drawn between the obligations of a state towards the
international community as a whole, and those arising vis-à-vis another state in the field of
diplomatic protection. By their very nature the former are the concern of all states. In view of
the importance of the rights involved, all States can be held to have a legal interest in their
protection; they are obligations erga omnes.
Such obligations derive, for example, in contemporary international law, from the outlawing
of acts of aggression, and of genocide, as also from the principles and rules concerning the
basic rights of the human person, including protection from slavery and racial
discrimination. [119]
Page 54
As the International Court was itself responsible for first expounding the notion of
obligations erga omnes, it is not surprising that it was more willing to recognise this
concept than it was to recognise the concept of jus cogens. In the East Timor case, the
Court accepted that the right of peoples to self-determination ‘has an erga
omnes character’, [120] and, in its advisory opinion on Legal Consequences of the
Construction of a Wall in the Occupied Palestinian Territory, [121] the Court held that by
constructing a wall within Palestinian territory, Israel had violated certain
obligations erga omnes, notably ‘the obligation to respect the right of the Palestinian
people to self-determination, and certain of its obligations under international
humanitarian law’. [122] It is noteworthy that all these obligations with an erga
omnes character flow from norms that are recognised as being jus cogens. [123] It would
thus seem that there is an inextricable link between jus cogens norms and
obligations erga omnes in that the erga omnes nature of an obligation flows from
the jus cogens character of the relevant norm.
In his dissenting opinion in the case Certain Activities Carried Out by Nicaragua in
the Border Area (Costa Rica v Nicaragua), [124] Judge ad hoc John Dugard stated that
‘the release of carbon into the atmosphere and the loss of gas sequestration services is
certainly an obligation erga omnes’. [125] In Judge ad hoc Dugard’s opinion, the
determination that an obligation is one of erga omnes must be relevant in assessing the
quantum of damages for internationally wrongful acts. [126]
The International Law Commission has given recognition to the concepts of jus
cogens and obligations erga omnes in its 2001 Draft Articles on the Responsibility of
States for Internationally Wrongful Acts. [127] The Commission carefully constructs a
scheme for the responsibility of states in the case of a breach of higher norms, which
recognises both the notion of peremptory norms and the notion that certain obligations
are owed to the international community as a whole. It attaches serious consequences
to the breach of a peremptory
Page 55
norm and recognises the right of non-injured states to institute proceedings and take
measures on behalf of the international community as a whole against a state that has
violated a ‘higher norm’.
Articles 40 and 41 provide that states are obliged to co-operate in bringing to an
end, through lawful means, ‘serious breaches by a state of an obligation arising under a
peremptory norm of a general international law’. They are also obliged to refrain from
recognising as lawful a situation created by a serious breach of a peremptory norm.
Moreover, in contrast to the decision of the International Court in the 1966 South West
Africa Cases, [128] a non-injured state is entitled to invoke the responsibility of another
state if the other state violates an obligation ‘owed to the international community as a
whole’. [129] The possibility of such a non-injured state applying measures against a
delinquent state ‘to ensure cessation of the breach and reparation in the interest of the
injured state or of the beneficiaries of the obligation breached’ is also
contemplated. [130]
The notions of jus cogens and obligations erga omnes have had a profound effect on
international law. Together they have transformed international law from a system in
which all rules carried equal weight to a system of ‘graduated normativity’ [131] in which
certain norms enjoy a higher status. Although this development has been challenged on
the ground that international society has not evolved to the point where such a two-tier
system of norms can be sustained, [132] there can be little doubt that it has transformed
the nature and structure of international law. [133]
It is inconceivable that a state committed to compliance with international law,
respect for human rights and the promotion of the rule of law under its own
constitutional order would tolerate the violation of a norm of jus cogens. The obiter
dictum of the Cape Provincial Division in Azanian Peoples’ Organisation (AZAPO) v Truth
and Reconciliation Commission, [134] that South Africa’s constitutional rules ‘would, it
would seem, enable Parliament to pass a law even if such law is contrary to the jus
cogens’, [135] is therefore unfortunate. The
Page 56
failure of the Court to consider the full implications of its dictum suggests that it had
little understanding of the role of jus cogens in the international legal order, and of the
importance of ensuring harmony between domestic constitutional norms and the
peremptory norms of international law.

[1] See further on this subject H Thirlway The Sources of International Law (2014); J
d’Aspremont Formalism and the Sources of International Law (2011); C Parry The Sources and Evidences
of International Law (1965); ME Villiger Customary International Law and Treaties (1985); A
Cassese International Law in a Divided World (1986) 169–99; P Szasz ‘General law-making processes’ in
O Schachter and C Joyner (eds) United Nations Legal Order (1995) vol 1 35; A Boyle and C Chinkin The
Making of International Law (2007); A Pellet ‘Article 38’ in A Zimmerman & T Thienel (eds) The Statute of
the International Court of Justice: A Commentary 3 ed (2018).
[2] O Spiermann and M Fitzmaurice ‘The history of article 38 of the Statute of the International Court of
Justice’ in S Besson and J d’Aspremont (eds) The Oxford Handbook of the Sources of International
Law (2017).
[3] J Pauwelyn Conflict of Norms in Public International Law (2003) 133.
[4] In the North Sea Continental Shelf Cases (Germany/Denmark; Germany/Netherlands) 1969 ICJ
Reports 3, the International Court of Justice adopts a methodical approach of first trying to identify
whether there is an applicable treaty rule and then assessing the possibility of a rule of customary
international law before, having found no applicable or customary international rule, seeking other
applicable sources for the delimitation of the continental shelf.
[5] In the Lotus Case (France v Turkey) 1927 PCIJ Reports Series A no 10 at 18, the Permanent Court
of International Justice stated that ‘[t]he rules of law binding upon states . . . emanate from their own
free will as expressed in conventions or by usages generally accepted as expressing principles of law’.
[6] See article 103 of the UN Charter.
[7] These are discussed in Chapter 20.
[8] See eg Case Concerning Kasikili/Sedudu Island (Botswana/Namibia) 1999 ICJ Reports 1045 para
18; Case Concerning the Territorial Dispute (Libyan Arab Jamahiriya/Chad) 1994 ICJ Reports 6 para
41; Case Concerning Oil Platforms (Islamic Republic of Iran v United States of America) 1996 ICJ Reports
803 para 23.
[9] This was accepted by Conradie J in S v Petane 1988 (3) SA 51 (C) 61E–F. See further Y Dinstein
‘The interaction between customary international law and treaties’ (2006) 322 Hague Recueil des
cours 247.
[10] For a discussion of customary international law as a source of law, see A D’Amato The Concept of
Custom in International Law (1971); M Akehurst ‘Custom as a source of international law’ (1974–1975)
47 BYIL 1; HA Strydom ‘Customary international law: The legacy of false prophets’ (1994) 27 CILSA 276;
M Byers Custom, Power and the Power of Rules (1999); AE Roberts ‘Traditional and modern approaches
to customary international law: A reconciliation’ (2001) 95 AJIL 757; MP Scharf Customary International
Law in Times of Fundamental Change (2013); CA Bradley Custom’s Future: International Law in a
Changing World (2016).
[11] Draft Conclusion 4 of the Draft Conclusions on the Identification of Customary International Law
adopted by the International Law Commission during its Seventieth Session (A/CN.4/L.908).
[12] S v Petane (n 9) 61D–E.
[13] Ibid 59F–G.
[14] South West Africa Cases (Ethiopia v South Africa; Liberia v South Africa), Second Phase 1966 ICJ
Reports 6 at 169.
[15] See, for discussion, memorandum by the UN Secretariat Ways and Means for Making the Evidence
of Customary International Law More Readily Available 2018 (A/CN.4/710) paras 26–122.
[16] Ibid. Annexes I–VI.
[17] Draft Conclusion 6, para 1 of the Draft Conclusions on the Identification of Customary
International Law (n 11).
[18] Ibid para 3 of the commentary to Draft Conclusion 6.
[19] Asylum Case (Colombia/Peru) 1950 ICJ Reports 266.
[20] Asylum Case (n 19) 277. The Court seemed to relax this strict approach to settled practice in
the Case Concerning Military and Paramilitary Activities in and Against Nicaragua (Nicaragua v USA) 1986
ICJ Reports 14 when it stated that a custom did not require ‘absolutely rigorous conformity with the rule’.
It is sufficient that ‘the conduct of states should, in general, be consistent with such rules, and that
instances of state conduct inconsistent with a given rule should generally have been treated as breaches
of that rule, not as indications of the recognition of a new rule’ (at 98).
[21] Asylum Case (n 19).
[22] Fisheries Jurisdiction Case (United Kingdom v Iceland) 1974 ICJ Reports 3.
[23] Case Concerning the Arrest Warrant of 11 April 2000 (Democratic Republic of Congo v
Belgium) 2002 ICJ Reports 3.
[24] Ibid paras 53–4.
[25] See the dissenting opinion of Judge ad hoc Van den Wyngaert in this case: ibid 143–51, paras 11–
23.
[26] Arrest Warrant case (n 23) para 58.
[27] Case Concerning Pulp Mills on the River Uruguay (Argentina v Uruguay) 2010 ICJ Reports 14 para
203.
[28] Minister of Justice and Constitutional Development v Southern Africa Litigation Centre 2016 (3) SA
317 (SCA) (‘Minister of Justice v SALC’) paras 66 et seq. See, for discussion, D Tladi ‘Interpretation and
international law in South African courts: The Supreme Court of Appeal and the Al Bashir Saga’ (2016)
16 African Human Rights Law Journal 310.
[29] Resolution 1962 (XVIII). See Chapter 18.
[30] That local customary rules are permissible is clear from the Case Concerning Right of Passage over
Indian Territory 1960 ICJ Reports 6 at 39.
[31] 1980 (2) SA 894 (O). See J Dugard ‘Jurisdiction over persons on board an aircraft landing in
distress’ (1981) 30 ICLQ 902.
[32] Here, the applicant, a leading member of the outlawed ANC, wanted by the South African Police in
connection with the commission of crimes against the state, was arrested in South Africa when the
Lesotho Airways flight on which he was travelling was forced to land in Bloemfontein due to bad weather.
[33] Asylum Case (n 19).
[34] Nkondo v Minister of Police (n 31) 908F–G.
[35] Fisheries Jurisdiction Case (n 22) 23–6.
[36] See the dictum of Judge Lachs in the North Sea Continental Shelf Cases (Germany/Denmark;
Germany/Netherlands) 1969 ICJ Reports 3 at 229.
[37] 1978 (1) SA 893 (A) 906D.
[38] See Inter-Science Research and Development Services (Pty) Ltd v Republica Popular de
Mocambique 1980 (2) SA 111 (T) 124–5; S v Petane (n 9) 56–7. Although the Supreme Court of Appeal
in Minister of Justice v SALC (n 28) para 70 refers to ‘an established universal practice’, this is likely just
an unfortunate choice of words rather than an attempt to establish universal practice as a standard for
customary international law.
[39] Dissenting opinion of Judge Tanaka, South West Africa Cases, Second Phase (n 14) 291.
[40] S v Petane (n 9) 64A–B.
[41] See the Anglo-Norwegian Fisheries Case (United Kingdom v Norway) 1951 ICJ Reports 116 at
131; North Sea Continental Shelf Cases (n 36) 26–7, 131 (separate opinion of Judge Ammoun); Asylum
Case (n 19) 277–8; Nicaragua case (n 20) 107.
[42] J Crawford Brownlie’s Principles of Public International Law 8 ed (2012) 28. Cf JI Charney ‘The
persistent-objector rule and the development of customary international law’ (1985) 56 BYIL 1.
[43] JI Charney ‘Universal international law’ (1993) 87 AJIL 529 at 539–40; L Henkin International
Law: Politics and Values (1995) 39; O Schachter ‘International law in theory and practice’ (1984 V)
178 Hague Recueil 9, 119, 130–1.
[44] North Sea Continental Shelf Cases (n 36).
[45] Ibid 45.
[46] Ibid 44.
[47] Crawford (n 42) 26. Dissenting opinion of Judge Tanaka in North Sea Continental Shelf Cases 1969
ICJ Reports 176; dissenting opinion of Judge ad hoc Sorensen (246–7).
[48] Nicaragua case (n 20) 108–9.
[49] Draft Conclusion 10 of the Draft Conclusions on the Identification of Customary International Law
(n 11).
[50] Ibid at Draft Conclusion 12. See, for discussion, I McGibbon ‘Means for the identification of
international law: General Assembly resolutions: Custom, practice and mistaken identity’ in Bin Cheng
(ed) International Law: Teaching and Practice (1982) 10; and B Sloan ‘General Assembly resolutions
revisited’ (1987) 58 BYIL 39. For a critical perspective, see P Weil ‘Towards relative normativity in
international law’ (1983) 77 AJIL 413 at 417.
[51] 1966 ICJ Reports 6 at 170.
[52] See the dissenting opinion of Judge Jessup 432–3.
[53] At 291–4.
[54] S v Petane (n 9) 58A–F.
[55] At 59F–G.
[56] At 58G–J.
[57] 1989 (3) SA 368 (E) 376A–B.
[58] 630 F 2d 876 (1980).
[59] At 882–4.
[60] See N Botha ‘Justice Sachs and the interpretation of international law by the Constitutional Court:
Equity or expediency?’ (2010) 25 South African Public Law 235. See D Tladi ‘Interpretation of treaties in
an international law-friendly framework: The case of South Africa’ in Helmut Aust and Georg Nolte
(eds) The Interpretation of International Law by Domestic Courts: Uniformity, Diversity,
Convergence (2016).
[61] S v Makwanyane 1995 (3) SA 391 (CC) para 35.
[62] Nicaragua case (n 20) 14.
[63] Ibid 99–100.
[64] 1996 ICJ Reports 226.
[65] 1996 ICJ Reports paras 70–1.
[66] See further discussion of this subject in Chapter 23.
[67] JH van Hoof Rethinking the Sources of International Law (1983) 179; M Olivier ‘Informal
agreements under the 1996 Constitution’ (1997) 22 SAYIL 63 at 69; M Olivier ‘Relevance of “soft law” as
a source of international human rights’ 2002 CILSA 289; D Shelton (ed) Commitment and Compliance:
The Role of Non-binding Norms in the International Legal System (2000).
[68] (1975) 14 ILM 1292.
[69] (1992) 31 ILM 874.
[70] P Birnie and A Boyle International Law and the Environment (1992) 26.
[71] Case Concerning Military and Paramilitary Activities in and Against Nicaragua (Nicaragua v United
States) 1984 ICJ Reports 392.
[72] Ibid para 67.
[73] Ibid para 73. See also Nicaragua case (n 20) para 175: ‘[E]ven if a treaty norm and a customary
norm relevant to the present dispute were to have exactly the same content, this would be no reason for
the Court to take the view that the operation of the treaty process must necessarily deprive the
customary norm of its separate applicability.’
[74] North Sea Continental Shelf Cases (n 36) paras 60 et seq.
[75] Ibid para 76.
[76] S v Petane (n 9).
[77] Lena Goldfields Arbitration (1930) 5 AD 3.
[78] Chorzow Factory (Merits) (1928) PCIJ Reports Series A no 17 at 29.
[79] Effect of Awards of Compensation Made by the UN Administrative Tribunal 1954 ICJ Reports 47 at
53; Case Concerning the Application of the Convention on the Prevention and Punishment of the Crime of
Genocide (Bosnia and Herzegovina v Serbia and Montenegro) 2007 ICJ Reports 43, 89–102.
[80] Barcelona Traction Light and Power Company, Ltd (Belgium v Spain) 1970 ICJ Reports 3 at 33–5.
[81] Case Concerning the Temple of Preah Vihear (Cambodia v Thailand) 1962 ICJ Reports 6 at 23, 31,
32.
[82] Mosul Boundary Case (1925) PCIJ Reports Series B no 12 at 32.
[83] 1950 ICJ Reports 128 at 148.
[84] South West Africa Cases (n 14).
[85] Ibid 170.
[86] Corfu Channel (Merits) 1949 ICJ Reports 4 at 22; Nicaragua case (n 20) 114.
[87] Crawford (n 42) 44–7.
[88] See para 3 of the commentary to Draft Conclusion 13 of the Draft Conclusions on the Identification
of Customary International Law (n 11).
[89] Crawford (n 42) 416–18; AP Rubin ‘The international legal effects of unilateral declarations’ (1977)
71 AJIL 1; W Fiedler ‘Unilateral acts in international law’ in R Bernhardt (ed) Encyclopedia of Public
International Law (2000) vol IV, 1018; E Suy ‘Some unfinished new thoughts on unilateral acts of state
as a source of international law’ (2002) 27 Tydskrif vir Regswetenskap 1.
[90] Legal Status of Eastern Greenland 1933 PCIJ Reports Series A/B no 53.
[91] Nuclear Tests Case (Australia v France) 1974 ICJ Reports 253 at 267–8.
[92] The South African Yearbook of International Law includes an annual survey of the work of the ILC.
See, eg D Tladi ‘Progressively developing and codifying international law: The work of the International
Law Commission in its sixty-seventh session’ (2015) 40 SAYIL 432. For a more general overview of the
work of the ILC, see R Jennings and A Watts (eds) Oppenheim’s International Law 9 ed, vol 1 (1992)
103–10; J Dugard ‘How effective is the International Law Commission in the development of international
law?’ (1998) 23 SAYIL 34; A Watts The International Law Commission 1949–1998 (1999) 3 vols; A
Pronto and M Wood The International Law Commission 1999–2009 (2010) vol 4; A Pronto ‘Some
thoughts on the making of international law’ (2008) 19 EJIL 601.
[93] Report of the International Law Commission, General Assembly Official Records 56th Session,
Supplement 10 (A/56/10) 29 (2001); Report of the International Law Commission, General Assembly
Official Records 58th Session, Supplement 10 (A/61/10) 13 (2006).
[94] See, for example, the Case Concerning the Gabcikovo-Nagymaros Project 1997 ICJ Reports 7 at
39–41 in which the International Court of Justice relied on the ILC’s Draft Articles on State Responsibility.
[95] I Sinclair The International Law Commission (1987) 121.
[96] Watts (n 92) vol 1, 15.
[97] See P Weil ‘Towards relative normativity in international law?’ (1983) 77 AJIL 413.
[98] First Report on Jus Cogens by Dire Tladi, Special Rapporteur (2016) (A/CN.4/693) para 64. For
more on jus cogens, see A Cassese International Law 2 ed (2005) 198–212; D Shelton ‘Normative
hierarchy in international law’ (2006) 100 AJIL 291; A Bianchi ‘Human rights and the magic of jus cogens’
(2008) 19 EJIL 491; A Orakhelashvili Peremptory Norms in International Law (2008).
[99] First Report on Jus Cogens (n 98) paras 18 et seq.
[100] CL Rozakis The Concept of Jus Cogens in The Law of Treaties (1976); J Sztucki Jus Cogens and
the Vienna Convention on the Law of Treaties: A Critical Appraisal (1974).
[101] See, eg, Besluß des Zweiten Senats 7 April 1965, BVerfGE 18, 441 at 449 where the German
Constitutional Court upheld a treaty provision because the jus cogens norm advanced to impugn a
provision in the treaty did not belong to the category of jus cogens; R v Bow Street Metropolitan
Stipendiary Magistrate, Ex parte Pinochet Ugarte (No 3) [1999] 2 All ER 97 (HL); Ferrini v Federal
Republic of Germany (Italian Court of Cassation) 11 March 2004 (reported in (2005) 99 AJIL 242); Yousuf
v Samantar 699 F.3d 763; Kenya Section of the International Commission of Jurists v Attorney-
General [2011] E-Kenyan Law Reports; Case Concerning Armed Activities on the Territory of the Congo
(Democratic Republic of Congo v Rwanda) 2006 ICJ Reports 6 para 64. See also the International
Criminal Tribunal for the former Yugoslavia in Furundzija, Case IT–95–17/1–A (121 ILR 213, 260).
[102] National Commissioner of Police v Southern African Human Rights Litigation Centre 2015 (1) SA
315 (CC) para 14.
[103] See, eg, para 5 of the commentary to Draft Article 26 of the Draft Articles on the Responsibility
of States for Internationally Wrongful Acts (2001).
[104] Case Concerning Armed Activities on the Territory of the Congo (Democratic Republic of Congo v
Rwanda) 2006 ICJ Reports 6 para 64 and para 78.
[105] Ibid 611 (para 10).
[106] Draft Conclusion 2 of Draft Conclusions on Peremptory Norms of General International Law (Jus
Cogens), adopted by the Drafting Committee of the International Law Commission; statement of the
Chair of the Drafting Committee on Peremptory Norms of General International Law (Jus Cogens) (2017)
Annex.
[107] Draft Conclusion 4 of Draft Conclusions on Peremptory Norms of General International Law (Jus
Cogens).
[108] See D Costelloe Legal Consequences of Peremptory Norms in International Law (2017); Third
Report on Peremptory Norms of General International Law (Jus Cogens) by Dire Tladi, Special
Rapporteur (2018) (A/CN.4/714).
[109] See Chapter 9 for more discussion.
[110] Arrest Warrant case (n 23).
[111] Jurisdictional Immunities of the State (Germany v Italy: Greece intervening) 2012 ICJ Reports 99
paras 92 et seq.
[112] See, for instance, the decisions of the European Court of Human Rights in Al-Adsani v United
Kingdom (2001) 34 EHRR 273, 123 ILR 24. See also Naït-Liman v Switzerland, Judgment of the Grand
Chamber of the European Court of Human Rights of 15 March 2018.
[113] Jurisdictional Immunities of the State (n 111).
[114] Jurisdictional Immunities of the State (n 111) para 91. See also Al-Adsani (n 112) para 61.
[115] Minister of Justice v SALC (n 28).
[116] South West Africa Cases (n 14).
[117] See, for discussion, J Allain ‘Decolonisation as the source of the concepts of jus cogens and
obligations erga omnes’ 2016 Ethiopian Yearbook of International Law 35.
[118] 1970 ICJ Reports 3. See further J Dugard ‘1966 and all that: The South West Africa judgment
revisited in the East Timor case’ (1996) 8 African Journal of International and Comparative Law 549.
 [119] At 32.
[120] East Timor (Portugal v Australia) 1995 ICJ Reports 90 at 102. See, too, Application of the
Convention on the Prevention and the Punishment of the Crime of Genocide, Preliminary Objections 1996
ICJ Reports 595 at 616.
[121] 2004 ICJ Reports 136, (2004) 43 ILM 1009.
[122] Ibid para 155.
[123] Third Report on Peremptory Norms of General International Law (n 108) paras 108 et seq.
[124] Judgment of the International Court of Justice of 2 February 2018 (not yet reported).
[125] Ibid. Dissenting opinion of Judge ad hoc Dugard para 36.
[126] Ibid.
[127] 2001 Draft Articles on the Responsibility of States for Internationally Wrongful Acts, Report of the
International Law Commission, General Assembly Official Records 56th Session, Supplement 10
(A/56/10) 29 (2001).
[128] Above (n 14).
[129] Draft Articles on State Responsibility (n 127).
[130] Ibid article 54.
[131] Weil (n 50).
[132] Ibid.
[133] On the impact of peremptory norms on the international legal order, see A Ferreira-Snyman
‘Sovereignty and the changing nature of international law: Towards a world law?’ (2007) 40 CILSA 395.
[134] 1996 (4) SA 562 (C).
[135] At 574B–C.

Page 57

Chapter 4
The Place of International Law in South
African Municipal Law

John Dugard and Andreas Coutsoudis

The relationship between international law and municipal law troubles both theorists
and courts. There are two main approaches to this subject—the monist and the
dualist. [1]
The monist school, whose leading exponents are Kelsen, Verdross, and Scelle,
maintains ‘that international and municipal law, far from being essentially different,
must be regarded as manifestations of a single conception of law’. [2] Consequently,
monists argue that municipal courts are obliged to apply rules of international law
directly without the need for any act of adoption by the courts, or transformation by the
legislature. For them, international law is incorporated into municipal law without any
act of adoption or transformation—hence the fact that the monist position is often
described as lending support to a ‘doctrine of incorporation’.
Dualists, led by Triepel and Anzilotti, see international law and municipal law as
completely different systems of law, with the result that international law may be
applied by domestic courts only if ‘adopted’ by such courts, or transformed into local
law by legislation. [3] Lauterpacht portrays the dualist position as follows: [4]
 According to the dualistic view, international and municipal law differ so radically in the matter
of subjects of the law, its sources and its substance, that a rule of international law can
never per se become part of the law of the land; it must be made so by the express or implied
authority of the state. Thus conceived, the dualistic view is merely a manifestation of the
traditional positivist attitude.
While maintaining that international law is not foreign law,
Page 58
monists have been compelled to accept that the whole body of international law binding
on a state cannot be directly applied by municipal courts. This has led to the emergence
of the ‘harmonisation theory’, which qualifies the absolute monist position by
acknowledging that in cases of conflict between international law and municipal law the
judge must apply his country’s own jurisdictional rules. [5] This means that customary
international law is to be applied directly as part of the common law, but that conflicting
statutory rules and acts of state may prevail over international law. In this way,
‘harmony’ is achieved between international law and municipal law.
Whatever the jurisprudential basis for the application of international law in
municipal law may be, the undeniable fact is that international law is today applied in
municipal courts with more frequency than in the past. In so doing, courts seldom
question the theoretical explanation for their recourse to international law. [6]

Roman-Dutch Law

The monist–dualist debate postdates the classical period of Roman-Dutch scholarship.

Nevertheless, it is illuminating to examine the attitude of Roman-Dutch law to
international law, particularly as it now has constitutional endorsement.
Grotius saw international law and municipal law as components of a universal legal
order premised on natural law. [7] He drew no clear distinction between international
law and municipal law, and certainly did not regard international law as a foreign legal
system. Thus, De Jure Belli ac Pacis, which claims to be a treatise on international law,
expounds on a number of municipal law topics, such as contract, delict, family relations
and criminal law. Although Van Bynkershoek did not share Grotius’s faith in natural law,
he likewise failed to distinguish between international law and municipal law. Thus, his
major work Quaestionum Juris Publici contains an integrated study of problems of
international law and Dutch public
Page 59
law. According to De Louter, in his introduction to Tenney Frank’s translation of this
work, Van Bynkershoek
disdains the important demarcation between international and national public law and freely
intermingles questions of real international relations with those which only concern the
constitution of his country and are ruled by national laws and customs. The result is a medley
of materials not always easy to disentangle. [8]
Roman-Dutch law writers generally accepted Grotius’s approach to the nature of law
and consequently drew no sharp distinction between international law and municipal
law. [9] Thus, they did not hesitate to apply international law rules in the municipal law
of Holland. [10] The contention that international law and municipal law are inherently
different legal orders—to be treated as such—was raised only at the end of the 19th
century and, consequently, did not concern the Roman-Dutch jurists. Moreover, it is
unlikely that they would have been persuaded by this argument as their loyalty to
natural law was largely unquestioning, [11] and, according to the tenets of natural law,
international law and municipal law are components of a universal legal order. In these
circumstances, it is possible to assert that under Roman-Dutch law international law
formed part of municipal law. [12]
After the British occupation in 1806, the Cape retained Roman-Dutch law as its
common law, and this common law was in due course accepted by the other colonies
and states in southern Africa. This meant that international law remained part of the
common law of South Africa and was applied directly by the courts without any
statutory incorporation. [13] The extent to which this principle applied in the South
African Republic is illustrated by the extravagant comments of Kotzé CJ and Jorissen J
in CC Maynard et alii v The Field
Page 60
Cornet of Pretoria. [14] After citing with approval a passage from Wharton’s Digest that
‘the Law of Nations makes an integral part of the laws of the land’, Kotzé CJ continued
that it followed
as put by Sir Henry Maine ‘that the state which disclaims the authority of International Law
places herself outside the circle of civilized nations’. It is only by a strict adherence to these
recognized principles that our young state can hope to acquire and maintain the respect of all
civilised communities, and so preserve its own national independence. [15]
In a similar vein, Jorissen J declared in respect of ‘the great principles of modern
international law’: [16]
There is no doubt that the laws of this Republic must be viewed and interpreted, in case of
ambiguity, with due regard to the higher law, which is accepted in all civilized countries as
ideal without further proof.
The Anglo-Boer War (1899–1902) raised a wide range of international law
problems, [17] some of which came before the municipal courts of southern Africa. The
rights of belligerents, [18] the confiscation of property for war purposes, [19] the seizure
of enemy property, [20] and the legality of the annexation of the Boer republics [21] were
considered by municipal courts within the context of international law. The views of
Roman-Dutch jurists, particularly Grotius, and other publicists were examined in the
search for the relevant rule of international law, and in no case was it suggested that
international law was a foreign system of law.

Anglo-American Law

Before 1994, South Africa’s constitutional system was modelled on that of Britain.
Consequently, South African courts frequently turned to English law, rather than
Roman-Dutch law, for guidance on
Page 61
questions of public law, including public international law. [22] As English law, like
Roman-Dutch law, treats customary international law as part of municipal
law, [23] recourse to English law simply confirmed the common-law rule governing the
relationship between international law and municipal law. The occasional reference to
American law [24] further cemented the position as American law likewise treats
customary international law as part of domestic law. [25]

South African Law before 1994 [26]

International law is essentially made up of treaties, reflecting the express agreement of
states, and custom, which comprises those rules
Page 62
of international conduct to which states have given their tacit consent. Different rules
applied to the applicability of treaties in South African law before 1994 because treaties
were entered into by the executive alone without the endorsement of the
legislature. [27]

1 Customary international law

For over 100 years, South African courts simply assumed that the rules and principles
of customary international law might be applied by municipal courts as if they were in
some way part of South African law. Consequently, they did not require international
law to be proved as a foreign legal system. Indeed, in 1971 in South Atlantic Islands
Development Corporation Ltd v Buchan, [28] the Court refused to admit an affidavit from
an expert on international law on the ground that international law was not foreign law
and therefore could not be proved by affidavit. South African courts therefore showed
strong support for the monist approach (the doctrine of incorporation) in respect of
customary international law.
In most cases, courts applied customary international law without questioning its
place in the legal order. [29] But in a number of cases, commencing in 1971 with South
Atlantic Islands Development Corporation Ltd v Buchan, [30] the courts expressly
asserted that international law ‘forms part of our law’ and that it was the duty of a
municipal court ‘to ascertain and administer the appropriate rule of international law’.
While the Cape Provincial Division in South Atlantic Islands [31] found support for the
proposition that international law is part of our law in
Page 63
Anglo-American law, the Appellate Division stressed, in Nduli v Minister of
Justice, [32] that ‘the fons et origo of this proposition must be found in Roman-Dutch
law’. [33] This served as an important reminder that the rule favouring the incorporation
of customary international law into South African law is derived from Roman-Dutch law
and not English law.
Before 1994, there was one isolated case in which customary international law was
transformed into municipal law by legislation. The Prize Jurisdiction Act, [34] which
confers prize jurisdiction on all divisions of the Supreme Court of South Africa, defines
‘prize’ for the purposes of the Act as, inter alia, ‘a ship or aircraft captured as prize jure
belli’. This statutory incorporation of the law of war in prize cases was unnecessary, as
the Minister of Justice himself admitted when he explained, in introducing the second
reading of the Bill, that ‘[t]he present position is that our courts have to consult
international law in determining what is a prize’ and that the purpose of the statutory
reference to the law of war was to make it clear that this remained the position. [35] This
exceptional case of statutory incorporation ex abundanti cautela therefore lends support
to the monist position rather than that of the dualists.
As customary international law is a species of common law, it was subordinate to all
forms of legislation. [36] There was, however, a statutory presumption that the
legislature did not intend to violate international law. [37] Although some decisions
asserted that customary international law was subordinate to the common
law, [38] there was only one case in which this matter was considered and there the
Court gave priority to the rule of customary international law. [39]
Page 64

2 Treaties and municipal courts [40]

In South Africa, [41] as in the United Kingdom, the power to enter into treaties was
entrusted completely to the executive. The legislature played no part in the treaty-
making process. Consequently, if treaties were to have become part of South African
law without legislative endorsement, wide law-making powers would have been
conferred on the executive. This explains why treaties, in most instances, did not
become part of municipal law without some act of legislative transformation. [42]
The need for legislation to transform a treaty into South African law was clearly
spelled out by Steyn CJ in Pan American World Airways Incorporated v SA Fire and
Accident Insurance Co Ltd [43] when he stated that it was
trite law . . . that in this country the conclusion of a treaty, convention or agreement by the
South African government with any other government is an executive and not a legislative act.
As a general rule, the provisions of an international instrument so concluded, are not embodied
in our law except by legislative process . . . In the absence of any enactment giving [its]
relevant provisions the force of law, [it] cannot affect the rights of the subject.
The principle of transformation was also extended to resolutions of the General
Assembly and Security Council of the United Nations. [44]
Numerous treaties were transformed into South African law by legislative means. But
the most important multilateral treaty in the modern world, the Charter of the United
Nations, which South Africa signed and ratified in 1945, was not and still has not been
enacted into our law.
Page 65

The New Constitutional Order [45]

1 The 1993 and 1996 Constitutions

In 1993, 26 political groups assembled at Kempton Park to draft a constitution to bring

an end to the apartheid legal order. As these groups were in most instances unelected,
and simply reflected the political realities of the time, it was considered inappropriate to
confer on them the power to draft a final constitution. Instead, the constitution which
they fashioned in negotiations that lasted some six months was to serve as an ‘interim’
constitution, pending the drafting of a constitution by a democratically elected
constitutional assembly. As the Interim Constitution represented a political compromise
between rival groups, notably the National Party (which had
Page 66
ruled South Africa since 1948) and the African National Congress (outlawed from 1960
to 1990), it was agreed at Kempton Park that the ‘final’ constitution would comply with
34 constitutional principles contained in a schedule to the Interim Constitution and that
the Constitutional Court created by the 1993 Interim Constitution would be empowered
to pronounce on the issue of compliance. This Interim Constitution, approved at
Kempton Park, was duly endorsed by the last apartheid Parliament and became the
Constitution of the Republic of South Africa Act 200 of 1993.
 On 27 April 1994, the Interim Constitution came into effect to govern South Africa’s
first democratic elections. The Parliament thus served the dual role of legislature and
Constitutional Assembly. From January 1995 to May 1996, the Constitutional Assembly
met regularly to draft the ‘final’ constitution in accordance with the 34 constitutional
principles agreed upon at Kempton Park. A draft constitution was approved by the
required two-thirds majority vote in the Constitutional Assembly on 8 May and was
forwarded to the Constitutional Court for certification. The Constitutional Court,
however, found fault with a number of provisions in the draft constitution, [46] on the
grounds that they failed to comply with the constitutional principles contained in the
Interim Constitution, and referred it back to the Constitutional Assembly. After these
faults had been remedied by the Constitutional Assembly, the Constitutional Court gave
its final approval to the Constitution on 4 December 1996. [47] The new Constitution—
the Constitution of the Republic of South Africa, 1996 [48]—was signed into law by
President Mandela on 10 December 1996.
Previous South African constitutions made no mention of the place of international
law in the South African legal order. Both the 1993 and the 1996 Constitutions
remedied this omission to ensure that ‘international law has a special place in our
law’. [49] As the two Constitutions are similar in most respects in their treatment of
Page 67
international law, the present study will focus attention principally on the 1996
Constitution. Divergences between the two Constitutions will, however, be examined.

2 Customary international law

South African common law treats international law as part of municipal law, as shown
above. The common law is given constitutional endorsement by s 232 of the 1996
Constitution, which provides that: ‘Customary international law is law in the Republic
unless it is inconsistent with the Constitution or an Act of Parliament.’ [50]
There can be little doubt that the ‘constitutionalisation’ of this rule gives it additional
weight. Moreover, customary international law is no longer subject to subordinate
legislation. Only a provision of the Constitution or an Act of Parliament that is clearly
inconsistent with customary international law will trump it. This is emphasised by s 233
of the 1996 Constitution, which provides that:
When interpreting any legislation, every court must prefer any reasonable interpretation of the
legislation that is consistent with international law over any alternative interpretation that is
inconsistent with international law.
Common-law rules and judicial decisions are now subordinate to customary
international law as it is only the Constitution and Acts of Parliament that enjoy greater
legal weight. There can be no suggestion therefore that a new rule of customary
international law must give way to South African judicial decisions recognising an
earlier rule. Consequently, the doctrine of stare decisis cannot be invoked as an
obstacle to the application of a new rule of international law. This accords with
the dictum of Eksteen J in Kaffraria Property Co (Pty) Ltd v Government of the Republic
of Zambia, [51] in which he applied the principle expounded by Lord Denning MR in
the Trendtex case [52] that ‘international law knows no rule of stare decisis’. [53]

3 Proof of customary international law

As noted above, s 232 expressly provides that ‘customary international law’ (in its
totality) is law in South Africa, unless the customary international law in question is
inconsistent with the Constitution or an Act of Parliament. This makes clear that s 232
compels courts to undertake a two-stage inquiry. First, a court must determine what
the relevant customary international law is in relation to the issue facing
Page 68
the court. Secondly, having determined the customary international law position, the
court must determine whether the relevant customary international law is consistent
with the Constitution and domestic legislation. Therefore, it is necessary to turn to both
pre-constitutional and post-constitutional judicial precedent in relation to how rules of
customary international law are to be determined or proved and then applied.
Since international law is not foreign law, courts may take judicial notice of it as if it
were part of our own common law. In practice, this means that courts turn to the
judicial decisions of international tribunals and domestic courts, both South African and
foreign, [54] and to international law treatises for guidance as to whether a particular
rule is accepted as a rule of customary international law on the ground that it meets the
twin qualifications of usus and opinio juris. [55]
Some pre-constitutional South African decisions suggest that our law requires a
more stringent test for the acceptance of custom than international law itself demands.
In 1905, De Villiers CJ suggested that for a customary rule to be applied, it was
necessary that it be ‘universally accepted’. [56] Similar language was employed by
Rumpff CJ in Nduli when he stated: [57]
It was conceded by counsel for appellants that according to our law only such rules of
customary international law are to be regarded as part of our law as are either universally
recognized or have received the assent of this country . . . . I think that this concession was
rightly made.
This test is too strictly formulated. It is therefore necessary to add
Page 69
the qualification introduced by Margo J in Inter-Science Research and Development
Services (Pty) Ltd v Republica Popular de Mocambique: [58]
The concept of universal recognition in this context is obviously not an absolute one, despite
the ordinary meaning of the word ‘universal’, for, ‘if a custom becomes established as a
general rule of international law, it binds all states which have not opposed it, whether or not
they themselves played an active part in its formation’.
This qualification was confirmed by Conradie J in S v Petane: [59]

It is not clear to me whether Rumpff CJ in giving the judgment [in Nduli] meant to lay down
any stricter requirements for the incorporation of international law usages into South African
law than the requirements laid down by international law itself for the acceptance of usages by
states. International law does not require universal acceptance for a usage of states to become
a custom. . . . Margo J, in giving the judgment of the full Transvaal court in Inter-Science
Research and Development Services (Pty) Ltd v Republica Popular de Mocambique 1980 (2) SA
111 (T), did not think that the word ‘universal’, despite its ordinary meaning, was really
intended to mean universal. I do not think so either. In the present case, however, the
distinction between universal and general recognition makes no difference. I am prepared to
accept that where a rule of customary international law is recognized as such by international
law it will be so recognized by our law.
The correct approach to be adopted is well illustrated by Conradie J’s judgment in S v
Petane, in which he considered the question whether the 1977 Protocol I to the Geneva
Conventions of 1949 had become part of customary international law by examining
resolutions of the General Assembly, state practice and the writings of jurists. In the
course of this judgment, he stated: [60]
I am . . . prepared to accept that customary international law may . . . be created very quickly,
but before it will be considered by our municipal law as being incorporated into South African
law the custom, whether created by usus and opinio juris or only by the latter, would at the
very least have to be widely accepted.
Section 231(4) of the Interim Constitution provided that ‘the rules of customary
international law binding on the Republic shall, unless inconsistent with this Constitution
or an Act of Parliament, form part of the law of the Republic’ (emphasis added). The
omission of the word ‘binding’ from the 1996 Constitution has led one commentator to
argue that all rules of customary international law, including those to which South
Africa may have ‘persistently objected’, are part of
Page 70
municipal law. [61] This, so it is argued, accords with a 1995 dictum of the Constitutional
Court that the reference to international law in the Bill of Rights ‘includes non-binding
law as well’. [62]
The better view is that the word ‘binding’ was dropped from the 1996 Constitution on
the grounds that it was considered to be unnecessary and, indeed, tautologous. [63] As
far as South Africa is concerned, a practice to which it has persistently objected is
simply not a customary rule. On the other hand, there can be little doubt that the
omission of the word ‘binding’, with its undertones of strict consent, lends support to
the proposition that widespread or general acceptance, as opposed to universal
acceptance, is sufficient for proof of customary international law. This is the standard
set by the International Court of Justice [64] and there is no reason why a South African
court should demand a higher standard.
The Supreme Court of Appeal in Minister of Justice v SALC [65] (a matter which dealt
with the South African government’s failure to arrest Sudanese President al-Bashir
pursuant to two arrest warrants issued by the International Criminal Court) was
required to consider, for the purposes of domestic application, whether there was an
international crimes exception to the personal immunity of heads of state as a matter of
customary international law. Wallis JA held that one of the factors that a court would
consider when seeking to determine whether such an exception existed (together, inter
alia, with judgments by international courts) was whether there was an ‘established
universal practice in the affairs of nations’. [66] It is nevertheless clear from the
judgment, read as a whole, that Wallis JA did not intend to suggest that such ‘universal
practice’ was now to be regarded as the necessary standard for South African courts to
accept a customary international law rule (or an exception thereto). Wallis JA, in
concluding his analysis of the state of customary international law, emphasised that the
constitutional role of domestic courts in South Africa was not to develop customary
international law, but merely to ascertain the current state of that law and then to
‘apply it’. Wallis JA held:
Page 71
[O]ur task is to assess the state of customary international law as it stands at the present time
and apply it. That is what the Constitution requires us to do. While in other areas of the law
the court’s function includes the development of the law, in the area of customary international
law its task is one of discerning the existing state of the law, not developing it. As Lord
Hoffmann said in Jones:
‘It is not for a national Court to develop international law by unilaterally adopting a
version of that law which, however desirable, forward-looking and reflective of values it
may be, is simply not accepted by other states.’
 Development of customary international law occurs in international courts and tribunals, in the
contents of international agreements and treaties and by general acceptance by the
international community of nations in their relations with one another as to the laws that
govern that community. However tempting it may be to a domestic court to seek to expand
the boundaries of customary international law by domestic judicial decision, it is not in my view
permissible for it to do so. [67]
Wallis JA therefore made it clear that he accepted that if a rule is recognised by
international law as part of customary international law, then s 232 makes that rule law
in South Africa (save if it is inconsistent with the Constitution or an Act of Parliament).
Thus, South African courts are not entitled to include additional, more stringent
requirements for the recognition and application of customary international law in South
Africa than those accepted by international law itself. Conversely, it is not permissible
for a South African court to expand the boundaries of customary international law
beyond that recognised by international law. Therefore, s 232, as confirmed by Wallis
JA in Minister of Justice v SALC, effectively gives the Constitution’s imprimatur to
Conradie J’s statement in S v Petane that ‘where a rule of customary international law
is recognised as such by international law it will be so recognised by our
law’ [68] (obviously subject to the requirement of constitutional and legislative
consistency).
In National Commissioner v SALC, [69] a case which dealt with the state’s obligation
to investigate allegations of torture committed in Zimbabwe by Zimbabwean officials,
the Constitutional Court held that since torture is an international crime under
customary international law (its prohibition is a peremptory norm of international law) s
232 of the Constitution makes torture a crime in South Africa. [70]
Page 72

4 Treaties [71]

Before 1994, South Africa followed the dualist approach to the incorporation of treaties.
Treaties were negotiated, signed, ratified and acceded to by the executive. Only those
treaties incorporated by Act of Parliament became part of South African law. Thus,
treaty-making fell exclusively within the competence of the executive.
The 1993 Kempton Park negotiators were strongly motivated by considerations of
transparency and accountability—which had played little role in the apartheid state.
Thus, influenced by the Namibian Constitution, [72] they departed radically from the pre-
1993 position in respect of the treaty-making power and incorporation of treaties. While
the executive retained its power to negotiate and sign treaties under the Interim
Constitution, [73] the National Assembly and Senate were required to ‘agree to the
ratification of and accession to’ treaties. [74] Moreover, where the two houses of
Parliament agreed to the ratification of or accession to a treaty, it became part of
municipal law, ‘provided Parliament expressly so provides’. [75]
The clear purpose of the Interim Constitution was to facilitate the incorporation of
treaties into municipal law. The drafters of the Interim Constitution, however, failed to
take account of the bureaucratic mind. Government departments, required to scrutinise
treaties before they were submitted to Parliament, refused to present treaties to
Parliament for its agreement to the ratification until they were completely satisfied that
there would be no conflict between the provisions of the treaty and domestic law. The
result was that few treaties were presented to Parliament expeditiously. The
parliamentary procedures for dealing with treaties further delayed
ratification. [76] Consequently, few of the treaties which Parliament agreed to
Page 73
ratify were incorporated into municipal law. [77] The hopes of the drafters of the 1993
Interim Constitution were therefore not realised: the ratification of treaties proved to be
cumbersome and few treaties were incorporated into municipal law.
In these circumstances, the drafters of the 1996 Constitution elected to return to the
pre-1994 position relating to the incorporation of treaties, without abandoning the need
for parliamentary agreement to the ratification of or accession to treaties. Section 231
provides:
(1)
The negotiating and signing of all international agreements is the responsibility of the national
executive.
(2)
An international agreement binds the Republic only after it has been approved by resolution in
both the National Assembly and the National Council of Provinces, unless it is an agreement
referred to in subsection (3).
(3)
An international agreement of a technical, administrative or executive nature, or an agreement
which does not require either ratification of accession, entered into by the national executive,
binds the Republic without approval by the National Assembly and the National Council of
Provinces, but must be tabled in the Assembly and the Council within a reasonable time.
(4)
Any international agreement becomes law in the Republic when it is enacted into law by
national legislation; but a self-executing provision of an agreement that has been approved by
Parliament is law in the Republic unless it is inconsistent with the Constitution or an Act of
Parliament.
(5)
The Republic is bound by international agreements which were binding on the Republic when
this Constitution took effect.
Although this provision ensures that Parliament will continue to play an active role in
treaty-making, it is unfortunate that the realities of the bureaucratic process compelled
the Constitutional Assembly to require an Act of Parliament or other form of ‘national
legislation’, in addition to the resolution of approval, for the incorporation of treaties
into municipal law. It represents an abandonment of the idealism of 1993 that sought
‘to bring international law and domestic law in harmony with each other’. [78]
The Interim Constitution suggested that all treaties signed by the executive were to
be agreed to by Parliament before they could be ratified or acceded to on the
international plane (by depositing of the necessary instruments with the relevant treaty
depositary) or otherwise
Page 74
made binding on South Africa. [79] This took no account of the fact that many treaties
are intended to come into operation immediately, and that a slow parliamentary
approval process would undermine the value of such treaties. Consequently,
government departments ignored the letter of the Interim Constitution and
distinguished between ‘formal’ treaties that required parliamentary approval, and less
formal treaties that did not. [80] The 1996 Constitution recognises this distinction. While
treaties that expressly or by necessary implication require ratification have to be
approved by Parliament after signature, agreements of ‘a technical, administrative or
executive nature’ and agreements that do not require ratification or accession do not
require parliamentary approval to become binding, but must merely be tabled before
Parliament in a reasonable time. [81] In practice, this may give rise to disputes about the
precise meaning of the terms ‘technical’, ‘administrative’ or ‘executive’ in the context of
treaty law. [82]
The only case that has dealt directly with the question of whether a particular
international agreement is technical, administrative or executive in nature is Earthlife
Africa v Minister of Energy. [83] At issue in the matter was an inter-governmental
agreement in relation to nuclear co-operation and procurement between the Russian
and South African governments. The agreement had been tabled before Parliament
under s 231(3) of the Constitution. A two-judge bench of the Western Cape High Court
was required to determine whether the agreement had been lawfully tabled under s
231(3) or whether the government should have tabled it under s 231(2) in order to
seek parliamentary approval of the agreement. The parties accepted that the
agreement did not make express provision for ratification; therefore, the only issue was
whether it could rightly be classified as an agreement of a technical, administrative or
executive nature. In finding that it could not be so classified, the Court held as follows:
Page 75
The tabling of an IGA under s 231(3) permits the executive to bind South Africa to an
agreement without parliamentary approval or the public participation that often accompanies
any such parliamentary-approval process. Limiting those international agreements which may
be tabled under s 231(3) to a limited subset of run-of-the-mill agreements (or as Prof Dugard
puts it, agreements ‘of a routine nature, flowing from daily activities of government
departments’) which would not generally engage or warrant the focused attention or interest of
Parliament would give optimal effect to the fundamental constitutional principles of the
separation of powers, open and accountable government, and participatory democracy. [84]
The Court’s interpretation is consistent with the purpose for referring to agreements of
‘a technical, administrative or executive nature’ in s 231(3) since it has been noted that
‘[a]t the constitutional assembly not much weight was attributed to the differences
between these terms. It was rather the intention to provide for an expedited way of
dealing with the many minor every-day issues that can be the subject of agreement
between two states, and to save Parliament from having to consider technical and
departmental matters.’ [85] The approach taken by the Court also accords with that
adopted by state international law advisers. They ‘understand these terms to refer to
agreements of a routine nature, flowing from the daily activities of government
departments. The approach suggested by the Office of the President is that where there
is any doubt as to whether an agreement falls under s 231(3), the longer,
parliamentary route should be followed’. [86]
In Earthlife Africa, after considering the contents of the Russia–South Africa inter-
governmental agreement, the Court found:
[T]hree hallmarks of the agreement are its degree of specificity, the frequent use of
peremptory language and the scope and importance of key elements which form the bedrock
of the agreement. All these factors combine to suggest a firm legal commitment by the
contracting parties to the ‘strategic partnership’ which the agreement establishes between the
two countries, as well as in relation to the future steps and developments which the far-
reaching agreement clearly foreshadows. . . .
It may well be difficult to delineate the precise line between an agreement relating to the
procurement of new nuclear reactor plant as distinct from one dealing with cooperation
towards this end. In my view, however, seen as a
Page 76
whole, the Russian IGA stands well outside the category of a broad nuclear cooperation
agreement and, at the very least, sets the parties well on their way to a binding, exclusive
 agreement in relation to the procurement of new reactor plants from that particular
country. [87]
The Court therefore held that the inter-governmental agreement was not an agreement
of a technical, administrative or executive nature. Accordingly, the Court found that the
Minister of Energy’s decision to table the agreement under s 231(3) was
unconstitutional, and accordingly set the tabling aside.
Whether a treaty requires ratification is ultimately a question of intention. Generally,
this will be determined by the express inclusion of a requirement for ratification in the
treaty. [88] Where parties intend that an agreement is to come into force immediately,
without ratification at the international level, then parliamentary approval for the
executive to bind the Republic to the agreement at the international level would only be
required if, given the content of the agreement, it would not be a run-of-the-mill
agreement (those that are of a technical, administrative, or executive nature). This is
so since, whatever the intention of the executive in negotiating an international
agreement, the principle of separation of powers, woven into s 231, [89] requires that all
but run-of-the-mill agreements (that do not warrant Parliament’s attention) must be
approved by Parliament. Put differently, it would not be constitutionally permissible for
the executive, when negotiating a treaty, to intend to provide for the treaty to
immediately bind South Africa on signature if, given its content, the agreement did not
fall within the ambit of an agreement of a ‘technical, administrative, or executive
nature’. This is not surprising. Merely because the executive may desire to bind South
Africa to a substantive agreement without first having to obtain Parliament’s approval
(which could be withheld, and which must generally include a public participation
process) [90] does not
Page 77
mean that the executive is constitutionally empowered to do so. [91] Thus, as made
clear by Earthlife Africa, even in instances where a treaty does not expressly require
ratification, whether the executive can dispense with the requirements of s 231(2) will
depend on an objective assessment of the nature and terms of the treaty. Thus, s 231
ensures that it will only be a limited subset of agreements that can be made binding
absent parliamentary approval. As held by the Court in Earthlife Africa, s 231(3) (which
allows the requirement for parliamentary approval to be dispensed with) only deals with
agreements that would not ‘engage or warrant the focussed attention or interest of
Parliament’. [92] Similarly, a full bench of the North Gauteng High Court in DA v the
Minister of International Relations (which dealt with the lawfulness of the government’s
withdrawal from the Rome Statute of the International Criminal Court) held that when
Parliament approves international agreements it does so ‘on behalf of the people of
South Africa as their elected representative’. [93]
There is some academic debate as to whether an agreement that is expressly
subject to ratification might still be tabled under s 231(3) if it could be argued to
nevertheless be an agreement that was technical, administrative, or executive in
nature. [94] However, without considering the competing academic arguments, the North
Gauteng High Court in DA v the Minister of International Relations [95] appeared to
accept that where an international agreement requires ratification, it must be tabled
under s 231(2), since the Court held that ‘ratification . . . requires prior parliamentary
approval in terms of s 231(2)’. [96] In practice, it would be rare for a run-of-the-mill
international agreement that would not engage or warrant parliamentary interest and
attention (thus constituting an agreement of a technical, administrative or executive
nature) to nevertheless make provision for ratification. This is so since the general
reason why states parties to an international agreement provide that the agreement
will only become binding on ratification is precisely to allow the signatory states to
apply their minds to, and comply with, necessary domestic constitutional obligations,
prior to agreeing on the international plane to be bound by the agreement. [97]
In Earthlife Africa, the Court noted that even if an agreement might
Page 78
in principle fall within the terms of s 231(3), and therefore not require Parliament’s
approval, the government would, if it so chose, be entitled to make use of the more
onerous procedure in s 231(2) to obtain parliamentary approval in order to make the
agreement binding. [98] Therefore, to the extent there is any doubt as to whether an
agreement falls within the ambit of s 231(2) or (3), it would be advisable for the
government to adopt the more exacting procedure in s 231(2).
In DA v Minister of International Relations, the North Gauteng High Court also
established that, properly interpreted, s 231 requires that the executive must first
obtain parliamentary approval prior to withdrawing from a treaty which Parliament had
approved in terms of s 231(2). The Court held that since the depositing of a notice of
withdrawal had ‘concrete legal effects in international law, as it terminates treaty
obligations’, [99] prior parliamentary approval was required before the instrument of
withdrawal could be deposited by the executive. In this respect, the Court found that
the depositing of a notice of withdrawal was the equivalent of depositing an instrument
of ratification, which required prior parliamentary approval in terms of s
231(2). [100] The Court held this was so since ‘[i]f it is Parliament which determines
whether an international agreement binds the country, it is constitutionally untenable
that the national executive can unilaterally terminate such an agreement’. [101] The
Court also found that since the Rome Statute had been domesticated, a notice of
withdrawal could not be given, prior to the repeal of the domesticating
legislation. [102] The Court therefore held that the national executive’s decision to deliver
the notice of withdrawal without first obtaining parliamentary approval violated s
231(2) and the separation of powers doctrine underlying that section. [103] In making
this finding, the Court rejected the suggestion by government that since it had sought
parliamentary approval for the notice of withdrawal after it was deposited, this would
cure the unlawfulness. The Court held that ‘[w]hatever Parliament does about the
subsequent request to it by the national executive to approve the notice of withdrawal
would not cure its invalidity . . . [since] “[a]n invalid act, being a nullity, cannot be
ratified, validated or amended”’. [104] Furthermore, the
Page 79
Court held that the failure to first consult Parliament and obtain its approval prior to
depositing the notice of withdrawal also rendered the depositing of the notice
procedurally irrational. [105]
While a number of the parties [106] had argued that it was substantively irrational and
unlawful for the government to have decided to withdraw from the Rome Statute, the
Court found that it was unnecessary to determine any of the substantive challenges
since it had already found the notice of withdrawal to be unconstitutional and invalid on
procedural grounds. [107] The Court therefore declared the decision to sign and deliver
the notice of withdrawal to be unconstitutional and invalid. [108] It also ordered the
government to revoke the notice of withdrawal. [109] The government elected not to
seek leave to appeal the decision and duly revoked the notice of withdrawal.
 Section 231(4) represents a return to the pre-1994 position expounded by the
Appellate Division in Pan American World Airways. [110] An international agreement or
treaty does not become part of domestic law until it is enacted into law by national
legislation. [111] An Act of Parliament is ‘national legislation’ but the term also includes
(a)
subordinate legislation made in terms of an Act of Parliament; and
(b)
legislation that was in force when the Constitution took effect and that is administered by the
national government. [112]
Three principal methods are employed by the legislature to transform treaties into
municipal law. In the first instance, the provisions of a treaty may be embodied in the
text of an Act of Parliament; [113] secondly, the treaty may be included as a schedule to
a
Page 80
statute; [114] and thirdly, an enabling Act of Parliament may give the executive the
power to bring a treaty into effect in municipal law by means of proclamation or notice
in the Government Gazette. [115] Mere publication of a treaty for general information
does not constitute an act of transformation. [116] In S v Tuhadeleni the Appellate
Division refused to accept that the publication of the Mandate for South West Africa for
general information in the Official Gazette of the territory made the mandate part of the
law of the territory. [117]
The place of treaties in our law was summed up in the following terms by Ngcobo CJ
in Glenister v President of the Republic of South Africa:
An international agreement that has been ratified by resolution of Parliament is binding on
South Africa on the international plane. And failure to observe the provisions of this agreement
may result in South Africa incurring responsibility towards other signatory states. An
international agreement that has been ratified by Parliament under section 231(2), however,
does not become part of our law until and unless it is incorporated into our law by national
legislation. An international agreement that has not been incorporated in our law cannot be a
source of rights and obligations. [118]
It may have been more accurate for Ngcobo CJ to have said that: ‘An international
agreement that has been [approved] [119] by resolution of Parliament [and thereafter
ratified by the executive on the international plane in accordance with the requirements
of the agreement (usually by depositing the instrument of ratification with
Page 81
the depositary)] [120] is binding on South Africa on the international plane.’ Given the
wording of s 231(2), and having regard to the relevant domestic constitutional law and
international law contexts, once Parliament has approved an international agreement,
the executive is then obligated to convey South Africa’s agreement to be bound by the
treaty to the other states parties, in whatever way provided for in the treaty (usually by
depositing an instrument of ratification). [121] Both as a matter of practice and of law, it
is the executive and not Parliament that formally conducts South Africa’s foreign affairs
on the international plane, and it is therefore the executive that takes the necessary
international steps (once empowered by Parliament) [122] to advise other countries of
South Africa’s agreement to be bound by an international agreement. [123] As the North
Gauteng High Court held in DA v Minister of International Relations, ‘prior
parliamentary approval is required before instruments of ratification may be deposited
with the United Nations’. [124]

5 Self-executing treaties
The proviso to s 231(4) introduces the concept of self-executing treaties—that is,
treaties that automatically become part of municipal law, and enforceable by municipal
courts, without any act of legislative incorporation—into South African law. The
provisions of a treaty approved by Parliament, but not incorporated into municipal law
by an Act of Parliament, that are self-executing become part of municipal law unless
inconsistent with the Constitution or an Act of Parliament. It is important to emphasise
that s 231(4) only provides for self-executing treaties to become municipal law in the
absence of legislation if they have been ‘approved by Parliament’. This refers to treaties
that have been approved by both houses of Parliament in terms of s 231(2). Treaties
covered by s 231(3) that have not been
Page 82
approved by Parliament will not have domestic force even if they could be argued to be
self-executing.
Whether the provisions of a treaty are self-executing has troubled the courts of the
United States for many years. [125] In 1951, Professor Myres McDougal declared in
respect of the position of the United States: ‘[T]his word self-executing is essentially
meaningless, and . . . the quicker we drop it in our vocabulary the better for clarity and
understanding.’ [126] The soundness of this advice was confirmed by the US Supreme
Court in Medellin v Texas, [127] in which the Court divided sharply over the meaning of
self-executing treaties in a case involving the enforcement of a judgment of the
International Court of Justice holding that an alien sentenced to death in the United
States was entitled to review of his sentence. [128] American law therefore offers no
panacea for the problems that are likely to confront South African courts in the
interpretation of s 231(4).
Scholars have expressed divergent views on self-executing treaties. Botha states
that s 231(4) was taken over unwisely from US jurisprudence ‘with no regard to its
suitability to the South African context’. [129] Other writers have described it as
‘nonsensical’ [130] and ‘farcical’. [131] Ngolele [132] and Olivier [133] adopt a different
position, in favour of the concept of self-executing treaties, and suggest that human
rights treaties, such as the International Covenant on Civil and Political Rights (ICCPR),
may be directly applied by South
Page 83
African courts as self-executing treaties. [134] Ngolele rejects the view that the adequacy
of existing South African law is a condition precedent for self-execution of a treaty.
However, in acknowledging that self-execution has limited effect in South Africa,
because treaties will be self-executing only if consistent with the Constitution and Acts
of Parliament, he comes close to accepting that municipal law must be adequate for a
treaty to be self-executing, in the sense that it must not obstruct the application of
such a treaty. Olivier adopts a more radical position. She argues that in terms of s
231(2), a treaty approved by Parliament has direct application in South African law
without the need for incorporation—that is, it is self-executing—provided that it is not
inconsistent with the Constitution or an Act of Parliament. Olivier’s thesis is based on a
false assumption—namely that s 231(2) is designed to give domestic effect to treaties
approved by Parliament. [135] Section 231(2) is intended to establish that an
international agreement binds South Africa on the international level only after it has
been approved by both houses of Parliament. To apply domestically, a treaty must still
be ‘enacted into law by national legislation’, as specified in s 231(4), unless it is self-
executing. In addition, it will be self-executing only if the language of the treaty so
indicates and existing municipal law, either common law or statute, is adequate in the
sense that it fails to place any obstacle in the way of treaty application.
Judicial decisions likewise demonstrate a divergence of opinion. Nello Quagliani v
President of the RSA [136] concerned the question whether the 1999 extradition
agreement between the United States and South Africa was part of South African law,
despite the fact that it had not been incorporated into municipal law by national
legislation. In considering whether the agreement qualified as a self-executing treaty
and was thus part of South African law in terms of the proviso to s 231(4), Preller J
stated that he found s 231(4) ‘hard to understand’, [137] ‘at most of academic interest to
us’ in South Africa, [138]
Page 84
and ‘foreign to our legal system’. [139] He then found that the agreement in question had
not been validly incorporated in terms of s 231(4) and was not part of South African
law: that is, it was not self-executing. He stated ‘[f]rom a reading of the plain words of
section 231(4) of the Constitution it is simply not possible to have a statute in terms of
which any number of international agreements can subsequently be concluded that will
have the force of law in the Republic. What the plain language of the sub-section
requires is the enactment into law of every new treaty. In my view that clearly means a
new Act of Parliament for every new treaty.’ [140] In Goodwin v Director-General
Department of Justice and Constitutional Development, [141] however, Ebersohn AJ took
a completely different line in respect of the same treaty in holding that the treaty ‘is a
self-executing provision in its totality’ [142] and that Preller J’s decision in Quagliani was
‘clearly wrong’. [143]
The Constitutional Court refused to provide any guidance on the meaning to be given
to self-executing treaties in the appeals from Quagliani and Goodwin [144] when it found
that it was unnecessary to consider the question whether the extradition agreement
between the United States and South Africa should be regarded as self-
executing. [145] In making this finding, Sachs J stated on behalf of a unanimous court:
The question then is whether the Agreement ‘becomes law’ in South Africa as contemplated by
section 231(4) of the Constitution. There are two ways in which this question can be answered.
The first is to say that the Agreement itself does not become binding in domestic law, but the
international obligation the Agreement encapsulates is given effect to by the provisions in the
[Extradition] Act. The second approach is that once the Agreement has been entered into as
specified in sections 2 and 3 of the [Extradition] Act, it becomes law in South Africa as
contemplated by section 231(4) of the Constitution without further legislation by Parliament.
It is not necessary for the purposes of this case to decide which of these approaches is
correct, for their effect in this case is the same. Either the Agreement has ‘become law’ in
South Africa as a result of the prior existence of the [Extradition] Act which constitutes the
anticipatory enactment of the Agreement for the purposes of s 231(4) of the Constitution. Or
the Agreement
Page 85
has not ‘become law’ in the Republic as contemplated by s 231(4) but the provisions of the Act
are all that is required to give domestic effect to the international obligation that the
Agreement creates.
I conclude, therefore, that on either of the approaches identified above, no further
enactment by Parliament is required to make extradition between South Africa and the United
States permissible in South African law. [146]
This dictum is incomprehensible. [147] Section 231(4) makes it clear that for a treaty to
become law in South Africa it must be enacted into law by national legislation unless it
is self-executing. There is no third avenue for the incorporation of treaties. The
Constitution makes no exception to s 231(4) for extradition agreements, such as that
between the United States and South Africa, which had not been ‘enacted into law by
national legislation’. The highly convoluted reasoning of the Constitutional Court fails to
satisfactorily explain why extradition agreements come into existence by means not
contemplated by s 231(4). It is obviously undesirable to have to enact legislation for
every extradition agreement. But the solution is either to amend the Extradition Act 67
of 1962 or to consider extradition agreements to be self-executing. Clearly, amending
legislation is the most desirable course. Indeed, this was proposed in earlier editions of
the present study [148]—a proposal that was not considered by the Constitutional Court.
The result of these appeals is that the Court has given an opaque and confusing
interpretation of s 231(4) and failed to throw any light on the meaning to be attached
to the term ‘self-executing’.
South African courts are reluctant to explore the meaning of the term ‘self-
executing’. This is abundantly clear from the decision of the Constitutional Court
in Quagliani. It is also clear from the statement of Binns-Ward J in Claassen v Minister
of Justice and Constitutional Development that:
The ICCPR is not a self-executing legal instrument, in the sense that this country’s formal
adoption of its provisions did not, without more, amend our established domestic law. It seems
to me that the current case illustrates the need, if unqualified effect is to be given to article
9(5) of the ICCPR, for South Africa to enact legislation . . . . [149]
Courts must, however, address the meaning to be given to self-executing treaties in s
231(4) and not pretend that the proviso to s 231(4) does not exist or argue that
treaties can be incorporated into
Page 86
municipal law by means other than ‘national legislation’—as was done by the
Constitutional Court in Quagliani. The fact that American courts have had difficulty in
satisfactorily defining the term self-executing in the context of American law should not
deter South African courts from attaching meaning to the term in the context of South
African law. There is much to be said for identifying an extradition treaty and the ICCPR
as ‘self-executing’ as there is already legislation in place to enable them to be applied in
municipal law and they are not inconsistent with the Constitution or an Act of
Parliament. But, unfortunately, courts, with the exception of the lower courts
in Quagliani and Goodwin, have refused to address this issue satisfactorily. If full
meaning is to be given to s 231(4), it is essential that courts consider this issue in a
manner which promotes harmony between South Africa’s international obligations and
its municipal law.
******
A treaty that has been signed and ratified, but not enacted into local law, is binding on
South Africa on the international plane. Failure to observe the provisions of such a
treaty may result in South Africa incurring responsibility towards other states parties.

6 The national executive and the making of treaties

Section 231(1) confers on the ‘national executive’ the responsibility for the making of
treaties. The President, as head of state and head of the national executive, may be
given the power to enter into treaties, such as extradition treaties, in terms of a specific
Act, but he is obliged (given s 85(2)(e) read with s 231(1) of the Constitution) to act ‘in
a collaborative manner’ with the other members of the Cabinet in exercising this
authority. [150]
When the President enters into an agreement he does so as head of the national
executive. In the words of the Constitutional Court:
Given the provisions of section 231 of the Constitution, it is not improper for the President,
once the decision to enter into the treaty has been made by the President, to confer other
formal aspects relating to the accession to the treaty on other members of the national
executive. It is important that these provisions should not be applied in a formalistic manner
that will impair the ability of the national executive to function. [151]
In DA v Minister of International Relations, the North Gauteng High Court held that s
231(1) only gives the national executive a preliminary power to undertake the
‘exploratory work’ of negotiating and signing international agreements, but this does
not empower the
Page 87
executive to bind South Africa to such international agreements; ‘[t]he binding power
comes only once parliament has approved the agreement’ [152] (although, the Court was
only dealing with treaties that require parliamentary approval in terms of s 231(2)).

7 International agreements or treaties

The 1996 Constitution, like the Interim Constitution, uses the term ‘international
agreement’ instead of the more commonly used term ‘treaty’. This creates uncertainty
over the meaning of s 231, as there is strong support for the view that the term
‘international agreement’ is wide enough to include both legally binding agreements
(treaties) and non-binding, unenforceable informal agreements. [153] This uncertainty
surfaced in the Harksen case [154] in which it was argued that s 3(2) of the Extradition
Act, [155] authorizing the President to consent to the extradition of a person to a state
with which South Africa has no extradition agreement, gives rise to an international
agreement if not a treaty. Consequently, the procedure for the adoption of international
agreements contained in s 231(2) should be followed in respect of such an undertaking
by the President. While the Cape Provincial Division was prepared to contemplate the
existence of unenforceable, informal arrangements falling within the scope of an
‘international agreement’, [156] the Constitutional Court rejected the argument that a
presidential undertaking under s 3(2) of the Extradition Act could be categorized as
either an international agreement or as an informal agreement. [157] Unfortunately, the
Constitutional Court, in a judgment characterized by its brevity and a determination not
to consider the broader issues relating to treaty-making, failed to throw light on the
meaning to be given to ‘international agreement’. [158]
The prevailing view, however, is that the term ‘international agreement’ in s 231 is
synonymous with ‘treaty’ and refers to legally binding, enforceable agreements as
defined in article 2 of the Vienna
Page 88
Convention on the Law of Treaties of 1969. [159] According to this provision, a ‘treaty’ is
an international agreement concluded between states in written form and governed by
international law, whether embodied in a single instrument or in two or more related
instruments and whatever its particular designation [emphasis added].
‘Terminology is not a determinant factor as to the character of an international
agreement’—in the words of the International Court of Justice. [160] What is important is
that the agreement be between states, in writing, and that the state parties intend it to
be governed by international law. [161] Once these requirements are met, an
international agreement exists between the state parties and it matters not whether it
is called an ‘international agreement’ or ‘treaty’—or, for that matter, convention,
declaration, act, concordat, protocol, memorandum of understanding or exchange of
notes.
No provision is made for the recognition of oral agreements in the Constitution, nor
is express provision made for agreements with international organisations. [162]

8 Resolutions of international organisations

Resolutions of international organisations are not treaties and, in most instances, they
are not binding on member states. If South Africa wishes to translate such a resolution
into municipal law, it must do so by legislation. [163] Resolutions of the Security Council
of the United Nations adopted under Chapter VII of the Charter are, however, binding
on member states. The Application of Resolutions of the Security Council of the United
Nations Act [164] empowers the President to incorporate resolutions of the Security
Council into
Page 89
municipal law by proclamation in the Government Gazette and to provide for the
implementation of such resolutions under South African law. This Act has, however, not
yet been brought in force. [165]

9 Statutory (including constitutional) interpretation with

special reference to human rights

The 1996 Constitution reveals a clear determination to ensure that the Constitution and
South African law are interpreted to comply with international law, particularly in the
field of human rights. First, the common-law presumption requiring a court to interpret
legislation in compliance with international law [166] is given constitutional form in s 233,
which provides:
When interpreting any legislation, every court must prefer any reasonable interpretation of the
legislation that is consistent with international law over any alternative interpretation that is
inconsistent with international law.
Secondly, the Bill of Rights, which is modelled on international human rights
conventions and on occasion refers directly to international law, [167] is subjected to a
special interpretative regime which pays particular attention to international law.
Section 39 (s 35 of the Interim Constitution) [168] declares:
(1)
When interpreting the Bill of Rights, a court, tribunal or forum—

(a)

must promote the values that underlie an open and democratic society based on human
dignity, equality and freedom;

(b)

must consider international law; and

(c)

may consider foreign law

(2)
When interpreting any legislation, and when developing the common law or customary law,
every court, tribunal or forum must promote the spirit, purport and objects of the Bill of
Rights.
(3)
The Bill of Rights does not deny the existence of any other rights or freedoms that are
recognized or conferred by common law, customary law or legislation, to the extent that they
are consistent with the Bill.
Fears that international human rights law might be narrowly construed to cover only
clear rules of customary law and those
Page 90
human rights conventions to which South Africa is a party have been dispelled. [169] In
one of its earliest decisions, in a case involving the constitutionality of the death
penalty, the President of the Constitutional Court ruled:
In the context of s 35(1), public international law would include non-binding as well as binding
law. They may both be used under the section as tools of interpretation. International
agreements and customary international law accordingly provide a framework within which
[the Bill of Rights] can be evaluated and understood, and for that purpose, decisions of
tribunals dealing with comparable instruments, such as the United Nations Committee on
Human Rights, the Inter-American Commission on Human Rights, the Inter-American Court of
Human Rights, the European Commission on Human Rights, and the European Court of Human
Rights and, in appropriate cases, reports of specialised agencies such as the International
Labour Organisation, may provide guidance as to the correct interpretation of particular
provisions of [the Bill of Rights]. [170]
Since the establishment of the new constitutional order in 1994, both the Constitutional
Court and ordinary courts have shown a great willingness to be guided by international
human rights law. Decisions of the European Commission and Court of Human Rights
have provided the greatest assistance, but courts have on occasion also considered the
‘views’ of the United Nations Human Rights Committee, and United Nations reports on
human rights matters. [171]
Page 91
The manner in which South African courts have invoked international human rights
norms is considered in Chapter 15.
While a court must consider treaties to which South Africa is not a party in
interpreting the Bill of Rights, no such rule exists in respect of treaties to which South
Africa is not a party where the Bill of Rights is not in issue. A treaty to which South
Africa is not a party is res inter alios acta and may not be considered qua treaty,
although it may be considered as evidence of a customary rule. [172]
Different considerations apply in respect of a treaty to which South Africa is a party
but has not been incorporated into municipal law. In the first instance, a municipal
court may have recourse to an unincorporated treaty in order to interpret an ambiguous
statute. [173] However, the Constitution’s interpretative injunction, contained in s 233,
applies to all statutes, not only those that are ambiguous. The Constitutional Court’s
decision in S v Okah makes it clear that s 233 of the Constitution requires courts to
favour the interpretation of legislation that is in line with international law (including
unincorporated treaties binding on South Africa) even where another reasonable
interpretation of the legislation is possible but not in line with
Page 92
international law. [174] S v Okah involved the state’s prosecution of a Nigerian national,
who was resident in South Africa, for terrorist activities in Nigeria pursuant to South
Africa’s domestic anti-terrorism legislation, which provided for extraterritorial
jurisdiction. Cameron J held that the Court was required, in terms of s 233 of the
Constitution, to interpret the relevant provisions of the Protection of Constitutional
Democracy against Terrorist and Related Activities Act [175] in conformity with
international law. In determining what the relevant international law was, Cameron J
took account of a number of international treaties to which South Africa was party—
including treaties which had not been domesticated [176]—that established a duty to
prosecute or extradite. Cameron J pointed out that even if the Supreme Court of
Appeal’s interpretation of the legislation were reasonable, the Court was required by s
233 to interpret the legislation in line with international law. The Court rejected the
interpretation adopted by the Supreme Court of Appeal, which ‘would create instances
in which it would be possible to “prosecute the banker, but not the bomber”’, [177] and
adopted an interpretation which gave effect to South Africa’s international law
obligation to prosecute or extradite.
Secondly, an unincorporated treaty may be taken into account in a challenge to the
validity of delegated legislation on grounds of unreasonableness. [178] This was
confirmed by the Supreme Court of Appeal in Progress Office Machines CC v
SARS [179] when it held that delegated legislation, in the form of a notice by the Minister
of
Page 93
Finance imposing an anti-dumping duty, must be reasonable and that a court might
have regard to the state’s obligations under an unincorporated treaty—the WTO
Agreement of 1995—in making such a determination. (In casu the Court found that the
notice was unreasonable and, hence, invalid as it violated the WTO Agreement.)
Despite earlier failure of the Appellate Division to commit itself on this subject, [180] the
finding in Progress Office Machines is sound since the concept of reasonableness is
inextricably linked with presumptions of legislative intent and there is a presumption
that the legislature, in enacting a law, did not intend to violate South Africa’s
international obligations.
As South African courts are given the power of judicial review of legislation under the
1996 Constitution, [181] it is inevitable that international law will be invoked not only as
a guide to statutory interpretation but as a challenge to the validity of legislation. This
may take the form of a direct challenge where, for example, it is argued that the
procedures for ratification and incorporation of a treaty under s 231 have not been
followed. Or it may assume the form of an indirect challenge where international law is
invoked to support an interpretation in favour of the unconstitutionality of a statute.
The former type of direct challenge in relation to a failure to comply with the procedure
in s 231 is illustrated, to an extent, by Earthlife Africa (albeit that there had not been
any domestic incorporation of the relevant international agreement). As shown above,
in this case the Court set aside the tabling of an inter-governmental agreement
between Russia and South Africa in relation to nuclear procurement and co-operation
under s 231(3) of the Constitution because the international agreement was found,
given its substantive content, to be an agreement that fell under s 231(2), which
required parliamentary approval to be made binding. The Court also set aside the
tabling of inter-governmental agreements with the United States and South Korean
governments because of the significant delays (20 years and five years respectively) in
tabling them under s 231(3). The Court held that s 231(3) could only be used to table
international
Page 94
agreements to make them binding absent parliamentary approval (as required by s
231(2)) if they were tabled within a ‘reasonable time’ (as expressly required by s
231(3)) since this was a jurisdictional requirement for the use of s 231(3). Similarly,
in DA v Minister of International Relations (discussed above) the Court set aside the
government’s decision to deposit a notice withdrawing South Africa from the Rome
Statute (and ordered the government to revoke the notice) since the government, in
violation of s 231, had failed to first obtain parliamentary approval for the withdrawal
and had failed to repeal the Act that incorporated the Rome Statute into domestic law.
Two major decisions of the Constitutional Court illustrate the indirect type of
challenge: AZAPO v President of the Republic of South Africa, [182] decided under the
Interim Constitution of 1994, which contained no provision equivalent to s 233 of the
1996 Constitution, and Glenister v President of the Republic of South Africa. [183]
In AZAPO the applicants sought to set aside s 20(7) of the Promotion of National
Unity and Reconciliation Act, [184] providing for amnesty from criminal and civil
proceedings, on the ground that it was inconsistent with s 22 of the Interim
Constitution, which provides that every person shall have the right to have justiciable
disputes settled by a court of law or, where appropriate, another independent or
impartial forum. In support of this challenge, the applicants argued that the state was
obliged by international law, particularly the Geneva Conventions of 1949, to which
South Africa is a party but have not been incorporated into domestic law, to prosecute
those responsible for gross human rights violations and that the provisions of s 20(7),
which authorized amnesty for such offences, constituted a breach of international
law. [185]
In an eloquent judgment, written by Mahomed DP, the Constitutional Court held that
the epilogue to the Interim Constitution, providing for amnesty, trumped s 22 of the
Constitution and that
Page 95
s 20(7) of the Promotion of National Unity and Reconciliation Act, authorizing criminal
and civil amnesty, was therefore constitutional. From the perspective of international
law, the judgment is disappointing because it fails to adequately address the question
whether conventional and customary international law oblige a successor regime to
punish the officials and agents of the prior regime for international crimes and, thus,
gave support to the constitutional challenge advanced by the applicants. From the
perspective of the place of international law in the new constitutional order, the
judgment is disconcerting because of a strange dictum which may be construed as
suggesting that a ‘proper interpretation’ of the Constitution should be sought without
recourse to international conventions.
In the course of his judgment, Mahomed DP stated:
The issue which falls to be determined in this Court is whether s 20(7) of the Act is
inconsistent with the Constitution. If it is, the enquiry as to whether or not international law
prescribes a different duty is irrelevant to that determination. International law and the
contents of international treaties to which South Africa might or might not be party at any
particular time are, in my view, relevant only to the interpretation of the Constitution itself, on
the grounds that the lawmakers of the Constitution should not lightly be presumed to authorise
any law which might constitute a breach of the obligations of the state in terms of international
law. . . . [186]
 The exact terms of the relevant rules of public international law contained in the Geneva
Conventions relied upon on behalf of the applicants would therefore be irrelevant if, on a
proper interpretation of the Constitution, s 20(7) of the Act is indeed authorized by the
Constitution, but the content of these Conventions in any event does not assist the case of the
applicants. [187]
Fortunately, the place of treaties in a challenge to the constitutionality of legislation was
clarified in Glenister v President of the Republic of South Africa, [188] which considered
the constitutionality of legislation that brought into being the Directorate for Priority
Crime Investigation (DPCI) (popularly known as the ‘Hawks’) [189] and disbanded the
Directorate for Special Operations (popularly known as the ‘Scorpions’). [190] Moseneke
DCJ and Cameron J, writing for the majority, held that the legislation in question was
unconstitutional because the specialized unit it created to combat corruption failed to
meet the requirement of independence by not providing for secured conditions of
employment and by not removing executive oversight. Parliament was given 18 months
to remedy this constitutional defect. The Court held that corruption ‘undermines the
democratic ethos,
Page 96
the institutions of democracy, the rule of law and the foundational values of our nascent
constitutional project’. [191] Corruption, said the Court, stunts sustainable development
and economic growth and the state becomes unable to fulfil all the rights enshrined in
the Bill of Rights, particularly economic and social rights.
In reaching this conclusion, the Constitutional Court relied heavily on international
conventions aimed at combating corruption, particularly the United Nations Convention
against Corruption, [192] the African Union Convention on Preventing and Combating
Corruption [193] and the Southern African Development Community (SADC) Protocol
Against Corruption, [194] which South Africa has ratified. These conventions are given
effect to by the Prevention and Combating of Corrupt Activities Act 12 of 2004, which in
its preamble acknowledges the SADC Protocol against Corruption and expressly
declares that South Africa ‘desires to be in compliance with and become Party to the UN
Convention against Corruption’. The obligation contained in s 7(2) of the Constitution,
said the Court, requires the state to respect, protect and fulfil the rights in the Bill of
Rights and this creates the duty to create an independent and efficient anti-corruption
mechanism. The content of this constitutionally imposed requirement of independence
is to be found in the international agreements on corruption that bind South Africa
internationally although they have not been incorporated into South African law. The
anti-corruption agreements bind South Africa in terms of s 231(2) of the Constitution
and this had ‘significant impact in delineating the state’s obligations in protecting and
fulfilling the rights in the Bill of Rights’. [195] The Court continued:
The obligations in these Conventions are clear and they are unequivocal. They impose on the
Republic the duty in international law to create an anti-corruption unit that has the necessary
independence. That duty exists not only in the international sphere and is enforceable not only
there. Our Constitution appropriates the obligation for itself, and draws it deeply into its heart,
by requiring the state to fulfil it in the domestic sphere. In understanding how it does so, the
starting point is s 7(2), which requires the state to respect, protect, promote and fulfil the
rights in the Bill of Rights . . . . Implicit in s 7(2) is the requirement that the steps the state
takes to respect, protect, promote and fulfil constitutional rights must be reasonable and
effective . . . . [196]
Page 97
And it is here where the courts’ obligation to consider international law when interpreting
the Bill of Rights is of pivotal importance. Section 39(1)(b) states that when interpreting the
Bill of Rights a court ‘must consider international law’. The impact of this provision in the
 present case is clear, and direct. What reasonable measures does our Constitution require the
state to take in order to protect and fulfil the rights in the Bill of Rights? That question must be
answered in part by considering international law. And international law, through the
interlocking grid of conventions, agreements and protocols we set out earlier, unequivocally
obliges South Africa to establish an anti-corruption entity with the necessary independence.
That is the duty this country itself undertook when it acceded to these international
agreements. And it is an obligation that became binding on the Republic, in the international
sphere, when the National Assembly and the NCOP by resolution adopted them, more
especially the UN Convention.
That the Republic is bound under international law to create an anti-corruption unit with
appropriate independence is of the foremost interpretive significance in determining whether
the state has fulfilled its duty to respect, protect, promote and fulfil the rights in the Bill of
Rights, as s 7(2) requires. Section 7(2) implicitly demands that the steps the state takes must
be reasonable. To create an anti-corruption unit that is not adequately independent would not
constitute a reasonable step. In reaching this conclusion, the fact that s 231(2) provides that
an international agreement that Parliament ratifies ‘binds the Republic’ is of prime significance.
It makes it unreasonable for the state, in fulfilling its obligation under s 7(2), to create an anti-
corruption entity that lacks sufficient independence.
This is not to incorporate international agreements into our Constitution. It is to be faithful
to the Constitution itself, and to give meaning to the ambit of the duties it creates in
accordance with its own clear interpretive injunctions. The conclusion that the Constitution
requires the state to create an anti-corruption entity with adequate independence is intrinsic to
the Constitution itself . . . . [197]
A further provision of the Constitution that integrates international law into our law
reinforces this conclusion. It is s 233, which . . . demands any reasonable interpretation that is
consistent with international law when legislation is interpreted. There is, thus, no escape from
the manifest constitutional injunction to integrate, in a way the Constitution permits,
international law obligations into our domestic law. We do so willingly and in compliance with
our constitutional duty. [198]
Section 39(2) of the Constitution requires a court, when it develops the common law or
customary law, to ‘promote the spirit, purport and objects of the Bill of Rights’.
In Carmichele v Minister of Safety and Security (Centre for Applied Legal Studies
Intervening) [199] the Constitutional Court invoked decisions of the European Court of
Human Rights, [200] the Convention on the Elimination of All Forms of
Page 98
Discrimination against Women, [201] and United Nations guidelines (’soft law’) [202] to
develop a new rule of common law. Commenting on this decision, Neville Botha writes
that the Constitutional Court has
shown clearly that the spirit, purport and objects of the Bill of Rights—which reflects the
underlying precepts of the Constitution and the fabric of South African society—are inextricably
linked to international law and the values and approaches of the international community and
international role players. [203]
The Supreme Court of Appeal later followed the approach of the Constitutional
Court. [204]
In Government of Zimbabwe v Fick and Others [205] the Constitutional Court had to
determine whether a cost order granted by the SADC Tribunal against the Zimbabwean
government in favour of certain Zimbabwean farmers could be enforced in South Africa.
The cost order had been granted by the Tribunal in a matter where the farmers
challenged the Zimbabwean government’s expropriation of their land without
compensation. The Constitutional Court relied on South Africa’s obligations under the
SADC Treaty [206] and Tribunal Protocol [207] to develop the common-law rule allowing for
the enforcement of foreign court judgments to include the enforcement of judgments
by the SADC Tribunal. [208] In particular, Mogoeng CJ, writing for the majority, held that
‘[t]he Amended [SADC] Treaty, incorporating the Tribunal Protocol, places an
international law obligation on South Africa to ensure that its citizens have access to
the Tribunal and that its decisions are enforced. Section 34 of the Constitution must
therefore be interpreted, and the common law developed, so as to grant the right of
access to our courts to facilitate the enforcement of the decisions of the Tribunal in this
country. This, as said, will be achieved by regarding the Tribunal as a foreign court, in
terms of our common law.’ [209]
An increasing number of statutes refer expressly to international
Page 99
law and some make it clear that the statute is to be interpreted to accord with
international law. For instance, the Promotion of Equality and Prevention of Unfair
Discrimination Act [210] provides that any person interpreting the Act may be ‘mindful’ of
international law, [211] and the Implementation of the Rome Statute of the International
Criminal Court Act [212] provides that a court applying the Act must consider both
conventional and customary international law. [213] Such statutes confirm the principle
expounded in s 233 of the Constitution that legislation is to be interpreted in
accordance with international law. Of course, given the interpretative obligation
imposed by s 233 of the Constitution, the inclusion in statutes of a specific requirement
to have regard to international law is, to an extent, unnecessary.

10 Hierarchy of treaties in municipal law

The Constitution is the supreme law of South Africa. A treaty enacted into law by
national legislation in accordance with s 231(4) of the Constitution will enjoy the status
accorded to it by the act of incorporation: a treaty enacted into law by an Act of
Parliament will be treated as an Act of Parliament, whereas a treaty enacted into law by
subordinate legislation will be treated as subordinate legislation. A non-self-executing
treaty binding on South Africa internationally but not incorporated into municipal law
will have no direct force of law. However, (a) national legislation must be interpreted in
light of and in line with such non-self-executing treaties, as made clear by s 233 of the
Constitution and confirmed by the Constitutional Court, inter alia, in National
Commissioner v SALC [214] and in S v Okah; [215] (b) such treaties may be used as part
of the basis upon which to challenge legislation and government action, along the lines
indicated in Glenister v President of the Republic of South Africa, [216] Progress Office
Machines CC v SARS, [217] and Law Society of South Africa v President of the Republic of
South Africa (discussed below); [218] and (c) such non-self-executing treaties may be
used as the basis for developing the common law, as was done in Government of
Zimbabwe
Page 100
v Fick. [219] A self-executing treaty will obviously, in terms of s 231(4), give way to both
the Constitution and an Act of Parliament. Such a self-executing treaty will probably
take priority over delegated legislation, in the event of a conflict.

11 The executive certificate [220]

The executive is responsible for the conduct of South Africa’s foreign relations, and, in
the exercise of this function, it will frequently make decisions on subjects governed by
international law. These include [221] the recognition of a foreign state or
government, [222] the recognition of territorial acquisitions by another state, the
commencement or termination of a state of war with another country, [223] whether or
not a person is entitled to diplomatic status, [224] whether any territory is a constituent
part of a federation and whether any person is to be regarded as head of state or
government of a foreign state. [225] Obviously, it is undesirable that different organs of
state should pronounce on the same subject, particularly if their assessment of the
legal implications of the matter should differ. As Lord Atkin said in Government of the
Republic of Spain v SS ‘Arantzazu Mendi’ (The Arantzazu Mendi), [226] ‘[o]ur state
cannot speak with two voices . . . the judiciary saying one thing, the executive
another’. Consequently, in order to avoid confusion of this kind, the courts have
deferred to the judgment of the executive on certain acts or facts of state. The
judgment of the executive is generally given in an executive certificate handed in to
court, and the effect of such a certificate is ‘to substitute the view of the government
for an independent judicial investigation into the factual position’. [227] In this way, the
executive in effect seeks to usurp the power of a municipal court to apply rules of
customary international law to a particular factual situation that comes before it.
Page 101
The leading case on the subject of an executive certificate is S v Devoy in which the
Natal Provincial Division [228] and the Appellate Division [229] considered the continued
existence of an extradition agreement between South Africa and Nyasaland (later
Malawi) after the dissolution of the Federation of Rhodesia and Nyasaland. Here, the
Department of Justice issued a certificate stating that, as far as the South African
government was concerned, the treaty continued to be in force after the dissolution of
the Federation and that its validity was unaffected by Nyasaland’s reversion to colonial
status and subsequent emergence as the independent state of Malawi. Although the
capacity of Nyasaland to enter into treaties in the pre-independence period was
questioned, both the Court a quo and the Appellate Division held that they were bound
by the executive certificate on this subject and could not ‘go behind’ it. [230] In the
Court a quo, James JP stated: [231]
It is clear from the certificate produced from the Minister of Justice the South African
Government recognised the competence of Nyasaland after its secession from the Federation
to be a party to an extradition agreement entered into by the government of the Federation.
This recognition is a function of the executive branch of government; it is a political act
entailing legal consequences. Once that recognition has been granted by the executive branch
of any country it is not for the judicial branch to consider whether that recognition was
competent.
In Harksen v President of the Republic of South Africa, [232] a case decided under the
Interim Constitution of 1993, the Cape Provincial Division was confronted with an
executive certificate from the Minister of Justice declaring that there was no extradition
agreement between South Africa and the Federal Republic of Germany. The Court noted
that ‘as it is undesirable that different organs of the state should pronounce on the
same subject, particularly if their assessments of the legal implications of the matter
should differ . . . it may well be that this Court should accept the Minister of Justice’s
certificate as binding on it’. [233] However, with no consideration of the law governing
the conclusiveness or otherwise of an executive certificate, the Court stated:
Having regard to the view which we take of this matter, it is unnecessary to decide whether
the certificate by the Minister of Justice is binding on the
Page 102
Court and we accordingly proceed on the basis that it is not. [234]
The court then embarked on a thorough examination of extradition treaty relations
between South Africa and Germany, and of the law on this subject, from which it
concluded that there was indeed no valid treaty in existence.
The source of the executive’s power to issue a conclusive (or prima facie) certificate
has a statutory basis in the case of immunities conferred on foreign diplomats and
other foreign state representatives [235] and on foreign states. [236] In other cases, the
executive arguably relies on its non-statutory discretionary common-law powers to
conduct foreign relations as neither the Interim Constitution [237] nor the 1996
Constitution [238] appear to expressly confer powers on the executive in matters such as
the recognition of foreign states or governments or the determination of the existence
of treaties between South Africa and foreign states. The continued validity of such
powers, sometimes described as prerogative powers, is highly questionable, [239] and
the suggestion that the executive retains any of the old prerogative powers other than
where those are now provided for in the Constitution or in legislation has been
conclusively rejected by the Constitutional Court. [240] However, it is impossible to
contemplate the denial of such powers to the executive, [241] as without them it could
not carry out its management of the country’s foreign relations. Therefore, it is
preferable to view these powers, or many of them, as
Page 103
now flowing from ss 84(2)(h) and (i), [242] 85(2)(b), [243] 92(1) [244] and/or 231 [245] of
the Constitution when interpreted purposively. This approach has been endorsed by the
Constitutional Court. [246] And, as the Supreme Court of Appeal held in Minister of
Justice v SALC, albeit in the context of statutory interpretation, ‘[t]he ordinary principle
of interpretation is that the conferral of a power conveys with it all ancillary powers
necessary to achieve the purpose of that power’. [247]
Yet, outside of the limited situations governed by statute, [248] it is difficult to assert
that a certificate issued by the executive on its assessment of an act or fact of state
retains its conclusiveness under our new constitutional rules. As shown above, the
courts have made clear that all exercises of public power are now subject to
constitutional control, and s 232 of the 1996 Constitution provides that ‘customary
international law is law in the Republic unless it is inconsistent with the Constitution or
an Act of Parliament’. This
Page 104
means that an executive certificate issued in terms of any general executive powers in
relation to foreign affairs flowing from the Constitution (assuming such power exists)
that expresses an opinion on a question of international law—such as whether an entity
meets the customary international law requirements of statehood or whether a treaty is
in force between South Africa and another state—is now subject to judicial review. The
same holds true for executive certificates issued in terms of any statute. Executive
certificates issued in terms of a statute, whether the statute provides for them to be
conclusive proof (as in the case of the Foreign States Immunities Act) or merely prima
facie proof (as in the case of the Diplomatic Immunities and Privileges Act), are
nevertheless the exercise of public power that will be reviewable, at least on grounds of
irrationality or bad faith. [249] In other words, a certificate issued in terms of s 17 of the
Foreign States Immunities Act will, as provided for in that section, indeed be conclusive
while extant; however, it would always be open to a litigant to seek to review the
certificate for failure to comply with the principle of legality, and have it set aside.
 In practice, courts will no doubt extend a margin of appreciation to the executive in
matters such as the recognition of states and governments in which it is undesirable
that the state should ‘speak with two voices’. In other cases, such as the continued
existence of a treaty, different considerations apply. Sections 232 and 39(1) (where
human rights are in issue) require a court to examine whether the statement in an
executive certificate correctly reflects customary international law. The court
in Harksen was therefore correct in declining to accept the Minister’s certificate as
binding upon it and in making its own determination of the question of customary
international law before it.

12 The justiciability of acts of state

Before 1994, the application of international law in South Africa was subject to
constitutional rules and prerogative powers derived from English law. The position has
now changed dramatically. South Africa is a constitutional democracy in which the
principles of executive accountability and transparency feature prominently and the
courts are given wide powers of review of executive and administrative action and
legislation. Although the Constitution confers on the executive broad powers to enable
it to conduct South Africa’s foreign relations, these powers are no longer beyond the
reach of judicial review. [250]
Page 105
This issue is of particular importance in respect of ‘acts of state’—that is the acts of
foreign states within their own territories and the acts of the South African government
in the field of foreign affairs—which have hitherto been beyond judicial scrutiny.

(a) The justiciability of acts of the South African government in foreign

relations
The starting point for the doctrine of the non-justiciability of acts of the state to whose
structures the courts belong is Van Deventer v Hancke & Mossop [251] in which the
Transvaal Supreme Court was asked to rule that the British annexation of the South
African Republic on 1 September 1900 was premature on the ground that the British
forces had not established effective control over the territory—a prerequisite for
annexation under customary international law. To this argument, Innes CJ
responded: [252]
It is no doubt correct as a general rule of international law that two circumstances are
necessary to create a complete title by conquest: the conqueror must express in some clear
manner his intention of adding the territory in question to his dominions, and he must by
exercise of military force demonstrate his power to hold it as part of his own possessions. It is
also true that in March, 1901, large portions of the Transvaal, including the district of Vryheid,
were neither occupied nor dominated by British troops; but on the contrary were under the de
facto control of the Boer forces. And if this were a foreign court engaged in trying a cause in
regard to which the question of when the conquest of the Transvaal was complete became
relevant to the inquiry, it is possible that points of considerable intricacy and difficulty would
present themselves. But those considerations are not present here. This is a court constituted
by the British Crown, exercising powers and discharging functions derived from the Crown. In
its dealings with other states the Crown acts for the whole nation, and such dealings cannot be
questioned or set aside by its courts. They are acts of state into the validity or invalidity, the
wisdom or unwisdom, of which domestic courts of law have no jurisdiction to inquire.
Subsequent decisions adopted a similar approach. [253] The refusal of courts to ‘go
behind’ an executive certificate on a matter of foreign relations was a further
manifestation of the non-justiciability of acts of the South African government by
domestic courts. [254]
Page 106
However, over two decades of post-constitutional jurisprudence would appear to
confirm that there are now no longer any acts of the South African government in
relation to foreign affairs that remain non-justiciable today. First, courts are given the
power to apply rules of customary international law not inconsistent with the
Constitution or an Act of Parliament. A court therefore has the power to apply a rule of
customary international law contrary to an executive decision, which gives it the power
to enquire whether an executive act complies with international law. Secondly, s 34 of
the 1996 Constitution grants everyone ‘the right to have any dispute that can be
resolved by the application of law decided in a fair public hearing before a court’
without qualification. Thirdly, the Constitutional Court has held that ‘every exercise of
public power, including every executive act’ must comply with the principle of
legality, [255] and that the ‘exercise of all public power is subject to constitutional
control’. [256] The principle of legality requires decisions to be procedurally and
substantively rational and lawful (intra vires the executive’s powers). [257] This applies
equally to exercises of public power in relation to foreign affairs, which the
Constitutional Court has held are justiciable. [258] In these circumstances, it appears
that, in principle, all the executive’s decision-making in relation to foreign affairs is now
justiciable, and the only question is the standard of review, and the extent to which the
Court will give deference to the executive, in light of separation of powers
considerations. [259]
Page 107
One of the earlier post-constitutional High Court cases that dealt with the issue
is Kolbatschenko v King NO and Another. [260] Here, the state sought to prevent judicial
review of a decision on the part of the appropriate authorities to request legal
assistance from a court in Liechtenstein in a criminal investigation by arguing:
The respondents’ requests for foreign assistance, directed as they are to foreign governments,
constitute the conduct of foreign affairs by the Republic. Consequently, neither the decisions to
make the requests nor the requests themselves are justiciable in the sense of being
susceptible to rescission, review or declaratory proceedings in a South African court. [261]
In support of this argument, it was contended that ‘as a general rule, South African
courts should exercise judicial restraint when dealing with South Africa’s foreign
relations with other countries, in particular because of the lack of judicial or
manageable standards by which to judge the issue’. [262]
The Court held that there was no absolute rule of non-justiciability in respect of the
conduct of foreign relations. It stated:
[E]ven if one were to accept that the Executive retains certain discretionary non-statutory
powers to enable it to conduct foreign relations (such as, for example, the recognition of
foreign States and governments or the determination of the existence of treaties between
South African and foreign States), it would appear that such powers are no longer per
se beyond the scrutiny of the South African courts (see, in this regard, Dugard International
Law: A South African Perspective 2nd ed (2000) at 64–9). Whether or not a court will, in any
particular case, sit in judgment upon, and, if necessary, interfere with, a decision or action of
the executive depends, in our view, on the nature and subject-matter of the decision or action
concerned. [263]
On the one hand, the Court said it should be borne in mind that ‘the requirement of
accountable, responsive and transparent government is one of the founding values of
our constitutional democracy’ and that it is ‘only in highly exceptional cases that a court
will adopt a “hands-off” approach where a discretion has been exercised or an
Page 108
executive or administrative decision made which directly affects the rights and interests
of an individual applicant’. [264]
On the other hand, the Court acknowledged that:
South African courts have refused to evaluate decisions or actions in the realm of foreign
relations involving issues of a ‘high executive nature’. Thus, for example, matters such as the
recognition by the South African Government of a foreign State or of a foreign government, or
of the status of diplomatic representatives of a foreign State, have generally been regarded as
non-justiciable (see, for example, Inter-Science Research and Development Services (Pty) Ltd
v Republica Popular de Mocambique 1980 (2) SA 111 (T) at 117D–G). Such decisions usually
involve the relationship between the South African State and the foreign State concerned,
directly affecting the interests of such States as States, and are often of so ‘political’ a nature
that the courts have ‘no judicial or manageable standards’ by which to judge them (per Joffe J
in Swissborough Diamond Mines (Pty) and Others v Government of the Republic of South
Africa and Others (1999 (2) SA 279 (T) 334F–G), citing the judgment of Lord Wilberforce
in Buttes Gas and Oil Co v Hammer and Another (Nos 2 and 3); Occidental Petroleum Corp and
Another v Buttes Gas and Oil Co and Another (Nos 1 and 2) [1981] 3 All ER 616 (HL) at 633a–
f. This type of decision, which falls four-square within the political arena, would include matters
such as the making, or the determination of the existence, of treaties between South Africa
and foreign States, the declaration of war and the making of peace. In such cases, it is indeed
undesirable that the State should ‘speak with two voices’ and the latitude extended by the
Judiciary to the Executive in such matters will be correspondingly large. [265]
The Court held, however, that considerations of this kind did not apply in the present
case, as the decision to request foreign legal assistance did ‘not involve the evaluation
of social or economic policy among competing claims’. [266] Nor did it concern ‘matters
of a “high executive nature” directly impacting upon or affecting the relationship
between South Africa and the foreign State as States’. [267] The Court concluded by
stating that it could not be said that in assessing the regularity of a request for foreign
legal assistance in a criminal investigation ‘a court would be in a “judicial no-man’s
land” in that it would lack “judicial or manageable standards” by which to judge the
issues involved’. [268] Consequently, it held that the request for legal assistance from
Liechtenstein was justiciable.
The Constitutional Court adopted a similar approach in Mohamed v President of the
Republic of South Africa (Society for the Abolition of the Death Penalty in South Africa
Intervening) [269] in which it reviewed the
Page 109
decision of the government to deport a Tanzanian national, who was wanted by the
United States in connection with the 1998 bombing of the United States embassy in Dar
es Salaam, to the United States. The Court held that the government had acted
unconstitutionally by deporting Mohamed without securing an undertaking from the
United States that, if convicted, the death penalty could not be imposed on him, and
ordered the government to obtain an assurance that such a sentence would not be
imposed. [270] The Court stated:
To stigmatise such an order as a breach of the separation of State power as between the
Executive and the Judiciary is to negate a foundational value of the Republic of South Africa,
namely supremacy of the Constitution and the rule of law. The Bill of Rights, which we find to
have been infringed, is binding on all organs of State and it is our constitutional duty to ensure
that appropriate relief is afforded to those who have suffered infringement of their
constitutional rights. [271]
That the powers of the executive are not unlimited in foreign affairs was confirmed by
the Constitutional Court in Geuking v President of the Republic of South Africa when it
declared:
The President . . . must be free to take into account any matter considered relevant to what is
a policy decision relating to foreign affairs. It is not for the courts to determine what matters
are appropriate or relevant for that purpose. The Court could intervene only if the President
were to abuse the power vested in him or use it in a manner contrary to the provisions of the
Constitution. [272]
Kaunda and Others v President of the Republic of South Africa and Others (Society for
the Abolition of the Death Penalty in South Africa intervening as Amicus
Curiae) [273] raised the question whether a court might direct the government to protect
its nationals abroad who were threatened with serious human rights violations in
Zimbabwe or Equatorial Guinea. Although the Constitutional Court found that the
applicants had failed to establish that the government’s response to requests for
assistance was inconsistent with international law, the Court held that South African
nationals facing adverse state action in a foreign country are entitled to request the
government to protect them and the government is obliged to consider such requests
and to deal with them appropriately. The Court stated:
Decisions made by the government in these matters are subject to constitutional control.
Courts required to deal with such matters, will, however, give particular weight to the
Government’s special responsibility for and
Page 110
particular expertise in foreign affairs, and the wide discretion that it must have in determining
how best to deal with such matters. [274]
Although the Court recognised that the government should be given a wide margin of
appreciation in its decision to extend diplomatic protection to South African nationals,
the Court stated:
If the Government refuses to consider a legitimate request, or deals with it in bad faith or
irrationally, a court could require Government to deal with the matter properly. Rationality and
bad faith are illustrations of grounds on which a court may be persuaded to review a
decision. [275]
Tladi and Dlagnekova are correct in saying:
The essence of the judgment is, therefore, that while the executive has a broad discretion
when conducting foreign affairs, the courts can review such decisions on the grounds of, for
example, irrationality and bad faith. Thus, in contrast to the act of state doctrine, the exercise
of power by the executive in the conduct of foreign affairs is reviewable by the courts. The
courts can declare such exercise to be illegitimate or invalid. Although, the margin of discretion
afforded to the state in the exercise of such power is extremely wide, the exercise of the power
is nevertheless reviewable. [276]
Von Abo v Government of the Republic of South Africa and Others [277] must be seen in
the context of the strictures in Kaunda relating to review of executive action taken in
bad faith or irrationally. Here, the government failed to provide diplomatic protection to
Von Abo, a South African national whose farms were seized without compensation by
the Zimbabwe government. Numerous requests for assistance over a period of six years
were not taken seriously [278] by the Department of Foreign Affairs. To aggravate
matters, when Von Abo initiated legal proceedings against the government arising out
of its failure to grant diplomatic protection, the respondent government Ministers, and
particularly the Minister of Foreign Affairs, failed to
Page 111
offer any explanation to the Court by way of affidavit, leaving it to a senior law adviser
to attest to the opposing affidavit. This ‘shocking dereliction of duty’, [279] coupled with
the failure to afford protection to Von Abo, led Prinsloo J to conclude that the
government had acted irrationally and in bad faith [280] and to issue an order declaring
that the failure of the respondents ‘rationally, appropriately and in good faith’ to
consider Von Abo’s request for diplomatic protection was inconsistent with the
Constitution. He further held that Von Abo had a right to diplomatic protection and that
the government had a constitutional obligation to provide diplomatic protection.
Prinsloo J accordingly ordered the government to remedy the violation of rights of the
applicant by the government of Zimbabwe and to report within 60 days on the steps
that had been taken to do so. No appeal was lodged against this order. [281]
The government’s response to the above order was remarkable. First, neither the
Minister for Foreign Affairs nor any of the other respondent ministers produced an
affidavit on steps they had taken to protect Von Abo. Nor was there any explanation as
to why they had failed to do so. Instead, the Deputy Director of the Department of
Foreign Affairs produced an affidavit that provided information about meetings between
relatively junior officials from the Department of Foreign Affairs and the Zimbabwe
government at which no serious effort had been made to assert the need to provide
relief to Von Abo. (This contrasted with the more assertive and effective actions of
other states on behalf of their nationals.)
In these circumstances, Prinsloo J declared the conduct of the Minister of Foreign
Affairs to be contemptuous [282] and found that the respondents had failed to comply
with the main order.
It is clear that a court cannot instruct the government on how to make diplomatic
interventions or on what interventions to make. But it is not true to say that any
intervention will suffice. In response to the government’s argument that it had
discharged its constitutional obligations by making some representations even though
they were unsuccessful, Prinsloo J stated:
I disagree. On this argument . . . it would mean that a government which has the prerogative
to decide on the nature of the diplomatic interventions to be made, can opt for the most
ineffective and weak measures, which have no prospect of achieving the desired result, and
still insist that their feeble efforts
Page 112
pass constitutional muster because they have the prerogative to decide what measures to
adopt . . . . To argue that the measures [adopted by respondents] comply with the court order
because it is the prerogative of the government to decide what measures to adopt, is
untenable. It does not pass the test as expressed in Kaunda, Mohamed and Fose. The task
must be performed properly. The remedy afforded to an aggrieved individual whose
fundamental rights have been impaired . . . must be an effective one. [283]
An appeal to the Supreme Court of Appeal in Von Abo [284] succeeded on the ground
that Von Abo had failed to establish that he was entitled to diplomatic protection and
that Prinsloo J had erred in ordering the government to remedy the situation within
such a limited time frame and in ordering the government to pay damages to a national
in respect of a wrong committed by Zimbabwe in the territory of Zimbabwe. The Court
did, however, accept the finding in Kaunda that a court may review the decision of
government not to exercise diplomatic protection when it had acted irrationally or in
bad faith. [285] Here, the SCA held that it was unable to assess the reasonableness and
appropriateness of the government’s response to Von Abo’s request for diplomatic
protection because it had not been ‘entrusted with the content of the government policy
or specific steps taken by particular officials in terms of recognised procedures and
protocols’. [286] This led the SCA to state that ‘this case is an example of how a
government, founded on a constitutional dispensation and a culture of human rights, is
not supposed to treat its citizens and its courts’. [287] How, in these circumstances, the
Court could fail to make a finding that Von Abo was entitled to diplomatic protection on
the grounds of irrationality and bad faith is not clear.
In DA v Minister of International Relations, which is discussed in more detail above, a
full bench of the North Gauteng High Court found that the government’s decision to
deposit a notice with the United Nations Secretary-General withdrawing South Africa
from the Rome Statute was unconstitutional. In reviewing the government’s decision,
the Court pointed out that ‘[i]t is now axiomatic that the exercise of all public power,
including the conducting of international relations, must accord with the
Constitution. . .. While the notice of withdrawal was signed and delivered in the conduct
of international relations and treaty-making as an executive act, it still remained an
exercise in public power, which must comply with the
Page 113
principle of legality and is subject to constitutional control.’ [288] The Court ultimately
found that the decision should be set aside because it violated s 231(2) of the
Constitution and was irrational.
In Earthlife Africa the Western Cape High Court was faced with challenges to
decisions in relation to the government’s entering into of international agreements
dealing with nuclear procurement and co-operation (as well as challenges to certain
determinations in relation to new energy generation capacity under domestic
legislation). One of the challenges the Court was faced with was a challenge to the
constitutionality of the Minister of Energy’s signature of an inter-governmental
agreement (IGA) with Russia relating to nuclear procurement (and the President’s
authorisation thereof). As shown above, the Court was willing to hold that the
international agreement was unlawfully tabled, after being signed, under s 231(3) of
the Constitution (avoiding the need for parliamentary approval) when it should have
been tabled under s 231(2). However, the Court held that it was premature to
determine whether the President and the Minister violated the Constitution by signing
the agreement (which would have required the Court to effectively determine that the
terms of the agreement were inconsistent with the Constitution). [289] The basis for this
finding was that given the relief to be granted due to the unconstitutional tabling
(setting aside the tabling under s 231(3)) the international agreement would then still
need to be placed before Parliament under s 231(2) (for parliamentary approval) before
it became binding on South Africa. The Court therefore held that it was appropriate, in
light of the principle of the separation of powers, to refuse, at that stage, to determine
whether the President’s approval of the signature of the international agreement was
irrational and in violation of s 217 of the Constitution (which deals with the
requirements for lawful public procurement). [290] The Court explained its decision as
follows:
Should the executive then choose to table the Agreement before Parliament in terms of [s]
231(2), a parliamentary/political process will follow in which the Agreement will be debated in
both the NA and the NCOP with a view to its approval or disapproval by Parliament. It may
very well also be the subject of a process of public participation conducted through Parliament.
The outcome of this process cannot be foreseen nor should it be anticipated. In these
circumstances it would be invidious if the Court were, at this stage, to declare that certain of
its provisions are inconsistent with the Constitution and, more specifically, [s] 217 thereof. This
is not to suggest, however, that the Court will lack
Page 114
 jurisdiction to deal with such a question in future if the need should arise. [291]

In Law Society of South Africa and Others v President of the Republic of South Africa
and Others [292] the North Gauteng High Court was faced with a challenge arising from
the suspension of the SADC Tribunal. The applicants (the Law Society of South Africa
and a number of Zimbabwean nationals and companies) sought to have the President’s
participation in the SADC Summit’s decision to suspend the SADC Tribunal in
2011 [293] and the President’s signature of the 2014 Protocol for the Tribunal (which
would replace the extant 2000 Protocol for the Tribunal and limit the Tribunal’s
jurisdiction to hear only inter-state complaints) declared unconstitutional. Although a
number of grounds of review were raised, the President’s conduct was challenged
primarily on the basis that it was irrational, in bad faith, and violated South Africa’s
obligations under the SADC Treaty and the 2000 Tribunal Protocol.
The Court held that both the President’s participation in the Summit’s decision to
suspend the Tribunal (effectively failing to object to the consensus decision to suspend
the Tribunal) and his signature of the 2014 Protocol were unlawful, irrational and,
therefore, unconstitutional.
The Court’s finding that the President’s participation in the Summit’s suspension of
the Tribunal was unlawful, irrational and, therefore, unconstitutional was based on a
number of grounds (albeit, the judgment does not always clearly delineate between its
findings of unconstitutionality in relation to the President’s participation in the
suspension and those in relation to his signature of the 2014 Protocol). First, the Court
found that the President’s participation in the suspension was unlawful. The Court held
that this was so since it determined that the suspension violated the SADC Treaty
(which had been amended to incorporate the 2000 Tribunal Protocol) since ‘[t]he
Tribunal and its jurisdiction lie at the heart of the SADC Treaty and fulfil one of its main
purposes. Its emasculation by way of its de facto suspension was therefore similarly in
conflict with the Founding Treaty and South Africa’s constitutional
obligations’. [294] Flowing from the violation of the SADC Treaty, the Court found that
the participation in the suspension was also therefore in violation of ‘constitutional
obligations’. On this point, the judgment is not a picture of clarity. It appears that the
Court’s finding was based on one
Page 115
or more of the following constitutional violations: [295] (a) a violation of the rule of law
(the Court appeared to accept that an executive action or decision that violated South
Africa’s international obligations would be in violation of the rule of law enshrined in s
1(c)); (b) a violation of s 7(2) of the Constitution (the obligation on the state to protect
and promote the rights in the Bill of Rights)—although why this section was violated
was not explained; [296] or (c) a violation of s 231 since there had been no
parliamentary approval for the President’s conduct in not seeking to prevent the
Summit from suspending the Tribunal. [297] Secondly, the Court found that the decision
was procedurally and substantively irrational, since the President had not properly
explained the reasons for failing to object to the suspension and had not sought to
consult with the public in relation to a decision that effectively nullified individuals’
vested rights to approach the Tribunal under the existing SADC Treaty and 2000
Protocol (at least while the suspension remained in place).
In relation to the President’s signature of the 2014 Protocol, the Court’s finding that
the decision was irrational and unlawful appears to have been primarily predicated on
its acceptance, without clearly determining the issue, that the 2014 Protocol bound
South Africa merely on the President’s signature. This appears, inter alia, from the fact
that the Court held that ‘the Tribunal’s jurisdiction was simply signed away’ by the
President [298] and that the President’s signature ‘severely undermined the crucial SADC
institution, the Tribunal’. [299] The difficulty with this finding is that the 2014 Protocol is
expressly made subject to ratification. Member states that wish to bind themselves to
the Protocol have to ratify it ‘in accordance with their constitutional
procedures’. [300] The Protocol only comes into force when two-thirds of the member
states have deposited their instruments of ratification. [301] While the President signed
the 2014 Protocol, the government had not yet sought to ratify the Protocol.
Importantly, the 2014 Protocol expressly provided that it is only when it
Page 116
enters into force (after the depositing of the requisite number of instruments of
ratification) that it will change the Tribunal’s jurisdiction by repealing the 2000 Tribunal
Protocol. This is so since article 48 of the 2014 Protocol states that ‘[t]he 2000 Protocol
on the Tribunal in the Southern African Development Community is repealed with effect
from the date of entry into force of this Protocol’. Thus, to the extent that the Court’s
decision was implicitly predicated on accepting that the President bound South Africa to
the 2014 Protocol by his mere signature, the Court does not explain how this could be
reconciled with the express provisions of the Protocol.
The Court’s declarations of unconstitutionality in respect of the President’s
participation in the suspension of the Tribunal and his signature of the Protocol were, in
terms of s 172 of the Constitution, referred by the High Court to the Constitutional
Court for confirmation. [302] This was required since the conduct declared
unconstitutional amounted to ‘conduct of the President’, which declarations must, in
terms of s 172(2)(a), be confirmed by the Constitutional Court. At the time of writing,
the Constitutional Court had not yet determined the matter.

(b) The justiciability of acts of foreign governments

Anglo-American courts, whose decisions have guided South African courts on matters of
international law, have produced a substantial body of decisions on the justiciability of
acts of foreign states. In South Africa there was no reported decision until 1999.
In the United States the act of state doctrine prohibits a US court from declaring
invalid the official act of a foreign sovereign performed within its own
territory. [303] Although the doctrine is premised on a policy of judicial restraint, the
Supreme Court has held that it ‘does not establish an exception for cases and
controversies that may embarrass foreign governments’. [304]
The position in England is less clear. Although the term ‘act of state’ is used, courts
prefer to see the refusal to pronounce on acts of
Page 117
foreign governments as the exercise of a policy of judicial restraint. In Buttes Gas and
Oil Co v Hammer (No 3) Lord Wilberforce declared:
There exists in English law a more general principle that the courts will not adjudicate upon the
transactions of foreign states. Though I would prefer to avoid argument on terminology, it
seems desirable to consider this principle, if existing, not as a variety of ‘act of state’ but one
for judicial restraint or abstention. [305]
The existence of a broad principle of judicial restraint in matters affecting the
transactions of foreign states has been criticised by scholars [306] and courts in
England. [307] Recent decisions, however, suggest that English courts will be prepared to
pronounce on the compatibility of foreign law with international law, [308] particularly in
the field of human rights. [309]
In the Belhaj case the United Kingdom Supreme Court recently had the opportunity
to undertake a detailed review of the law in relation to the act of state doctrine and the
inter-linked issue of state immunity. [310] In Belhaj the claimants sought to bring claims
in tort against the United Kingdom government and certain officials for alleged
complicity in their rendition and mistreatment at the hands of foreign states. The
Supreme Court found that in English law there are three types of foreign acts of
state. [311] It held, first, that a court would normally treat a foreign state’s legislation as
valid in so far as it affects movable or immovable property within the foreign state’s
jurisdiction, and, secondly, a court would not normally question the validity of a foreign
governmental act in respect of property within
Page 118
the foreign state’s jurisdiction. It was the scope of the third type of foreign act of state
that was principally in issue. The third type is the rule of non-justiciability or judicial
abstention whereby a domestic court will not adjudicate upon sovereign acts committed
by a foreign state abroad, which was developed in Buttes—and which Lord Sumption, in
his separate judgment, styled as an ‘international act of state’. [312] Such a situation
may arise, inter alia, where a court cannot properly hear a claim due to a lack of
judicial or manageable standards, or as a function of the separation of powers, where a
court considers that it should not hear a claim since it is outside the proper bounds of
its constitutional functions.
Neither the United States act of state doctrine nor the English doctrine of judicial
restraint are rules of international law. [313] They are principles of domestic law that limit
justiciability in foreign relations in accordance with their own constitutional rules.
Consequently, they are not binding on South African courts in terms of s 232 of the
Constitution.
A South African court may decide to follow the judicial policies of the United States
or the United Kingdom in respect of non-justiciability but it can only do so within the
framework of its own constitutional rules in general, and ss 34 and 232 in particular. A
court cannot fashion a principle of judicial restraint or non-justiciability for South Africa
that takes no account of this framework, particularly when it differs so fundamentally
from that of the United States or the United Kingdom.
It is against this background that the decision of the Transvaal Provincial Division
in Swissborough Diamond Mines (Pty) Ltd v Government of the Republic of South
Africa [314] must be viewed. Here, the applicant, a company registered in Lesotho but
controlled by South African shareholders, sought to obtain discovery of documents
relating to an alleged conspiracy between the South African and Lesotho governments
to dispossess the applicant of its rights to diamond leases in Lesotho. With no mention
of South Africa’s constitutional rules, Joffe J found that the ‘true agreement’ between
South Africa and Lesotho was non-justiciable. [315] Although he found it unnecessary to
make any finding on the conduct of the government of Lesotho, he found that the
relationship between South Africa and Lesotho ‘belonged’ to international law and was
‘not an area for the
Page 119
judicial branch of government’. It was therefore ‘a matter in respect of which this court
should exercise judicial restraint’. [316] In reaching this conclusion, he declared:
 The basis of the application of the act of state doctrine or that of judicial restraint is just as
applicable to South Africa as it is to the USA or England. The comity of nations is just as
applicable to South Africa as it is to other sovereign states. The judicial branch of government
ought to be astute in not venturing into areas where it would be in a judicial no-man’s land. It
would appear that in an appropriate case, as an exercise of the court’s inherent jurisdiction to
regulate its own procedure, the court could determine to exercise judicial restraint and to
refuse to entertain a matter, notwithstanding it having jurisdiction to do so, in view of the
involvement of foreign states therein. [317]
Referring to Swissborough as authority, the Supreme Court of Appeal held that ‘[c]ourts
should act with restraint when dealing with allegations of unlawful conduct ascribed to
sovereign States’. [318]
More recently, the question of whether a matter should be dismissed on the basis of
the act of state doctrine, and the linked basis of state immunity, arose in the case
of Saharawi Arab Democratic Republic and Another v Owners and Charterers of the MV
NM Cherry Blossom and Others. [319] In this case the applicants—the Saharawi Arab
Democratic Republic (SADR) (a state recognised by South Africa but regarded by the
United Nations as the non-self-governing territory of Western Sahara) and the Polisario
Front, a national liberation movement—applied ex parte (without notice to the other
parties) for an urgent interim interdict to restrain a cargo of phosphate jhat had been
mined by a Moroccan mining company, pursuant to Moroccan law, in Western Sahara.
(Morocco claims that Western Sahara, which it refers to as the Southern Provinces,
forms part of Morocco’s territory.) The phosphate mined in Western Sahara was on
board the NM Cherry Blossom, a ship which had docked in Port Elizabeth to refuel on
route to New Zealand. The applicants sought to prevent the phosphate from being
removed from the jurisdiction of the Court, pending the determination of an action in
which the applicants claimed, inter alia, delivery of the cargo (based on alleged
ownership of the phosphate, as a natural resource extracted from Western Sahara,
which belonged to the people of Western Sahara). The High Court granted the ex
parte interim order preventing the cargo from being removed from the Court’s
jurisdiction pending a return date,
Page 120
when the Court, after service on the parties, would determine whether to grant an
interdict pending the determination of a vindicatory action to be instituted for recovery
of the phosphate. In the subsequent opposed hearing, the Moroccan mining company
and its subsidiary (collectively ‘OCP’) opposed the application and sought the dismissal
of the interim interdict.
OCP raised two principal legal defences to the applicants’ claims. First, that in terms
of the act of state doctrine, the issues that must be determined by the Court for the
application to succeed were non-justiciable before any South African court. Secondly,
that the application indirectly impleaded Morocco (since it required the Court to
effectively determine that Morocco’s de facto administration and law making for
Western Sahara, in terms of which OCP conducted its mining, was unlawful and invalid)
and thus the Court was barred from hearing the case as a result of Morocco’s state
immunity.
In considering the act of state defence, the Court found that it was premature to
consider whether the doctrine would preclude the applicants’ claim to the phosphate.
This was so, the Court reasoned, because courts that are dealing with interlocutory
proceedings where the issues are of particular complexity should only decide such
issues as are strictly necessary to be determined at an interlocutory stage. The Court
found that it was only once the issues had been fully and precisely pleaded at the trial
stage that it would be appropriate to consider whether the claim made it necessary for
the Court to determine the lawfulness of a foreign act of state. [320] However, while not
applying the doctrine in the case before it, the Court confirmed a number of features of
the doctrine of act of state (or ‘foreign act of state’) as a matter of South African law.
First, it pointed out that ‘[u]nlike, state immunity, which is a rule of public international
law, the doctrine of a foreign act of state is a municipal law rule which derives from
common law principles as developed in Anglo-American courts. It is founded upon the
principle of mutual respect for equality of sovereign states, the principle of
comity.’ [321] Secondly, with reference to Swissborough and the Supreme Court of
Appeal’s decision in Van Zyl v Government of the Republic of South Africa, it accepted
that the act of state doctrine was applicable in South Africa. [322] Thirdly, it pointed out
that the guiding principles of the doctrine are that ‘[a] court which is called upon to
exercise restraint or to refrain from adjudicating a matter in respect of which it
Page 121
otherwise has jurisdiction will do so with caution and then only in circumstances where
it is necessary to determine the particular issue engaged by a foreign act of
state’. [323] Fourthly, the Court emphasised that the doctrine’s application and scope are
subject to constitutional discipline. It emphasised that a court ‘bound as it is to apply
the Constitution as supreme law and to give effect to the spirit, purport and objects of
the Constitution, will be mindful of the fundamental rights contained therein,
particularly the right of access to the courts enshrined in s 34, in determining the
circumstances in which and the ambit of the exercise of its discretion to decline
adjudication in circumstances where an act of a foreign sovereign is engaged’. [324]
The Court, in dealing with the issue of state immunity, [325] held that the
determination of the applicants’ claim would have ‘an [e]ffect upon the interests of
Morocco but such [e]ffect falls within the realm of political or moral interests and
cannot have legal affect. It follows therefore that the claim to state immunity cannot be
upheld’. Thus, importantly, the Court accepted that, even though Morocco itself was not
directly being sued, if the claim would have had an effect on Morocco’s legal rights and
interests, then state immunity (under the Foreign States Immunities Act 87 of 1981
and customary international law) would preclude the claim. The Court relied on [326] the
Supreme Court of Appeal’s decision in Minister of Justice v SALC, where it accepted that
state immunity ‘is available when it is sought to implead a foreign state, whether
directly or indirectly, before domestic Courts, and also when action is taken against
state officials acting in their capacity as such. They enjoy the same immunity as the
state they represent’. [327] The Court accepted that the concept of indirect impleading,
as articulated in Minister of Justice v SALC, was recognised (and precluded by) s 2(2) of
the Foreign States Immunities Act. [328] However, the Court noted that since there was
no direct South African judicial authority on the principles of indirect impleading
Page 122
(presumably other than the brief statement in Minister of Justice v SALC) it was
necessary to have regard to ‘customary international law and, in particular, the manner
in which the principle is interpreted and applied’.
The Court, in analysing the customary international law position, relied heavily on
the interpretation and application of the principle by the UK Supreme Court
in Belhaj (discussed above). When one considers the Court’s application of the
principles of indirect impleading to the facts of the case, it is not entirely clear why the
Court suggested that Morocco’s interest in the matter was merely ‘political’, when it had
held earlier that ‘[t]he exercise of administrative control [by Morocco] and the
application of Moroccan law to the portion of the territory of Western Sahara under
Moroccan control . . . is at the heart of the dispute as to the exercise of the right to
self-determination [which includes the claim to the Cargo] by the Saharawi
people’. [329] The Court partly explained that the reason why it found that the legal
rights of Morocco were not affected was that Morocco was not a party to the
proceedings and had no proprietary right to the cargo. [330] But neither of these are
necessary requirements for indirect impleading (in fact, quite the opposite: in indirect
impleading, the foreign state is not a party). [331] And, at a practical level, it is quite
easy to see what the effect on Morocco’s legal rights would be. The judgment would
mean that Morocco could no longer guarantee that OCP would have lawful and
enforceable title over the minerals that OCP mined in Western Sahara—thus, effectively
rendering Morocco’s granting of title and its regulation thereof invalid (at least within
South Africa). Conversely, the SADR would henceforth be entitled, as a matter of South
African law, to have attached within South Africa’s jurisdiction any minerals mined in
Western Sahara, notwithstanding the minerals having been mined by entities given
legal title to do so by Morocco. This would appear, as a matter of law, at least within
South Africa, to render Morocco’s laws and the passing of title invalid.
Having found that state immunity did not preclude the Court from determining the
matter, and having found that the question of whether the act of state doctrine
precluded the claim should be decided at the trial stage, the Court interdicted the
removal of the phosphate from the jurisdiction of the Court pending a vindicatory
Page 123
action for delivery of the cargo, which the applicants were given leave to institute.

13 The extraterritorial application of the Constitution

In Mohamed v President of the Republic of South Africa (Society for the Abolition of the
Death Penalty in South Africa Intervening) [332] the Constitutional Court held that the
deportation of Tanzanian national Mohamed to the United States—with the collusion of
US officials, for him to stand trial in that country—violated South Africa’s Constitution
because the South African government failed to obtain a prior undertaking that, if he
were convicted, the death penalty would not be imposed on him. This infringed his
constitutional rights to human dignity, life and not to be punished in a cruel, inhuman
or degrading manner. [333] The court ordered its judgment to be delivered to the US
federal court trying Mohamed in New York. Mohamed was not sentenced to death.
In a thoughtful analysis of the case, Max du Plessis points out that this case did not
really constitute an extraterritorial application of the Constitution as the harm to
Mohamed in the United States was caused by the action of public officials in South
Africa: ‘The “extra-territorial” application of the Constitution is thus an application of
the Bill of Rights, triggered by effects abroad, which would be the end-result of acts of
public officials begun in South Africa.’ [334]
This interpretation of Mohamed was confirmed by the Constitutional Court in Kaunda
and Others v President of the Republic of South Africa and Others. [335] In this case, the
Court refused to order the South African government to extend the protection offered
by the South African Constitution to South African nationals whose rights were
threatened in a foreign country (Zimbabwe or Equatorial Guinea). The court stated that
‘it is a general rule of international law that the laws of a state ordinarily apply only
within its own territory’. [336] It acknowledged that:
 There may be special circumstances where the laws of a State are applicable to nationals
beyond the State’s borders, but only if the application of the law does not interfere with the
sovereignty of other States. For South Africa to
Page 124
assume an obligation that entitles its nationals to demand, and obliges it to take action to
ensure, that laws and conduct of a foreign State and its officials meet not only the
requirements of the foreign State’s own laws, but also the rights that our nationals have under
our Constitution would be inconsistent with the principle of State sovereignty. Section 7(2)
should not be construed as imposing a positive obligation on government to do this. [337]
The Court, however, wisely left open the question whether the extraterritorial
infringement of a constitutional right by an organ of the South African state in
circumstances that do not infringe the sovereignty of a foreign state would be
justiciable. [338]
More recently, the question of the extraterritorial effect of the Constitution was
indirectly at issue in the case of Law Society of South Africa v President of the Republic
of South Africa. [339] The applicants challenged the constitutionality of two distinct
actions by the President. First, they challenged the President’s conduct in relation to the
SADC Summit’s suspension of the SADC Tribunal in 2011. Secondly, they challenged
the President’s signature of the new SADC Tribunal Protocol in 2014. The Court held
that both the President’s actions (his conduct in relation to the suspension and his
signature) were irrational and unlawful and therefore unconstitutional. Of relevance to
the issue of extraterritoriality is the fact that the Court was prepared to declare that the
President’s failure—through his representative, the High Commissioner to Namibia, who
attended the Summit’s meeting on his behalf—to object to the SADC Summit’s
consensus decision to suspend the Tribunal (taken by the Summit at its May 2011
meeting in Namibia) was unconstitutional. The Court made this declaration of
unconstitutionality on a number of grounds, including that the President’s conduct was
found to be irrational and in violation of constitutional and international law obligations.
Notwithstanding that the President, through his representative, was acting as a
member of an international organisation (the Summit, SADC’s highest policy-making
body) not based in South Africa, and was participating in the consensus decision-
making process governed by international law and not domestic law (the SADC Treaty
governed the decision making in question), the Court was still willing to find that the
President’s conduct was reviewable in terms of the Constitution. This decision would
therefore appear to be authority for the proposition that actions of South Africa’s
diplomats and other representatives in international organisations, such as the United
Nations, remain subject to constitutional control and must
Page 125
comply with the Constitution and, in particular, the principle of legality. Thus, these
officials’ decisions must be rational as a matter of South Africa law and must comply
with any relevant constitutional and international law obligations. If their decisions are
irrational or unlawful (either violating the Constitution or international law) they would
be regarded as violating the principle of legality. The mere fact that a decision or
conduct occurred abroad and as part of an international organisation’s decision-making
process would not preclude a South African court from reviewing the decision.
******
The 1996 Constitution includes provisions dealing with succession to treaties,
international human rights law, self-determination, humanitarian law and the use of
force. These provisions are considered in the appropriate chapters.
[1] For an examination of this debate, see JG Starke ‘Monism and dualism in the theory of international
law’ (1936) 17 BYIL 66; C Roodt ‘National law and treaties’ (1987–1988) 13 SAYIL 72.
[2] ‘International law and municipal law’ in E Lauterpacht (ed) International Law: Being the Collected
Papers of Hersch Lauterpacht vol I The General Works (1970) 216 at 217.
[3] For a discussion of the adoption and transformation theories, see F Morgenstern ‘Judicial practice
and the supremacy of international law’ (1950) 27 BYIL 42.
[4] Lauterpacht (n 2) 216.
[5] DP O’Connell International Law 2 ed (1970) vol 1 at 44–5; H Booysen Volkereg en sy Verhouding
tot die Suid Afrikaanse Reg (1989) 68–9.
[6] See S Fatima Using International Law in Domestic Courts (2005); Y Shany Regulating
Jurisprudential Relations between National and International Courts (2007); A Nollkaemper Domestic
Courts and the Rule of International Law (2009); D Sloss (ed) The Role of Domestic Courts in Treaty
Enforcement: A Comparative Study (2009); J Nijman and A Nollkaemper (eds) New Perspectives on the
Divide between National and International Law (2007); International Law Association Final Report of the
Study Group on Principles on the Engagement of Domestic Courts with International Law (2016).
[7] De Jure Belli ac Pacis, Prolegomena paras 16–18; Inleiding tot de Hollandsche
Rechtsgeleertheyd 1.2.10–13.
[8] Cornelis van Bynkershoek Quaestionum Juris Publici Libri Duo II, translation in J Brown Scott
(ed) The Classics of International Law (1930) xl–xli.
[9] Writers use the term ‘law of nations’ loosely to include both international law and the Roman law ius
gentium: see Huber Heedendaegse Rechtsgeleertheyt 1.2.21, 2; Voet Commentarius ad
Pandectas 1.1.18, 19.
[10] See, for example, Van der Linden Rechtsgeleerd Practicaal en Koopmans Handboek 4.2 (on
maritime law). See, too, AJGM Sanders ‘The applicability of customary international law in municipal
law—South Africa’s monist tradition’ (1977) 40 THRHR 147 at 148.
[11] See JW Wessels History of Roman-Dutch Law (1908) 285, 291–3. At 293, he declares that ‘natural
law . . . was the cornerstone of the whole fabric’ of Roman-Dutch law.
[12] This view is challenged by Hercules Booysen: ‘Is gewoonteregtelike volkereg deel van ons reg?’
(1975) 38 THRHR 315 at 316; Volkereg (n 5) 77.
[13] One of the earliest reported cases in which principles of international law were invoked
was Ncumata v Matwa (1881–1882) 2 EDC 272 at 279.
[14] (1894) 1 SAR 214.
[15] At 223.
[16] At 232.
[17] For a discussion of some of the problems that did not come to court, see T Baty International Law
in South Africa (1900). He deals with issues such as contraband for neutral ports (Delagoa Bay), the
sovereignty of the South African Republic, passage of troops over neutral territory, and the conduct of
hostilities. This fair-minded treatise is, surprisingly, singularly lacking in jingoism.
[18] Mshwakezele v Guduza (1901) 18 SC 167 at 171; Van Deventer v Hancke & Mossop 1903 TS 401
at 419, 424; Lemkuhl v Kock 1903 TS 451 at 454; Olivier v Wessels 1904 TS 235 at 241; R v
Louw (1904) 21 SC 36 at 40–1, 46–7.
[19] Alexander v Pfau 1902 TS 155 at 159–61, 163–4, 166.
[20] Du Toit v Kruger (1905) 22 SC 234 at 237, 239; Achterberg v Glinister 1903 TS 326 at 330.
[21] Van Deventer v Hancke & Mossop (n 18) 409–10.
[22] See, for example, Lendalease Finance Co (Pty) Ltd v Corporation de Mercadeo Agricola 1975 (4)
SA 397 (C). See further J Dugard ‘The purist legal method, international law and sovereign immunity’ in
JJ Gauntlett (ed) JC Noster: ’n Feesbundel (1979) 36 at 45–9.
[23] For the position in English law, see Lauterpacht (n 2); O’Connell (n 5). The strongest assertion of
this position appears in a dictum of Lord Denning MR in Trendtex Trading Corporation v Central Bank of
Nigeria (1977) QB 529 (CA) 553–4: ‘A fundamental question arises for decision: what is the place of
international law in our English law? One school of thought holds to the doctrine of incorporation. It says
that the rules of international law are incorporated into English law automatically and considered to be
part of English law unless they are in conflict with an Act of Parliament. The other school of thought holds
to the doctrine of transformation. It says that the rules of international law are not to be considered as
part of English law except insofar as they have been already adopted and made part of our law by the
decisions of judges, or by Act of Parliament, or long established custom. The difference is vital when you
are faced with a change in the rules of international law. Under the doctrine of incorporation, when the
rules of international law change, our English law changes with them. But, under the doctrine of
transformation, the English law does not change. It is bound by precedent. . . . As between these schools
of thought, I now believe that the doctrine of incorporation is correct. Otherwise I do not see that our
courts could ever recognise a change in the rules of international law.’
See, too, R v Jones [2006] UKHL 16, 132 ILR 668; R O’Keefe ‘The doctrine of incorporation revisited’ (2008)
79 BYIL 7.
[24] South Atlantic Islands Development Corporation v Buchan 1971 (1) SA 234 (C) 238B–F.
[25] The Paquete Habana 175 US 677 (1900) 700; Banco Nacional de Cuba v Sabbatino, Receiver, et
al 376 US 398 (1964) 423; ED Dickinson ‘The law of nations as part of the national law of the United
States’ (1952) 101 Univ of Pennsylvania LR 26, 792; L Henkin Foreign Affairs and the Constitution (1972)
222–3. JF Murphy The United States and the Rule of Law in International Affairs (2004).
[26] See RP Schaffer (Balkin) ‘The inter-relationship between public international law and the law of
South Africa: An overview’ (1983) 32 ICLQ 277; JW Bridge ‘The relationship between international law
and the law of South Africa’ (1971) 20 ICLQ 746; Booysen (n 12) 315 et seq and ch 3 respectively; DJ
Devine ‘Qualifications on the incorporation of international customary law into South African municipal
law’ (1973) 1 Natal Univ LR 58; J Dugard ‘International law is part of our law’ (1971) 88 SALJ 13;
Sanders (n 10); AJGM Sanders ‘The applicability of customary international law in South African Law—The
Appeal Court has spoken’ (1978) 11 CILSA 198; G Erasmus ‘The Namibian Constitution and the
application of international law’ (1989–1990) 15 SAYIL 81 at 85–92; N Botha ‘The coming of age of public
international law in South Africa’ (1992–1993) 18 SAYIL 36.
[27] In terms of s 6(2)(e) of the Republic of South Africa Constitution Act 110 of 1983.
[28] 1971 (1) SA 234 (C) 238B–F. Confirmed in Southern Africa Litigation Centre v Minister of Justice
and Constitutional Development 2015 (5) SA 1 (GP) para 5. For an early dictum that international law
need not be proved, see CC Maynard v The Field Cornet of Pretoria (n 14) 232.
[29] For example, Ex parte Belli 1914 CPD 742 at 745–6; Marburger v The Minister of Finance 1918
CPD 183 at 187; (1) De Howorth v The SS ‘India’; (2) Mann, George & Co (Delagoa) Ltd v The SS
‘India’ 1921 CPD 451 at 457–8; Crooks and Company v Agricultural Co-operative Union Ltd 1922 AD
423; R v Lionda 1944 AD 348 at 352; Ex parte Sulman 1942 CPD 407; S v Penrose 1966 (1) SA 5 (N) 10.
[30] Supra (n 28) 283C–D. See, also, Inter-Science Research and Development Services (Pty) Ltd v
Republica Popular de Mocambique 1980 (2) SA 111 (T) 124H; Kaffraria Property Co (Pty) Ltd v
Government of the Republic of Zambia 1980 (2) SA 709 (E) 712E–G, 715A; Yorigami Maritime
Construction Co Ltd v Nissho-Iwai Co Ltd 1977 (4) SA 682 (C) 696E; Ex parte Schumann 1940 NPD 251
at 254. Cf Parkin v Government of the République Démocratique du Congo 1971 (1) SA 259 (W), which
shows support for the adoption theory (261A).
[31] Supra (n 28) 238C–E.
[32] 1978 (1) SA 893 (A).
[33] At 906B. The suggestion by Booysen (n 5) 69–70 that the dictum in Nduli gives support to the
dualist adoption theory is discussed and dismissed in the first edition of this work at 43–4.
[34] Act 3 of 1968, s 1.
[35] House of Assembly Debates vol 22, col 337 (12 February 1968).
[36] Alexander v Pfau (n 19) 159, 164; Inter-Science Research and Development Services (Pty) Ltd v
Republica Popular de Mocambique (n 30) 124H; Kaffraria Property Co (Pty) Ltd v Government of the
Republic of Zambia (n 30) 712F, 715A; Binga v Administrator-General, South West Africa 1984 (3) SA
949 (SWA) 967F.
[37] GE Devenish Interpretation of Statutes (1992) 212; CC Maynard v The Field Cornet of Pretoria (n
14) 222–3, 232; Achterberg v Glinister (n 20) 334; Claassens v Wilkens 1905 ORC 139 141; R v
Lionda (n 29) 352; S v Penrose (n 29) 11E–F; Hajaree v Ismail 1905 TS 451 at 456; Ex parte Adair
Properties (Pvt) Ltd 1967 (2) SA 622 (R) 627B–F.
[38] See cases referred to in n 36.
[39] Liebowitz v Schwartz 1974 (2) SA 661 (T) 662A.
[40] Roodt (n 1).
[41] See s 6(2)(e) of the Republic of South Africa Constitution Act 110 of 1983.
[42] Minister of the Interior v Bechler; Beier v Minister of the Interior 1948 (3) SA 409 (A) 447. See,
too, Ex parte Savage 1914 CPD 827 at 830; Policansky v Minister of Agriculture 1946 CPD 860 at 865.
[43] 1965 (3) SA 150 (A) 161C–D. This dictum was confirmed by the Appellate Division in S v
Tuhadeleni 1969 (1) SA 153 (A) 173–5. See, also, Maluleke v Minister of Internal Affairs 1981 (1) SA 707
(B) 712; Binga v Administrator-General, South West Africa (n 36) 968B–C; Tshwete v Minister of Home
Affairs 1988 (4) SA 586 (A) 606; S v Muchindu 1995 (2) SA 36 (W) 38H–I; AZAPO v President of the
Republic of South Africa 1996 (4) SA 671 (CC) 688 (para 26).
[44] Binga v Administrator-General, South West Africa (n 36) 968E.
[45] There is a wealth of literature on the place of international law in the new constitutional order. The
following selection takes account of the writings on the Interim Constitution (1993), suggestions for the
‘final’ Constitution and the ‘final’ Constitution (1996) itself. N Botha ‘International law and the South
African Interim Constitution’ (1994) 9 SA Public Law 245; M Olivier ‘The status of international law in
South African municipal law: Section 231 of the 1993 Constitution’ (1993–4) 19 SAYIL 1; T Maluwa
‘International human rights norms and the South African Interim Constitution’ (1993–4) 19 SAYIL 14; N
Botha ‘Interpreting a treaty endorsed under the 1993 Constitution’ (1993–4) 19 SAYIL 148; DJ Devine
‘The relationship between international law and municipal law in the light of the Interim South African
Constitution 1993’ (1995) 44 ICLQ 1; DJ Devine ‘Some problems relating to treaties in the Interim South
African Constitution and some suggestions for the definitive Constitution’ (1995) 20 SAYIL 1; N Botha
‘Incorporation of treaties under the Interim Constitution: A pattern emerges?’ (1995) 20 SAYIL 196; J
Dugard ‘International law and the final Constitution’ (1995) 11 SAJHR 241; J Dugard ‘The influence of
international human rights law on the South African Constitution’ (1996) 49 Current Legal Problems 305;
R Keightley ‘Public international law and the final Constitution’ (1996) 12 SAJHR 405; E de Wet ‘The place
of public international law in the new South African constitutional order’ (1998) 1 Recht in Afrika 207; HA
Strydom ‘The international law openness of the South African Constitution’ in G Carpenter and N Botha
(eds) Suprema Lex: Essays on the Constitution Presented to Marinus Wiechers (1997); J Dugard
‘International law and the South African Constitution’ (1997) 8 European Journal of International Law 77;
RC Blake ‘The world’s law in one country: The South African Constitutional Court’s use of public
international law’ (1998) 115 SALJ 668; M Olivier ‘Informal international agreements under the 1996
Constitution’ (1997) 22 SAYIL 63; N Botha ‘Treaties after the 1996 Constitution: More questions than
answers’ (1997) 22 SAYIL 95; M du Plessis ‘The extra-territorial application of the South African
Constitution’ (2003) 120 SALJ 797; E de Wet ‘The friendly but cautious reception of international law in
the jurisprudence of the South African Constitutional Court: Some critical remarks’ (2005) 28 Fordham
International Law Review 101; E de Wet ‘The status of international law in the South African legal order’
in D Shelton (ed) International Law and Domestic Legal Systems: Incorporation, Transformation and
Persuasion (2011); E Cameron ‘Constitutionalism, rights, and international law: The Glenister decision’
(2013) 23 Duke Journal of Comparative & International Law 389; E de Wet, H Hestermeyer and R
Wolfrum (eds) The Implementation of International Law in Germany and South Africa (2015).
[46] Ex parte Chairperson of the Constitutional Assembly: In re Certification of the Constitution of the
Republic of South Africa 1996 (4) SA 744 (CC).
[47] Ex parte Chairperson of the Constitutional Assembly: In re Certification of the Amended Text of
the Constitution of the Republic of South Africa, 1996 1997 (2) SA 97 (CC).
[48] While initially the 1996 Constitution bore an Act number (Act 108 of 1996), in terms of s 1(1) of
the Citation of Constitutional Laws Act 5 of 2005, ‘no Act number is to be associated with the
Constitution’ from 27 June 2005 (the date of commencement of the Act) and in terms of s 2(2) ‘[a]ny
reference to the “Constitution of the Republic of South Africa, 1996 (Act No 108 of 1996 )” contained in
any law in force immediately prior to the commencement of that Act, must be construed as a reference to
the “Constitution of the Republic of South Africa, 1996”’.
[49] See the comment by Ngcobo CJ in Glenister v President of the RSA 2011 (3) 347 (CC) para 97.
[50] See the remarks of Sachs J on s 232 in S v Basson 2005 (1) SA 171 (CC) 216.
[51] Supra (n 30) 715. This matter was deliberately left open by Margo J in Inter-Science Research and
Development Services (Pty) Ltd v Republica Popular de Mocambique (n 30) 125G–H.
[52] Trendtex Trading Corporation v Central Bank of Nigeria [1977] QB 529 (CA).
[53] At 554.
[54] See Dugard (n 22). See also Minister of Justice and Constitutional Development v Southern Africa
Litigation Centre 2016 (3) SA 317 (SCA) (‘Minister of Justice v SALC’), a case relating to the
government’s failure to arrest President al-Bashir of Sudan, where Wallis JA, in considering whether there
was an exception to the customary international law rule of immunity for heads of state who are alleged
to have committed international crimes, held (para 70) that ‘[i]n the absence of a binding treaty or other
international instrument creating such an exception, or an established universal practice in the affairs of
nations, one looks to the decisions of international courts for guidance as to the existence of such an
exception’.
[55] See Chapter 3.
[56] Du Toit v Kruger (n 20). Here, the Chief Justice stated (at 238): ‘The modern authorities, to which
this court has been referred, on the rights of capture during war do not afford much assistance for the
decision of the appeal. The rules which are laid down by some writers for exempting the private property
of an enemy from capture have not been so universally accepted and acted upon as to justify this court in
treating them as binding principles of law’ (emphasis added).
[57] Supra (n 32) 906D (emphasis added). For a full examination of this dictum, see D Devine ‘What
international customary law is part of South African law?’ (1987–1988) 13 SAYIL 119. Support for
the dictum in Nduli is to be found in the judgment of Patel J in Van Zyl v Government of the RSA 2005
(11) BCLR 1106 (T) para 91.
[58] Supra (n 30) 125A–B.
[59] 1988 (3) SA 51 (C) 56–7 (emphasis added).
[60] At 57H–I. Another case that affords a good illustration of the manner in which customary
international law is to be ascertained is Nkondo v Minister of Police 1980 (2) SA 895 (O).
[61] R Keightley (n 45) at 408
[62] S v Makwanyane 1995 (3) SA 391 (CC) 413.
[63] N Botha ‘International law and the South African Interim Constitution’ (1994) 9 SA Public Law 245
at 255.
[64] Fisheries Jurisdiction Case 1974 ICJ Reports 3 at 23–6. The test of universal acceptance was
rejected by Judge Lachs in his separate opinion in the North Sea Continental Shelf Cases 1969 ICJ
Reports 3 at 229.
[65] Supra (n 54).
[66] Minister of Justice v SALC (n 54) para 70.
[67] Minister of Justice v SALC (n 54) para 74 (emphasis added) with reliance on Jones v Ministry of
the Interior (Saudi Arabia) [2006] UKHL 26 para 63.
[68] S v Petane (n 59) 57.
[69] National Commissioner of Police v Southern African Human Rights Litigation Centre 2015 (1) SA
315 (CC) (‘National Commissioner v SALC’).
[70] Paragraphs 35 and 77. As the Court held, torture is also criminalised in South Africa under two
domestication acts: the Prevention and Combating of Torture of Persons Act 13 of 2013 and the
Implementation of the Rome Statute of the International Criminal Court Act 27 of 2002. Thus, as the
Court held at para 39: ‘In effect, torture is criminalised in South Africa under s 232 of the Constitution
and the Torture Act whilst torture on the scale of crimes against humanity is criminalised under s 232 of
the Constitution, the Torture Act and the ICC Act.’ This case is further discussed in Chapter 9.
[71] See J Dugard ‘South Africa’ in D Sloss (ed) The Role of Domestic Courts in Treaty Enforcement: A
Comparative Study (2009) 448.
[72] See articles 32(3)(e), 63(2)(e) and 144. The text of this Constitution appears in (1989–1990)
15 SAYIL 301.
[73] Section 82(1)(i) of the Interim Constitution empowered ‘the President’ to carry out this task, but
in practice it was delegated to Ministers of state, particularly the Minister of Foreign Affairs.
[74] Section 231(2) of the Interim Constitution.
[75] Section 231(3) of the Interim Constitution.
[76] The post-apartheid Parliament relies heavily on committees. Thus, a treaty may have to be
approved by several parliamentary committees before it is presented for ratification. See J Dugard and I
Currie ‘Public international law’ 1995 Annual Survey 76 at 76–9.
[77] N Botha ‘Incorporation of treaties under the Interim Constitution: A pattern emerges’ (1995)
20 SAYIL 196.
[78] Keightley (n 45) 412.
[79] Section 231(2) of the Interim Constitution provided that ‘Parliament shall . . . be competent to
agree to the ratification of or accession to an international agreement negotiated and signed’ by the
executive.
[80] This interpretation was spelt out in a letter from the Minister of Foreign Affairs to other Ministers,
titled ‘Procedures for the Conclusion of International Agreements’ of 13 June 1994.
[81] Earthlife Africa v Minister of Energy 2017 (5) SA 227 (WCC) (‘Earthlife Africa’) paras 122–8.
[82] N Botha ‘Treaty making in South Africa: A reassessment’ (2000) 25 SAYIL 69 at 75–8; J
Schneeberger ‘A labyrinth of tautology: The meaning of the term “international agreement” and its
significance for South African law and treaty-making practice’ (2001) 26 SAYIL 1 at 5–7; W Scholtz ‘A few
thoughts on s 231 of the South African Constitution’ (2004) 29 SAYIL 202.
[83] Earthlife Africa (n 81).
[84] Earthlife Africa (n 81) para 114, with reference to the fourth edition of this work (at 417).
[85] Schneeberger (n 82) 5, quoting from the Manual on Executive Acts of the President May 1999
(emphasis added). This Manual was revised in 2007. It affirms that the executive views ‘technical,
administrative agreements’ as agreements which are ‘not of major political or other significance’, ‘have no
financial consequences and do not affect domestic law’. In short, they deal with ‘minor every-day issues’
(para 5.5).
[86] Olivier ‘Informal international agreements under the 1996 Constitution’ (n 45) 64; Botha (n 82)
77–8.
[87] Earthlife Africa (n 81) paras 109–10.
[88] See article 14 of the Vienna Convention on the Law of Treaties, which emphasises the intention of
parties in deciding whether a treaty requires ratification. This principle was approved by the South African
Appellate Division in S v Eliasov 1967 (4) SA 583 (A). Article 14(1) provides, as one of the means for
determining whether a treaty becomes binding on ratification, that ‘[t]he consent of a State to be bound
by a treaty is expressed by ratification when: (a) the treaty provides for such consent to be expressed by
means of ratification’.
[89] Glenister (n 49) para 89.
[90] See Earthlife Africa (n 81) para 114 and Democratic Alliance v Minister of International Relations
and Cooperation 2017 (3) SA 212 (GP) (‘DA v Minister of International Relations’) paras 61–3. The
Constitution obligates both houses of Parliament to facilitate public participation in their legislative and
other activities (ss 59(1) and 72(1)).
[91] See DA v Minister of International Relations (n 90) and Earthlife Africa (n 81).
[92] Earthlife Africa (n 81) para 114.
[93] DA v Minister of International Relations (n 90) para 55.
[94] See Botha (n 82) 76; and compare Schneeberger (n 82) 5.
[95] DA v Minister of International Relations (n 90).
[96] Ibid para 47.
[97] MN Shaw International Law 8 ed (2017) 691; A Aust Modern Treaty Law and Practice 3 ed (2013)
144.
[98] Paragraph 132.
[99] Paragraph 47.
[100] Paragraphs 47 and 51.
[101] Paragraph 51.
[102] Paragraph 53, read with para 44, and paras 65–6.
[103] Paragraph 57.
[104] Paragraph 59, quoting from C Hoexter Administrative Law in South Africa 2 ed (2012) 547.
[105] Paragraphs 70 and 71.
[106] In addition to the applicant (the Democratic Alliance, a political party) certain public interest
organisations (which had been cited as respondents, and one that had intervened) also challenged the
constitutionality of the notice of withdrawal.
[107] Paragraphs 72–7.
[108] Paragraph 84, orders 1 and 2.
[109] Paragraph 84, order 3.
[110] Supra (n 43).
[111] This principle was reaffirmed by the Constitutional Court in AZAPO v President of the Republic of
South Africa 1996 (4) SA 671 (CC) para 26 and National Commissioner v SALC (n 69) para 24; and by
the Supreme Court of Appeal in Progress Office Machines v SARS 2008 (2) SA 13 (SCA) para 6 (which the
Constitutional Court referred to with approval in International Trade Administration Commission v
SCAW 2012 (4) SA 618 (CC) para 25).
[112] Section 239 of the Constitution.
 [113] For example, s 133 of the Civil Aviation Act 13 of 2009 gives municipal effect to the Tokyo
Convention on Offences and Certain Acts Committed on Board Aircraft (1963), the Hague Convention for
the Suppression of Unlawful Seizure of Aircraft (1970), and the Montreal Convention for the Suppression
of Unlawful Acts Against the Safety of Civil Aviation (1971).
[114] For example, the Diplomatic Immunities and Privileges Act 37 of 2001 incorporates the Vienna
Convention on Diplomatic Relations of 1961 and the Vienna Convention on Consular Relations of 1963
into South African legislation by means of schedules.
[115] For example, s 2(3)(a) and (3)ter of the Extradition Act 67 of 1962, and s 108 of the Income Tax
Act 58 of 1962 (in this regard see the discussions in Commissioner, South African Revenue Service v Van
Kets 2012 (3) SA 399 (WCC) paras 11–15 and Commissioner, South African Revenue Service v
Tradehold 2013 (4) SA 184 (SCA) paras 15 and 16).
[116] Morgenstern (n 3) 51.
[117] 1969 (1) SA 153 (A) 173–5. In Binga v Cabinet for South West Africa 1988 (3) SA 155 (A) the
Appellate Division was asked to reconsider its finding in Tuhadeleni that the mandate had not been
incorporated into municipal law. Van Heerden JA found it unnecessary to decide on this matter but
‘assumed’, for the purpose of the appeal, that the mandate had been incorporated (182–3).
[118] Supra (n 49) para 92. Ngcobo CJ made this statement in a minority judgment. See, too, the
comments of Moseneke DCJ and Cameron J in the majority judgment in this case at paras 179–81.
[119] Section 231(2) quite appropriately and helpfully refers to ‘approval’ by Parliament, so as to
distinguish it from ‘ratification’, which is the action on the international plane by the executive once
parliamentary approval has been secured. See DA v Minister of International Relations (n 90) para 47.
[120] Article 16 of the Vienna Convention on the Law of Treaties provides that:
‘Unless the treaty otherwise provides, instruments of ratification, acceptance, approval or accession
establish the consent of a State to be bound by a treaty upon:
(a)
their exchange between the contracting States;
(b)
their deposit with the depositary; or
(c)
their notification to the contracting States or to the depositary, if so
agreed.’
[121] DA v Minister of International Relations (n 90) para 51.
[122] As the Court also pointed out in DA v Minister of International Relations (n 90) para 55, ‘[t]he
binding power comes only once parliament has approved the agreement on behalf of the people of South
Africa as their elected representative. It appears that it is a deliberate constitutional scheme that the
executive must ordinarily go to parliament (the representative of the people) to get authority to do that
which the executive does not already have authority to do.’
[123] See DA v Minister of International Relations (n 90) paras 51 and 55.
[124] Ibid para 51.
[125] Foster v Neilson 27 US (2 Pet) 253 (1829); Sei Fujii v California 242 P 2d 617 (1952), 19 ILR
312. See also Restatement (Third) of the Foreign Relations Law of the United States para 111 (1987); T
Buergenthal ‘Self-executing and non-self-executing treaties’ (1992 IV) Hague Recueil 343.
[126] ‘Remarks at the annual meeting of the American Society of International Law’ (1951)
45 Proceedings of the American Society of Law 102.
[127] 128 S Ct 1346 (2008), 170 L Ed 2d 190 (2008).
[128] Concerning Avena and Other Mexican Nationals (Mexico v US) 2004 ICJ Reports 12.
[129] N Botha ‘Treaty-making in South Africa: A reassessment’ (2000) 25 SAYIL 69 at 91; Botha in DB
Hollis, MR Blakeslee and LB Ederington (eds) National Treaty Law and Practice 2 ed (2005) 58.
[130] JD van Vyver ‘Universal jurisdiction in international criminal law’ (1999) 24 SAYIL 107 at 130.
[131] HA Strydom ‘The international law “openness” of the South African Constitution’ in G Carpenter
(ed) Suprema Lex: Essays on the Constitution Presented to Marinus Wiechers (1998) 93.
[132] E Ngolele ‘The content of the doctrine of self-execution and its limited effect in South African law’
(2006) 31 SAYIL 153.
[133] M Olivier ‘Exploring the doctrine of self-execution as enforcement mechanism of international
obligations’ (2002) 27 SAYIL 99.
[134] Claassen v Minister of Justice and Constitutional Development 2010 (6) SA 399 (WCC) expressly
rejected the suggestion that the ICCPR is a self-executing instrument. Cf Zealand v Minister of Justice and
Constitutional Development 2008 (4) SA 458 (CC) para 52.
[135] Olivier (n 133) 116.
[136] Unreported, TPD Case No 959/04, 18 April 2008 (‘Quagliani TPD’). Noted by N Botha in
‘Extradition, self-execution and the South African Constitution: A non-event? South African judicial
decisions’ (2008) 33 SAYIL 253 and ‘Public international law’ 2008 Annual Survey 30. See, too, G Ferreira
and W Scholtz ‘Has the Constitutional Court found the lost ball in the high weeds? The interpretation of
section 231 of the South African Constitution’ (2009) 42 CILSA 269.
[137] Quagliani TPD (n 136) 12.
[138] Ibid 13.
[139] Quagliani TPD (n 136).
[140] Ibid 18.
[141] Unreported, TPD Case No 21142/08, 23 June 2008 (‘Goodwin’). Noted by Botha (n 136).
[142] Goodwin (n 141) 13.
[143] Ibid 15.
[144] President of the Republic of South Africa v Quagliani; President of the Republic of South Africa v
Van Rooyen; Goodwin v Director-General, Department of Justice and Constitutional Development 2009
(4) BCLR 345 (CC).
[145] Ibid 359A–B.
[146] President of RSA v Quagliani (n 144) 363A–E.
[147] Neville Botha describes the judgment of the Constitutional Court in this matter as ‘a profoundly
unsatisfactory judgment’: ‘Rewriting the Constitution: The “strange alchemy” of Justice Sachs, indeed!’
(2009) 34 SAYIL 253.
[148] J Dugard International Law: A South African Perspective 3 ed (2005) 213–14.
[149] 2010 (6) SA 399 (WCC) para 36. Critically noted by Magnus Killander in ‘Judicial immunity,
compensation for unlawful detention and the elusive self-executing treaty provision’ (2010)
26 SAJHR 386.
[150] President of the Republic of South Africa v Quagliani (n 144) 355C.
[151] Ibid 356B–C. See, too, Quagliani TPD (n 136) 10, noted in Botha (n 136) 257.
[152] DA v Minister of International Relations (n 90) para 55.
[153] RR Baxter ‘International law in her “infinite variety”’ (1980) 29 ICLQ 549; A Aust ‘The theory and
practice of informal international instruments’ (1986) 35 ICLQ 787; Olivier ‘Informal international
agreements under the 1996 Constitution’ (n 45). Sed contra J Klabbers The Concept of Treaty in
International Law (1996) 63–4, 122–35, 216–17, 243, 245–50. For an overview of the literature, see
Schneeberger (n 82).
[154] S v Harksen; Harksen v President of the Republic of South Africa; Harksen v Wagner NO 2000
(1) SA 1185 (C); Harksen v President of the Republic of South Africa 2000 (2) SA 825 (CC).
[155] 67 of 1962.
[156] S v Harksen (n 154) paras 52–4 and 59.
[157] Harksen v President of the Republic of South Africa (n 154) para 21.
[158] Schneeberger (n 82) 32.
[159] Olivier ‘Informal international agreements under the 1996 Constitution’ (n 45) 74; Botha (n 82)
71; Schneeberger (n 82) 32–40.
[160] South West African Cases, Preliminary Objections, 1962 ICJ Reports 328 at 331.
[161] See S v Harksen (n 154) para 52, in which Van Zyl J stated: ‘It is this very intention and consent
that distinguishes treaties from informal or ad hoc agreements or arrangements’. The Cape Provincial
Division held that an arrangement to extradite a person under s 3(2) of the Extradition Act 67 of 1962 did
not qualify as an international agreement because of the absence of an intent to create reciprocal rights
and duties (para 59). See, too, the judgment of the Constitutional Court in Harksen v President of the
Republic of South Africa (n 154) para 21.
[162] See Botha (n 82) 72.
[163] Binga v Administrator-General, South West Africa (n 36) 968E; Welkom Municipality v Masureik
and Herman t/a Lotus Corporation 1997 (3) SA 363 (A) 371; Masureik and Another (t/a Lotus
Corporation) v Welkom Municipality 1995 (4) SA 745 (O). See further N Botha ‘Municipal application of
Annex 14 to the Chicago Convention: The role of recommended international practices and procedures in
South African municipal law’ (1997) 22 BYIL 112.
[164] 172 of 1993.
[165] See Botha (n 82) at 89–90; De Wet in Shelton (ed) International Law and Domestic Legal
Systems (n 45); H Strydom and T Huarka ‘South Africa’ in V Gowlland-Debbas (ed) National
Implementation of United Nations Sanctions: A Comparative Study (2004) 430–2.
[166] See above (n 37) and S v Basson (n 50) para 100.
[167] Section 37(4) provides that any legislation enacted in consequence of a declaration of a state
emergency may derogate from the Bill of Rights only to the extent that, inter alia, the legislation ‘is
consistent with the Republic’s obligations under international law applicable to states of emergency’.
Section 35(3)(l) recognises the right ‘not to be convicted of an act or omission that was not an offence
under either national or international law at the time when it was committed or omitted’.
[168] Section 35(1) of the Interim Constitution required ‘a court of law’ to ‘have regard to public
international law’.
[169] J Dugard ‘The role of international law in interpreting the Bill of Rights’ (1994) 10 SAJHR 208.
[170] S v Makwanyane 1995 (3) SA 391 (CC) 413–14. See the comment on this case by N Botha and M
Olivier ‘Ten years of international law in the South African courts: Reviewing the past and assessing the
future’ (2004) 29 SAYIL 42 at 44–8, 75. The Constitutional Court has confirmed that the principles
enunciated in Makwanyane in relation to s 35 of the Interim Constitution continue to apply to s 39 of the
1996 Constitution (see, eg, Government of the Republic of South Africa v Grootboom 2001 (1) SA 46
(CC) para 26; Glenister (n 49) para 178 fn 28). See, too, Prince v President of the Law Society, Cape of
Good Hope 1998 (8) BCLR 976 (C) 985C–D; Prince v President, Cape Law Society 2002 (2) SA 794 (CC)
824A–E, 837E–F, 851B–C, 858–9. Here, the Court considered non-human rights treaties under s
39(1)(b). In practice, human rights treaties will be most frequently invoked but the language of s
39(1)(b) does not exclude other treaties.
[171] For instance, the Constitutional Court, in SATAWU v Garvas 2013 (1) SA 83 (CC), had regard to a
report by the Special Rapporteur on extrajudicial, summary or arbitrary executions to assist it in
interpreting the right to freedom of assembly (para 30). In Rahim v Minister of Home Affairs 2015 (4) SA
433 (SCA) the Supreme Court of Appeal had regard to the report of the Special Rapporteur of the Human
Rights Council of the United Nations on the rights of migrants, a resolution by the Inter-American
Commission on Human Rights, and the International Convention on the Protection of the Rights of All
Migrant Workers and Members of Their Families (which South Africa was not a party to) in order to
establish the considerations that a South African official would have to take into account when deciding
whether to arrest and detain an illegal foreigner under the Immigration Act 13 of 2002 (see paras 18 and
19). And, in Glenister (n 49) the Constitutional Court relied on a report by the Organisation for Economic
Co-operation and Development (OECD) in relation to anti-corruption fighting institutions, even though it
was not binding. The Court explained that ‘[t]he OECD report is not itself binding in international law, but
can be used to interpret and give content to the obligations in the conventions that we’ve described’
(para 187). See also Helen Suzman Foundation v President of the Republic of South Africa 2015 (2) SA 1
(CC) paras 41, 160 and 167.
[172] This is borne out by the judgment in S v Petane 1988 (3) SA 51 (C), in which Conradie J
considered the question whether the principal provisions of a treaty to which South Africa was not a
party—Additional Protocol I of 1977—had been translated into customary law by usage.
[173] Glenister (n 49) paras 179–202. See further Maluke v Minister of Internal Affairs 1981 (1) SA
707 (B) 713; Mabuda v Minister of Co-operation and Development 1984 (2) SA 49 (Ck) 54–5 (discussed
in J Dugard ‘Public international law’ 1984 Annual Survey 68 at 82–3). For a discussion of this principle,
see J Dugard ‘International human-rights norms in domestic courts: Can South Africa learn from Britain
and the United States?’ in Ellison Kahn (ed) Fiat Iustitia: Essays in Memory of Oliver Deneys
Schreiner (1983) 221 at 234–6. Cf Binga v Cabinet for South West Africa 1988 (3) SA 155 (A) 185F
where Van Heerden JA restricted the scope of this argument in an obiter dictum in which he found that
the presumption in favour of compliance with an international treaty obligation applies only where the
statute seeks to give effect to the treaty in question. In limiting this presumption, Van Heerden JA
preferred the more restrictive approach of Diplock LJ in Salomon v Commissioner of Customs and
Excise [1966] 3 All ER 871 (CA) 875–6 to that of Lord Denning MR in R v Secretary of State for Home
Affairs, Ex parte Bhajan Singh [1975] 2 All ER 1083 (CA). In support of Lord Denning’s position, see PJ
Duffy ‘English law and the European Court of Human Rights’ (1980) 29 ICLQ 585 at 589.
[174] S v Okah 2018 (1) SACR 492 (CC) para 38; see also Bridon International GmbH v International
Trade Administration Commission 2013 (3) SA 197 (SCA) paras 12, 13 and 26 and International Trade
Administration Commission v SCAW (n 111) para 43, read with para 25. In SCAW the Court held that
‘[i]n Progress Office Machines the Supreme Court of Appeal correctly concluded that the Anti-Dumping
Agreement is binding on the Republic in international law, even though it has not been specifically
enacted into municipal law’, and the Court then went on to hold that ‘we are required by the Constitution
to interpret domestic legislation governing the duration of anti-dumping duties consistently with these
international obligations’.
[175] Act 33 of 2004.
[176] See also Bridon International (n 174) paras 12, 13 and 26.
[177] S v Okah (n 174) para 26, quoting the oral submissions of counsel for one of the amici.
[178] Molvan v Attorney-General for Palestine [1948] AC 351 (PC) 365; Morgenstern (n 3) 70.
[179] 2008 (2) SA 13 (SCA) para 11. Cf Association of Meat Importers v ITAC [2013] 4 All SA 253
(SCA) paras 56–61 where Nugent JA, writing for the majority, raises questions as to the nature and
extent of and the basis for what was decided in Progress Office Machines in relation to the indirect
application of the WTO Agreement, and, in essence, seeks to qualify the decision (see also the minority
judgment by Wallis JA at paras 109–18).
[180] In Winter v Minister of Defence 1960 AD 194 at 198 and S v Tuhadeleni (n 43) 176–7 the
Appellate Division left open the question whether proclamations may be tested against the terms of the
Mandate for South West Africa, a treaty that was not incorporated into municipal law. In Binga v
Administrator-General, South West Africa (n 36), especially at 973D–G, the Court wrongly assumed that
the Tuhadeleni case held that delegated legislation might not be tested against the mandate. See J
Dugard ‘The revocation of the mandate for South West Africa revisited’ (1985) 1 SAJHR 154 at 159; G
Erasmus ‘Mandates, military service and multiple choice’ (1985–1986) 11 SAYIL 115 at 133.
[181] Section 172 of the Constitution.
[182] 1996 (4) SA 671 (CC). For criticisms of this decision, see comments by J Dugard ‘Is the truth and
reconciliation process compatible with international law? An unanswered question’ (1997) 13 SAJHR 258;
C Braude and D Spitz ‘Memory and the spectre of international justice: A Comment on Azapo—AZAPO v
President of the Republic of South Africa 1996 focus—The Constitutional Court 1996 term—Cases and
comments’ (1997) 13 SAJHR 269; D Moellendorf ‘Amnesty, truth and justice: Azapo focus—The
Constitutional Court 1996 term—Cases and comments’ (1997) 13 SAJHR 283; Z Motala ‘The
Constitutional Court’s approach to international law and its method of interpretation in the
“Amnesty decision”: Intellectual honesty or political expediency?’ (1996) 21 SAYIL 29; and De Wet in
Shelton (ed) International Law and Domestic Legal Systems (n 45).
[183] Supra (n 49).
[184] Act 34 of 1995.
[185] AZAPO (n 182) para 25.
[186] AZAPO (n 182) para 26.
[187] Ibid para 28.
[188] Supra (n 49).
[189] South African Police Service Amendment Act 57 of 2008.
[190] National Prosecuting Authority Act 32 of 1998, as amended by Act 56 of 2008.
[191] Glenister (n 49) para 166.
[192] (2004) 43 ILM 37.
[193] (2004) 43 ILM 5.
[194] Available at http://www.sadc.int.
[195] Glenister (n 49) para 182.
[196] Paragraph 189.
[197] Glenister (n 49) paras 192–5.
[198] Paragraph 202.
[199] 2001 (4) SA 938 (CC).
[200] Ibid paras 45–8.
[201] Carmichele (n 199) para 62.
[202] Ibid para 73.
[203] ‘The role of international law in the development of South African common law’ (2001)
26 SAYIL 253 at 259.
[204] Minister of Safety and Security v Carmichele 2004 (3) SA 305 (SCA) 319–20.
[205] 2013 (5) SA 325 (CC).
[206] Treaty of the Southern African Development Community (which had been amended to
incorporate the Tribunal Protocol).
[207] Protocol on Tribunal in the Southern African Development Community.
[208] For a critical discussion of the case, see H Woolaver ‘Judicial enforcement of international
decisions against foreign states in South Africa: The case of Government of the Republic of Zimbabwe v
Louis Karel Fick and Others’ (2015) 6 Constitutional Court Review 217.
[209] Government of Zimbabwe v Fick (n 205) para 69.
[210] 4 of 2000.
[211] Section 3(2)(a).
[212] 27 of 2002.
[213] Section 2.
[214] 2015 (1) SA 315 (CC) para 24.
[215] S v Okah (n 174).
[216] Supra (n 49) paras 192–4.
[217] Supra (n 179) para 11.
[218] 2018 (6) BCLR 695 (GP).
[219] Supra (n 205). Neither the SADC Treaty nor the Tribunal Protocol has been domesticated and the
Constitutional Court, while not directly considering the issue, did not suggest that either international
agreement was self-executing.
[220] See further on this topic, AJGM Sanders ‘Our state cannot speak with two voices’ (1971)
88 SALJ 413; AJGM Sanders ‘The courts and recognition of foreign states and governments’ (1975)
92 SALJ 165; Booysen (n 5) 83–6.
[221] R Jennings and A Watts (eds) Oppenheim’s International Law 9 ed (1992) vol 1 at 1046–52.
[222] See Chapter 7.
[223] Minister of Home Affairs v Bickle 1984 (2) SA 439 (ZS) 450H.
[224] The executive publishes a list of persons entitled to diplomatic immunity in the Government
Gazette in terms of s 7(2) of the Diplomatic Immunities and Privileges Act 37 of 2001. See S v Penrose (n
29); J Dugard ‘Consular immunity’ (1966) 83 SALJ 126.
[225] Section 17 of the Foreign States Immunities Act 87 of 1981.
[226] [1939] AC 256 (HL) 264.
[227] Sanders (n 220) ‘Our state cannot speak with two voices’ 413–14.
[228] 1971 (1) SA 359 (N).
[229] 1971 (3) SA 899 (A).
[230] S v Devoy (n 228) 363D; S v Devoy (n 229) 906–7.
[231] S v Devoy (n 228) 362H.
[232] 1998 (2) SA 1011 (C).
[233] Ibid 1019–20.
[234] Harksen (n 232) 1020C. See the criticism of this statement by N Botha in ‘International law in
South African courts’ (1999) 24 SAYIL 330.
[235] Diplomatic Immunities and Privileges Act 37 of 2001, s 9(3).
[236] Foreign States Immunities Act 87 of 1981, s 17. Unlike s 9(3) of the Diplomatic Immunities and
Privileges Act, which merely provides that the certificate is ‘prima facie’ proof of any fact in relation to
‘any question [which] arises as to whether or not any person enjoys any immunity or privilege under this
Act or the [Vienna] Conventions’, s 17 of Foreign States Immunities Act provides that the certificate by
the Minister of Foreign Affairs (now the Minister of International Relations and Co-operation) is ‘conclusive
proof’, inter alia, of ‘whether any foreign country is a state for the purposes of this Act’ and ‘whether any
territory is a constituent part of a federal foreign state for the said purposes’.
[237] Section 82(1) of Act 200 of 1993.
 [238] Section 84 of the 1996 Constitution.
[239] President of the Republic of South Africa v Hugo 1997 (4) SA 1 (CC) 6–9; G Carpenter
‘Prerogative powers in South Africa—Dead and gone at last?’ (1997) 22 SAYIL 104.
[240] Mohamed v President of the Republic of South Africa 2001 (3) SA 893 (CC) paras 31 and
32; President of the Republic of South Africa v South African Rugby Football Union 2000 (1) SA 1 (CC)
para 144; and Mansingh v General Council of the Bar 2014 (2) SA 26 (CC) paras 4–6. Sections 84(1) and
85(2)(h) of the Constitution appear to make clear that the executive has no powers outside those
provided for in the Constitution or by legislation. See Mansingh para 25.
[241] Carpenter (n 239) 108; H Booysen ‘Has the act of state doctrine survived the 1993 Interim
Constitution?’ (1995) 20 SAYIL 189 at 191.
[242] Section 84(2)(h) and (i) provide that ‘[t]he President is responsible for—
...
(h)
receiving and recognising foreign diplomatic and consular representatives;
(i)
appointing ambassadors, plenipotentiaries, and diplomatic and consular representatives’.
[243] Section 85(2) provides that ‘[t]he President exercises the executive authority, together with the
other members of the Cabinet, by—
...
(b)
developing and implementing national policy’ (emphasis added).
Such national (as opposed to provincial or local government) policy would evidently include the
country’s international relations policy.
[244] Section 92(1) provides that ‘[t]he Deputy President and Ministers are responsible for the powers
and functions of the executive assigned to them by the President’. The President evidently can, and does,
assign foreign affairs powers and functions to Ministers (see Von Abo v President of the Republic of South
Africa 2009 (5) SA 345 (CC) paras 43 and 45).
[245] Section 231(1) provides that ‘[t]he negotiating and signing of all international agreements is the
responsibility of the national executive’.
[246] See Van Abo v The President (n 244) para 45 where the Constitutional Court held that the source
of the executive’s power to provide diplomatic protection was ‘s 85(2) read together with s 92(1) of the
Constitution’. In Kaunda and Others v President of the Republic of South Africa and Others 2005 (4) SA
235 (CC), O’Regan J pointed out (para 243) that ‘[i]t is clear, though perhaps not explicit, that under our
Constitution the conduct of foreign affairs is primarily the responsibility of the Executive. That this is so, is
signified by a variety of constitutional provisions, including those that state that the President is
responsible for receiving and recognising foreign diplomatic and consular representatives, appointing
ambassadors, plenipotentiaries and diplomatic and consular representatives, and that the national
executive is responsible for negotiating and signing international agreements. The conduct of foreign
relations is therefore typically an executive power under our Constitution.’
[247] Minister of Justice v SALC (n 54) para 95.
[248] The Diplomatic Immunities and Privileges Act 37 of 2001, s 9(3) and the Foreign States
Immunities Act 87 of 1981, s 17.
[249] See Kaunda (n 246) para 80; DA v Minister of International Relations (n 90) para 44.
[250] Ibid.
[251] Supra (n 18).
[252] At 409–10. See, to the same effect, Postmaster-General v Taute; Treasurer-General v Van
Vuuren; Postmaster-General v Parsons; Master of Supreme Court v Roth 1905 TS 582 at 586.
[253] Ex parte Belli 1914 CPD 742 at 747; Verein fur Schutzgebietsanleihen EV v Conradie NO 1937 AD
113 at 146–7; Haak v Minister of External Affairs 1942 AD 318 at 326; Vereeniging Municipality v
Vereeniging Estates Ltd 1919 TPD 159 at 163.
[254] See above, notes 228–31.
[255] Minister of Defence and Military Veterans v Motau 2014 (5) SA 69 (CC) para 69; see
also Democratic Alliance v President of the Republic of South Africa 2013 (1) SA 248 (CC) para
27; Pharmaceutical Manufacturers Association of SA: In re Ex Parte President of the Republic of South
Africa 2000 (2) SA 674 (CC) paras 51 and 85.
[256] National Treasury v Opposition to Urban Tolling Alliance 2012 (6) SA 223 (CC) para 64.
[257] See DA v Minister of International Relations (n 90) paras 54 and 64; Pharmaceutical
Manufacturers Association of SA (n 255) paras 17–20 and 85; Democratic Alliance v President of the
Republic of South Africa (n 255) para 34; Mansingh v General Council of the Bar (n 240 para 25.
[258] Kaunda (n 246) paras 78–80. See also the full bench High Court decision of DA v Minister of
International Relations (n 90) para 44.
[259] Some of the early academic debate and uncertainty appears to have been settled by cases such
as Kaunda (n 246), DA v Minister of International Relations (n 90) and Law Society v the President (n
218), as well as others discussed below, which all demonstrate that the courts now view all foreign affairs
decisions to be justiciable. For this early academic debate, see Booysen (n 241) who states that acts of
state ‘have not survived the Interim Constitution unscathed’ and that ‘an act of state has become
justiciable in terms of international law’ (196). Carpenter (n 239), however, argues acts of state are ‘not
subject to judicial scrutiny, in accordance with the doctrine of separation of powers’ (111). It is difficult to
understand how the doctrine of separation of powers can override ss 232, 34 and, possibly, s 39(1) (if
human rights are involved) or the principle of legality recognised by s 1(c) of the Constitution. See further
K Lehmann ‘The act of state doctrine in South African law: Poised for reintroduction in a different guise’
(2000) 15 SA Public Law 337; and K Lehmann ‘The foreign act of state doctrine: Its implications for the
rule of law in South Africa’ (2001) 16 SA Public Law 68.
[260] 2001 (4) SA 336 (C). See the comments on this case by N Botha ‘The post-Constitution “act of
state”: The need for further theoretical refinement’ (2002) 27 SAYIL 295; GN Barrie ‘Is the absolute
discretionary prerogative relating to the conduct of foreign relations alive and well and living in South
Africa?’ 2001 (3) TSAR 409.
[261] Kolbatschenko (n 260) 352F–G.
[262] Ibid 354C–D.
[263] Ibid 355F–H.
[264] Kolbatschenko (n 260) 355C–D.
[265] Ibid 356–7.
[266] Ibid 357C–D (emphasis in the original).
[267] Ibid 357F–G (emphasis in the original).
[268] Ibid 357H–I.
[269] 2001 (3) SA 893 (CC) para 31.
[270] See also Minister of Home Affairs v Tsebe 2012 (5) SA 467 (CC).
[271] Mohamed para 71.
[272] 2003 (3) SA 34 (CC) para 27.
[273] Supra (n 246). See, too, the judgment of the Transvaal Provincial Division, Kaunda v President of
the Republic of South Africa 2004 (5) SA 191 (T).
[274] Kaunda (n 246) para 144(6) (emphasis added). See, too, Van Zyl v Government of the RSA (n
57) paras 55–7.
[275] Kaunda (n 246) para 80. This dictum was confirmed in Von Abo v President of the Republic of
South Africa 2009 (5) SA 345 (CC). See further on the power to review the exercise of the government’s
discretion, the separate opinions of Judge Ngcobo in Kaunda (n 246) (para 172) and Judge O’Regan
(paras 243–5, 269–71).
[276] ‘The act of state doctrine in South Africa: Has Kaunda settled a vexing question?’ (2007) 22 SA
Public Law 444.
[277] 2009 (2) SA 526 (T); 2010 (3) SA 269 (GNP); 2009 (5) SA 345 (CC); 2011 (5) SA 262 (SCA).
For a strong criticism of Von Abo, see D Tladi ‘The right to diplomatic protection, the Von Abo decision,
and one big can of worms: Eroding the clarity of Kaunda’ (2009) 20 Stellenbosch LR 14.
[278] According to Prinsloo J, it was ‘difficult to resist the conclusion that the respondents were simply
stringing the applicant along and never had any serious intention to afford him proper protection. Their
feeble efforts, if any, amounted to little more than quiet acquiescence in the conduct of their Zimbabwean
counterparts and their “War veteran” thugs’ (2009 (2) SA 526 (T) para 112). See, too, para 143.
[279] Von Abo (n 278) para 41.
[280] Ibid para 143.
[281] In Von Abo (n 275) the Constitutional Court held that it was unnecessary for Prinsloo J’s order to
be confirmed by this Court.
[282] Von Abo v Government of the Republic of South Africa 2010 (3) SA 269 (GNP) paras 27 and 56.
[283] Von Abo (n 282) para 58(4).
[284] 2011 (5) SA 262 (SCA). This decision is more fully discussed in Chapter 13 in the section on the
discretionary nature of the right to diplomatic protection.
[285] Ibid para 28.
[286] Ibid para 39.
[287] Ibid para 39.
[288] DA v Minister of International Relations (n 90) para 44.
[289] Earthlife Africa (n 81) paras 117–21.
[290] Ibid para 120.
[291] Earthlife Africa (n 81) para 120 (emphasis added).
[292] 2018 (6) BCLR 695 (GP).
[293] The President was represented by South Africa’s High Commissioner to Namibia.
[294] Paragraph 67.
[295] See paras 66 and 67.
[296] However, the Court did not indicate which rights were violated, since the Court did not find that s
34 provided a right to access international tribunals (as opposed to domestic courts) (see para 21, where
the Court records that the Law Society made it clear that it was not arguing that s 34 guaranteed the
right of access to international tribunals).
[297] See paras 66 and 67.
[298] Paragraph 69.
[299] Paragraph 71.
[300] Article 52 provides that ‘[t]his Protocol shall be ratified by Member States who have signed the
Protocol in accordance with their constitutional procedures’.
[301] Article 53 provides that ‘[t]his Protocol shall enter into force thirty (30) days after the deposit of
the Instruments of Ratification by two-thirds of the Member States’. Article 55(1) specifies that ‘all
instruments of Ratification . . . shall be deposited with the Executive Secretary of SADC who shall
transmit certified copies to all Member States’.
[302] Order 3.
[303] Underhill v Hernandez 168 US 250 (1897) 252; Banco Nacional de Cuba v Sabbatino (n 25) 427–
8; J Harrison ‘The American act of state doctrine’ (2016) 47 Georgetown Journal of International
Law 507.
[304] Kirkpatrick v Environmental Tectonics 493 US 400 (1990) 405. See, too, Kadic v Karadzic (1995)
34 ILM 1592 at 1612: ‘[J]udges should not reflexively invoke these doctrines to avoid difficult and
somewhat sensitive decisions in the context of human rights.’
[305] [1982] AC 888 at 931. See, too, Al Jedda v SSD [2010] EWCA civ 758.
[306] FA Mann in Foreign Affairs in English Courts (1986) states: ‘[T]he great danger, not always
avoided, is that by holding a claim non-justiciable the court fails to perform its duty of deciding cases . . ..
It is possible that this judicial duty was violated by the most puzzling pronouncement on justiciability that
can be found in England’ at 69. See, too, JR Crawford ‘Public international law’ (1982) 53 BYIL 253, 267–
8; L Collins ‘Foreign relations and the judiciary’ (2002) 51 ICLQ 485.
[307] Maclaine Watson v Department of Trade and Industry [1988] 3 All ER 257 (CA) 291 (Kerr LJ);
[1989] 3 All ER 523 at 544–5 (Lord Oliver).
[308] Kuwait Airways Corporation v Iraqi Airways Company [2002] UKHL 19, [2002] 2 WLR 1353. At
para 26, Lord Nichols stated: ‘In appropriate circumstances, it is legitimate for an English Court to have
regard to the content of international law in deciding whether to recognize a foreign law.’ See further
paras 28–9, 114–15, 138–40, 148. This decision was cited with approval by Patel J in Van Zyl v
Government of the RSA (n 57) paras 70–1.
[309] Abbasi v Secretary of State for Foreign and Commonwealth Affairs [2002] EWCA Civ 1598,
(2003) 42 ILM 358 paras 57, 66. See, too, R v Bow Street Metropolitan Stipendiary Magistrate, Ex parte
Pinochet Ugarte [1998] 3 WLR 1456 (HL) 1480 (Lord Steyn).
[310] Belhaj and Another v Straw and Others; Rahmatullah v Minister of Defence and Others [2017]
UKSC 3; [2017] 3 All ER 337.
[311] Ibid paras 35–43 (judgment of Lord Mance).
[312] Belhaj (n 310) paras 234 et seq.
[313] IA Shearer (ed) Starke’s International Law 11 ed (1994) 100–1; R Jennings and A Watts
(eds) Oppenheim’s International law 9 ed (1992) vol I at 369; Banco Nacional de Cuba v Sabbatino (n
25) 427–8.
[314] 1999 (2) SA 279 (T).
[315] At 330C.
[316] At 334F–H.
[317] At 334D–E. See the criticism of this decision by N Botha ‘The foreign affairs prerogative and the
1996 Constitution’ (2000) 25 SAYIL 265.
[318] Van Zyl v Government of the Republic of South Africa 2008 (3) SA 294 (SCA) para 5.
[319] 2017 (5) SA 105 (ECP).
[320] Paragraph 92.
[321] Paragraph 86, read with para 95.
[322] Paragraphs 87–8.
[323] Paragraph 95.
[324] Paragraph 96.
[325] Paragraphs 60–85.
[326] Paragraph 61.
[327] Supra (n 54) para 66. By way of example of what would constitute indirect impleading precluded
by state immunity, the Supreme Court of Appeal noted that this would include ‘a civil action against an
individual in respect of actions on behalf of a foreign state, where permitting an action against the
individual would circumvent the state’s immunity’ (Minister of Justice v SALC para 66 fn 41,
referencing Jones v Ministry of the Interior (n 67) as authority).
[328] Section 2(1) and (2) provide as follows:
‘(1) A foreign state shall be immune from the jurisdiction of the courts of the Republic except as provided in this Act or in any
proclamation issued thereunder.
(2) A court shall give effect to the immunity conferred by this section even though the foreign state does not appear in the proceedings in
question.’

[329] Saharawi Arab Democratic Republic (n 319) para 58 (emphasis added).

[330] Ibid para 83
[331] Ibid para 74, where the Court quotes with approval from Lord Sumption’s judgment in Belhaj (n
310).
[332] 2001 (3) SA 893 (CC).
[333] The Court relied, inter alia, on Soering v UK (1989) 11 EHRR 439 and United States v
Burns [2001] 1 SCR 283.
[334] ‘The extra-territorial application of the South African Constitution’ (2003) 120 SALJ 797 at 799.
[335] Kaunda (n 246) paras 46–57. Cf the dissenting opinion of Judge O’Regan op cit paras 248–56.
[336] Ibid para 38.
[337] Kaunda (n 246) para 44. Confirmed in Rootman v President of the RSA, Oxford Reports on
International Law ILDC 469 (ZA 2006), [2006] SCA 80 para 11.
[338] Ibid para 45. Cf R v Cook [1998] SCR 597.
[339] Supra (n 292).

Page 126

Chapter 5
States (Including Recognition and Non-
recognition)

John Dugard
The state is the centre of the international legal order. Clear rules of customary
international law and treaty law proclaim the rights and duties and the privileges and
responsibilities of the state. Yet the birth and origins of the state are shrouded in legal
uncertainty. There is confusion over whether the creation of a state is a question of fact
or of law, whether the fulfilment of certain factual criteria suffices or whether the
imprimatur of the law is also required. The death or demise of a state is an area from
which international law keeps its distance. States are sovereign, but the notion of
sovereignty and its implications are better left for philosophical discourse rather than
legal definition. Although the concepts of the state and statehood are central to our
understanding of international law, this remains the most unsettled and uncharted
branch of international law.
Until recent times only states were subjects of international law. Today other
entities, such as inter-governmental organisations, are also accepted as international
persons with rights and duties under international law. [1] The state, however, remains
the principal actor [2] in the international arena and the raison d’être of the international
legal system. It is therefore necessary to examine the notion of statehood and to
consider how entities claiming to be states are accepted into the community of nations
as full subjects of international law. [3]
International law is a product of European, Christian civilisation and was for many
centuries a European law of nations. [4] When a new state appeared in Europe, it was
received into the community by the old members of the European society of nations.
After the decolonisation of the Americas, non-European, Christian states were admitted
to the ‘club’. With the admission of the Ottoman Empire to the community of nations in
1856, international law ceased to be a law between Christian states only. Thereafter,
non-European, non-Christian
Page 127
states such as Japan, China, Siam and Persia were accepted as states. The creation of
the League of Nations, and later the United Nations, provided a new mechanism for the
collective admission of states to the international community. [5] Although the existing
states retained the right to admit new members by unilateral acts of recognition on
their part, there is no doubt that the League of Nations and the United Nations have
both facilitated and accelerated the process of international acceptance.
The acceptance of new states into the international community is effected mainly by
means of ‘recognition’ on the part of the existing states. Although this process is
determined or influenced by political considerations, certain factual criteria for
statehood have been adopted to guide the decision to recognise states. Consequently, it
is necessary to consider both these criteria and the process of recognition for an
understanding of the creation of states.
South African history provides a good illustration of the process of state creation.
During the apartheid era, the South African government sought to promote the
independent statehood of four Bantustans—Transkei, Bophuthatswana, Venda and
Ciskei (the so-called TBVC states)—in order to give legitimacy to the policy of separate
development, as the apartheid regime preferred to term apartheid. [6] The refusal of the
international community to recognise these entities as states demonstrates how law
and politics interact in the creation—or non-creation—of states. More recently, the
process of state creation and its hazards have been illustrated by the recognition as
states of Palestine and Kosovo by some states but not others and by the manner in
which key states have obstructed their admission to the United Nations.
 Page 128

The criteria for statehood

The traditional criteria for statehood are described in the Montevideo Convention of
1933, [7] which provides:
The state as a person of international law should possess the following qualifications: (a) a
permanent population; (b) a defined territory; (c) government; and (d) capacity to enter into
relations with other states.
Although this is a regional convention, signed by 15 Latin American states and the
United States, it has come to be accepted as the definitive statement on the criteria for
statehood. Its requirements are factual by nature but ‘they have been accepted as the
normative starting point on state recognition, not only by scholars and lawyers, but
also, more importantly, by states’. [8] Consequently, they must be seen as constituting
the legal conditions for statehood—conditions that must be considered by third states in
deciding whether to grant recognition to a seceding entity.
More recently, since human rights and self-determination have become more
important in international law, it has been suggested that for a new entity to succeed in
a claim for statehood, it should meet the standards and expectations of the
international community on these subjects. This development was given support by
Guidelines on the Recognition of New States in Eastern Europe and in the Soviet Union
issued by the European Community in 1991, [9] and extended to Yugoslavia, which
sought to make recognition of states dependent on compliance with international norms
relating to self-determination, respect for human rights and the protection of minorities.

1 Permanent population

No minimum population size is required. Today, more than 50 states have populations
of less than one million, and Tuvalu and Nauru have less than 15 000 inhabitants each.
The microstate presents problems in the United Nations as it seems unfair that India,
with a
Page 129
population of 1.3 billion, should have the same voting rights as the Seychelles, with a
population of 95 000. Attempts to remedy this inequality in the United Nations,
however, have been abandoned and microstates are today accepted as full members of
the international community. [10]

2 Defined territory

It is not a necessary prerequisite of statehood that a state has clearly defined and
undisputed borders. [11] The borders of Israel and Palestine, for example, have been a
subject of dispute for over 50 years. The state, however, should have a stable
community within an area over which its government has control. Furthermore, it is not
necessary for a state to occupy a single territory. Prior to the creation of Bangladesh in
1971, Pakistan was divided into East and West Pakistan, separated by India; and,
Canada separates Alaska from the rest of the United States. It was therefore difficult to
argue that Bophuthatswana lacked statehood on the ground that it comprised a number
of separate territories. [12]
3 Government

In order to meet this requirement a state must have a government that is in effective
control of its territory and that is independent of any other authority. [13] The fact that a
government receives substantial financial aid from another state would not in itself
appear to affect its formal independence. This was, however, one of the reasons given
by the United Kingdom for its refusal to recognise Bophuthatswana. [14]
The requirement of effective government was relaxed in the process of
decolonisation, which saw many former colonies recognised as states with governments
still heavily dependent on the colonial power. Political considerations, too, have led to a
weakening of this requirement. The recognition of Croatia and Bosnia-Herzegovina by
the European Community in 1992, and their subsequent admission as ‘states’ to the
United Nations in the same year, cannot be reconciled with this criterion as both were
embroiled in a civil war,
Page 130
in which no authority exercised effective control over either territory, at the time of
their recognition. [15]
Difficulties arise when a recognised state degenerates into anarchy or civil war (as in
the instances of Yemen and Syria).
Although logic might suggest that such an entity—a ‘failed state’—should cease to be
a state, the practice of states provides no support for such a view. (The phenomenon of
the ‘failed state’ is considered below.)
A state does not cease to be a state when it is militarily occupied by another state.
During World War II, Germany occupied countries such as the Netherlands, Belgium
and Denmark, which clearly remained states even though they were subject to the
control of German forces. Palestine occupies an unusual position in this respect as it
had no government when it was first occupied by Israel in 1967. The subsequent
establishment of a government, the recognition of Palestine by some 138 states and its
admission to many international institutions reserved for states has remedied this
defect and the statehood of Palestine, albeit under Israeli occupation, is today widely
recognised.

4 Capacity to enter into relations with other states

The capacity of a state to enter into relations with other states is a consequence of
independence. [16] If an entity is subject to the authority of another state in the
handling of its foreign affairs, it fails to meet this requirement and cannot be described
as an independent state.
The position of the TBVC states illustrates the complexities of this criterion for
statehood. Although they enjoyed full constitutional independence under the respective
Status Acts, which conferred independence on them, [17] and the formal capacity to
enter into treaties [18] and diplomatic relations, they were unable to conduct
international relations with states other than South Africa on account of their non-
recognition by the international community. In S v Banda,
Page 131
in which the statehood of Bophuthatswana was challenged before a municipal court of
that territory in a case of treason, Friedman J held that this requirement of statehood
did not apply where the state had ‘an infrastructure to implement relations with other
states should it be given the opportunity to do so’ but was ‘precluded from so doing due
to political considerations’. [19] ‘An entity possessing all the other essentials of being a
state’, said the judge, ‘cannot be regarded as not having the capacity to enter into
relations with other states if it is denied the opportunity to demonstrate this capacity in
practice.’ [20] This view, which ignores the role played by recognition in the creation of
states, is out of touch with reality for, as Hedley Bull has written, a community that
claims to be sovereign ‘but cannot assert this right in practice, is not a state properly
so-called’. [21] Formal independence does not produce the capacity to enter into
relations with other states. In addition, the claimant entity must be recognised by at
least some states to make this theoretical capacity a reality to give
it functional independence. Although recognition may not create a state, it seems it
should be taken into account in deciding whether the fourth requirement is met.
The Turkish Republic of Northern Cyprus (TRNC) occupies the same position today as
did the TBVC states: it is recognised by no state other than its creator, Turkey.
In Caglar v Billingham (Inspector of Taxes), in which the British Special Commissioners
of Inland Revenue were concerned with the exemption from liability for income tax for
officials of a foreign state, the Commissioners adopted the above reasoning and held
that the TRNC failed to qualify as a state:
In view of the non-recognition of the Turkish Republic of Northern Cyprus by the whole of the
international community other than Turkey we conclude that it does . . . not have functional
independence as it cannot enter into relations with other states. It does not therefore satisfy
the fourth requirement of statehood. [22]
Page 132
Palestine and Kosovo present no difficulties in this respect. Although neither are
members of the United Nations, they are both recognised by over 100 states and
conduct relations with these states, thereby demonstrating their capacity to conduct
international relations.

The evolving statehood of South Africa and its capacity to enter into relations with other
states
In 1910 the four British colonies of the Cape, Natal, Transvaal and the Orange River
joined to form the Union of South Africa. [23] This Union met the first three requirements
of statehood—population, territory and effective government—but lacked the capacity
to enter into relations with other states. The Union was not able to enter into treaties
without the assistance of the United Kingdom, [24] to establish diplomatic relations with
other states, [25] and to make war or peace. [26] Moreover, constitutionally, the South
African Parliament was subordinate to Westminster as a result of the Colonial Laws
Validity Act, which recognised the right of the British Parliament to legislate for the
Dominions. [27] In these circumstances, South Africa could not claim to enjoy the right
to enter into relations with other states or to be independent.
After World War I, South Africa evolved towards a position in which it enjoyed the
capacity to enter into relations with other states free from the control of Britain. In
1919 it became an original member of the League of Nations [28] and thereafter
asserted its independence from Britain by pursuing independent policies within that
organisation. [29] Moreover, in 1920 South Africa was appointed by the League of
Nations as Mandatory Power over South West Africa with direct
Page 133
accountability to the Council of the League. [30] That South Africa was now free to
conduct its own foreign policy and to enter into treaties on its own account was
confirmed by the Imperial Conference of 1926, which resolved that the Dominions and
Britain were ‘equal in status, in no way subordinate one to another in any aspect of
their domestic or external affairs’. [31] In response, South Africa promptly established its
own Department of External Affairs and proceeded to establish diplomatic missions in
many parts of the world. [32] Treaties were now entered into by the South African
executive on its own. [33] South Africa’s constitutional position was brought into line with
this state of affairs in 1931 by the Statute of Westminster, which made it clear that
Westminster no longer enjoyed the right to legislate for the Dominions. [34] This was
confirmed by the South African Parliament in the Status of Union Act [35] and the Royal
Executive Functions and Seals Act. [36] By the mid-1930s, only one argument could be
seriously advanced to support South Africa’s subordination to Britain—namely the
alleged right of Britain to declare war on behalf of South Africa. [37] The correctness of
this argument was, however, disproved in 1939 when South Africa declared war against
Germany independently of Britain.
It is difficult to pinpoint the exact moment at which South Africa became an
independent state. Some may point to its admission to the League of Nations as the
decisive time, while others may prefer the declaration of war in 1939. The correct view,
it seems, is that South Africa acquired full international status at the moment that it
acquired the capacity to enter into relations with other states and this capacity was
recognised by Britain: 1926 therefore appears to be the year in which South Africa
became a fully sovereign independent state under international
law. [38] Constitutionally, independence was
Page 134
achieved only in 1931, with the passing of the Statute of Westminster, which removed
South Africa’s legislative subordination to the United Kingdom.

5 Respect for human rights and self-determination

The promotion of human rights has become a concern of international law only since
World War II. Before 1945, the manner in which a state treated its own citizens was
generally [39] not regarded as a factor to be considered in deciding whether to admit a
state to the community of nations. Neither the League of Nations nor any state raised
objections to South Africa’s racial policies when it became an independent member of
the community of states. Since 1945, many new states with poor human rights records
have been recognised and admitted to the United Nations. Moreover, there has been no
serious suggestion that the recognition of states with outrageous human rights records
should be withdrawn. [40]
Despite this, states in recent times have alluded to respect for human rights and
self-determination as a precondition for the recognition of statehood. When the Soviet
Union dissolved in 1991, the European Community indicated that it would recognise
only those parts of the former Soviet Union claiming to be independent states that
afforded some evidence of a willingness and capacity to protect and respect human
rights. [41] Similar assurances were sought from Slovenia, Croatia, Bosnia and
Herzegovina and Macedonia as a precondition for their recognition as states. [42]
The idea that an entity that denied human rights and self-determination would not
qualify as a state was initially mooted in the context of the requirement of effective
government. A government that denied basic rights, it was suggested, could not be
truly
Page 135
organised and effective. [43] But soon this was translated into a new criterion for
statehood. Writing on the non-recognition of Rhodesia after its unilateral declaration of
independence in 1965, JES Fawcett declared:
But to the traditional criteria for the recognition of a regime as a new state must now be added
the requirement that it shall not be based upon a systematic denial in its territory of certain
civil and political rights, including in particular the right of every citizen to participate in the
government of his country, directly or through representatives elected by regular, equal and
secret suffrage. This principle was affirmed in the case of Rhodesia by the virtually unanimous
condemnation of the unilateral declaration of independence by the world community, and by
the universal withholding of recognition of the new regime which was a consequence. [44]
Today, support for the linkage of effective government with respect for human rights is
to be found in the proposition that there is an emerging norm of democratic entitlement
in international law. [45]
Although Fawcett’s view has been well received, [46] it is open to a number of
criticisms. [47] First, if the systematic denial of human rights, including the right to
participate in government by means of free elections, is to become a bar to statehood,
it would mean that many states would cease to qualify as states and face withdrawal of
recognition. It would hardly be fair to limit this requirement to new states only and to
expect a higher moral standard of conduct from them than from the existing members
of the international community. Secondly, state practice does not provide support for
the proposition that compliance with human rights and self-determination norms is now
an additional criterion for statehood. This is particularly true of entities claiming
statehood outside the confines of decolonisation, which is the concern of the
decolonised world. The expectation that the recognition as states of entities emerging
from the dissolution of the former Yugoslavia would be guided by considerations of
human rights and respect for minorities was not fulfilled. The European Community
failed to implement its own guidelines for recognition, or to follow the advice of the
Arbitration Commission it had set up to monitor compliance with its guidelines. [48]
Page 136
Croatia was recognised before assurances relating to respect for minorities were given,
let alone implemented, while Bosnia was recognised at a time when one of the great
human rights tragedies of the 20th century had begun to unfold. In both cases, states
preferred to base the decisions to recognise on their perceptions of the political realities
of the region. [49]
State practice on this subject is largely confined to the political organs of the United
Nations, and from resolutions adopted by these bodies it appears that entities such as
Rhodesia, the TBVC states and the Turkish Republic of Northern Cyprus have not been
faulted for failure to comply with the requirements of statehood but denounced for
violation of certain peremptory norms of international law which result in their
‘illegality’, ‘invalidity’ and ‘nullity’. [50]

Recognition [51]

This chapter is concerned with the recognition of states as subjects of international law
and not with the recognition of the governments in control of states. The latter topic is
dealt with separately in Chapter 6. Although a government may be recognised as a de
facto or de jure government, no such distinction is possible in the case of a state. An
entity is either recognised as a state or it is not.
Recognition may be either unilateral or collective. In the former case, an individual
state, already accepted as a state, recognises that an entity claiming to be a state
meets the factual requirement of statehood and is therefore to be regarded as a state,
with the rights and duties attached to statehood. Recognition is a precondition for the
establishment of diplomatic relations, but it is not essential that the recognising state
enter into diplomatic relations with the new state. Collective recognition occurs when a
group of states, such as the European Community or the United Nations, recognises the
existence of a claimant state directly, by an act of recognition, or
Page 137
indirectly, by the admission of the state to the organisation in question.
Unilateral recognition is the most orthodox method of recognition. Collective
recognition is still relatively uncertain and controversial.

Unilateral recognition

Different views are held about the purpose and consequences of recognition. Two
principal schools of thought dominate this debate—the constitutive and the
declaratory. [52] According to the constitutive school, the recognition of a claimant entity
as a state creates or constitutes the state. [53] Recognition therefore becomes an
additional requirement for statehood. The declaratory school, on the other hand,
maintains that an entity becomes a state on meeting the factual requirements of
statehood and that recognition by other states simply acknowledges (declares) ‘as a
fact something that has hitherto been uncertain’. [54]
There are several serious objections to the constitutive view. First, if the claimant
state is recognised by state A and not by state B, it becomes in effect both a state and
a non-state. Palestine is recognised as a state by 138 states but is not recognised as a
state by the United States and some 50 other states. Is it a state only for those states
that recognise it but not for other states? Clearly such uncertainty is undesirable.
Secondly, if an unrecognised state is not a state, it is not entitled to the rights or
subject to the obligations of international law. In practice, this consequence is not real.
Many Arab states formally refuse to recognise Israel as a state but they nevertheless
accept that they have legal obligations towards Israel. Israel likewise accepts that it has
legal obligations towards states that refuse to recognise it.
What Hersch Lauterpacht described as the ‘grotesque spectacle’ [55] of an entity
being a state for some states and not for others could be avoided if states were to
recognise entities as soon as they complied with the requirements of statehood set out
in the Montevideo Convention. Thus, Lauterpacht contended that once these
requirements are met,
the existing states are under the duty to grant recognition. In the absence of an international
organ competent to ascertain and authoritatively to declare
Page 138
the presence of the requirements of full international personality, states already established
fulfil that function in their capacity as organs of international law. In thus acting they
administer the law of nations. This legal rule signifies that in granting or withholding
recognition states do not claim and are not entitled to serve exclusively the interests of their
national policy and convenience regardless of the principles of international law in the
matter. [56]
Unfortunately, Lauterpacht’s contention is not supported by state practice: states do
not regard themselves as being under a legal duty to recognise entities that comply
with the requirements of statehood, [57] and political considerations influence their
decisions. The notion that recognition is a political act appears in many guises.
Sometimes it is claimed that recognition is a matter of policy, not law, while on other
occasions it is argued that the facts of a particular case are sui generis, which means
that they are not governed by law or legal precedent.
In most cases, entities that meet the requirements of the Montevideo Convention are
recognised, so the process is not entirely arbitrary. It is, however, necessary to
appreciate that political considerations may prompt a state to recognise an entity
prematurely or to refuse to grant it recognition. For example, in 1903 the United States
immediately recognised Panama when it seceded from Colombia in order to prevent
Colombia from asserting its authority over its rebellious province—with the aim of
securing for the United States the right to build the Panama Canal. [58] In 1992 member
states of the European Community recognised Croatia and Bosnia-Herzegovina before
they fulfilled the requirements of statehood for a wide range of political reasons,
including the hope that recognition would prevent the escalation of violence. [59] On the
other hand, the United States, which is more influenced by political considerations in its
recognition practice than other states, refuses to recognise Palestine under the
influence of its domestic pro-Israel lobbies.
The political nature of recognition has prompted support for the
Page 139
declaratory school, [60] which accepts that an entity that meets the requirements of
statehood becomes a state regardless of recognition. [61] Inevitably this approach was
invoked by supporters of the TBVC states to justify their existence. In S v
Banda, [62] Friedman J conducted an enquiry into the respective merits of the
constitutive and declaratory approaches and found in favour of the latter on the
grounds that (i) it is supported by most writers; [63] and (ii) it is more objective and less
politically subjective than the constitutive approach. [64] Friedman J is correct in saying
that the declaratory approach is less arbitrary than the constitutive and that it enjoys
more support. He is wrong, however, to ignore the necessity for at least some
recognition by other states. A state without the recognition of any state other than its
creator [65] cannot demonstrate its capacity to enter into relations with other states and,
from a functional point of view, cannot be described as a state. [66] While it would be
ridiculous to deny the statehood of Palestine, which has demonstrated its capacity to
enter into relations with other states by entering into diplomatic relations with over 100
states, it would be equally ridiculous to accord statehood to an entity that produces no
evidence of such a capacity other than its governmental structure—as was the case
with Bophuthatswana and as is the case with the Turkish Republic of Northern Cyprus.
In the final resort, therefore, recognition does have a role to play in the creation of a
state. This explains the complexity of the constitutive versus declaratory debate. Most
declaratorists are
Page 140
compelled to acknowledge the need for at least some recognition on the part of existing
states as a precondition of statehood. [67]

The method of recognition [68]

No rules are prescribed for the act of recognition. Usually it will take the form of a
public declaration by the recognising state which is conveyed to the claimant state. In
some cases, it may be implied from the conduct of the recognising state, but such an
inference should not be drawn too readily. For conduct to imply recognition it must
provide clear evidence of an intention to recognise. Thus, while recognition may be
inferred from an exchange of diplomats and the signing of a bilateral treaty between
the two states, it should not be inferred from an exchange of consular agents or the
fact that both states are parties to a multilateral agreement or are members of the
same international organisation.
The question whether South Africa impliedly recognised Rhodesia from 1965 to 1980
was much debated. Despite calls from the United Nations for the isolation of Rhodesia,
South Africa maintained diplomatic relations with Rhodesia, continued to trade with the
rebellious colony, and sent police to Rhodesia to assist the Rhodesian security forces.
Although it was argued in some quarters that this amounted to implied
recognition, [69] the South African government persistently denied that it recognised
Rhodesia as an independent state and instead maintained an attitude of neutrality in
the dispute between Rhodesia and the United Kingdom. [70]

Collective recognition by the United Nations and

decolonisation [71]

In recent years the European Community (now the European Union) has sought to
speak with one voice on the recognition of new states in Europe. [72] This is a wise
policy that has helped to produce some consistency in recognition practice in Europe.
Here, states have exercised their individual right of recognition collectively in a manner
which does not depart substantially from traditional recognition
Page 141
practice. More controversial is the question whether admission to the United Nations
constitutes recognition.
The Charter of the United Nations provides for two categories of members: original
members and those subsequently admitted by the organisation. According to article 3:
The original Members of the United Nations shall be the states which, having participated in
the United Nations Conference on International Organization at San Francisco, or having
previously signed the Declaration by United Nations on January 1, 1942, sign the present
Charter and ratify it in accordance with Article 110.
Subsequent membership is regulated by article 4, which provides:
1.
Membership in the United Nations is open to all other peace-loving states which accept the
obligations contained in the present Charter and, in the judgment of the Organization, are able
and willing to carry out these obligations.
2.
The admission of any such state to membership in the United Nations will be effected by a
decision of the General Assembly upon the recommendation of the Security Council.
Membership in the United Nations is limited to states only. This is clear from articles 3
and 4, but it is reaffirmed by numerous other references to ‘state’ in the Charter which
indicate that the rights and obligations contained in the Charter are linked to
statehood. [73]
Fifty-one states, including South Africa, were original signatories to the Charter of
the United Nations. During the first decade of the organisation’s history only nine new
member states were admitted as a result of disagreements between the major powers,
but in 1955, following a ‘package deal’ between the major powers, 16 new states were
admitted. Once this impasse was broken, few restraints were placed on the admission
of applicant states to the world body, and, as decolonisation swept Asia, Africa, the
Caribbean and the Pacific Ocean, the membership of the United Nations multiplied. The
dissolution of the Soviet Union, Yugoslavia and Czechoslovakia in the early 1990s added
further to the proliferation of states. Today there are 193 members of the United
Nations.
In 1955 there were 76 member states of the United Nations. Thirty years later,
membership had more than doubled. Of the states admitted since 1955, most are the
products of the programme of decolonisation inspired and orchestrated by the United
Nations. Some 50 of these post-1955 member states have populations of less than one
million.
The Charter of the United Nations, as framed in 1945, implicitly recognised the
legitimacy of colonialism, but in Chapter XI it
Page 142
introduced the principle of international accountability for colonial administration, for
colonial powers undertook to transmit information concerning their territories to the
Secretary-General of the United Nations. [74] Moreover, in article 73, administering
states undertook to develop self-government (but not independence) in the territories
under their administration. Chapter XII, in establishing the International Trusteeship
System for mandated territories and the colonies of the vanquished states of World War
II, went still further and proclaimed ‘development towards self-government or
independence’ [75] of these trust territories to be an objective of the system. Finally, the
Charter affirmed the principle of ‘self-determination of peoples’. [76]
The Charter of the United Nations established a dynamic world organisation. Change
was foreseen in 1945. But it is unlikely that the founding fathers of the Charter could
have foreseen the changes which have been effected in the field of colonialism under
the mantle of the Charter’s carefully phrased provisions. Within two decades of the San
Francisco Conference, with no amendment to the Charter, the position had undergone a
complete transformation: the distinction between colony and trust territory had lost its
importance, self-determination was widely recognised as a legal right, the legitimacy of
colonialism was denied, and organs of the United Nations freely asserted the right to
decolonisation.
This transformation was the result of a gradual process, but if one event was to be
singled out for special attention, it was the adoption of the Declaration on the Granting
of Independence to Colonial Countries and Peoples (Resolution 1514(XV)) by the
General Assembly—by 90 votes to none with nine abstentions [77]—on 14 December
1960. This Declaration not only outlawed colonialism but it gave the blessing of the
United Nations to the rapid creation of new independent states, with little regard for
compliance with the traditional requirements of statehood. In its preamble, Resolution
1514(XV) considers ‘the important role of the United Nations in assisting the movement
for independence in Trust and Non-Self-Governing Territories’, welcomes ‘the
emergence in recent years of a large number of dependent territories into freedom and
independence’, and ‘solemnly proclaims the necessity of bringing to a speedy and
unconditional end colonialism in all its forms and manifestations’. The Declaration then
calls for ‘immediate steps’ to be taken in all
Page 143
non-independent territories for the transfer of ‘all powers to the peoples of those
territories, without any conditions or reservations, in accordance with their freely
expressed will and desire, without any distinction as to race, creed or colour, in order to
enable them to enjoy complete independence and freedom’. It adds that ‘[i]nadequacy
of political, economic, social or educational preparedness should never serve as a
pretext for delaying independence’.
Although Resolution 1541(XV), adopted in the same year, declares that non-self-
governing territories may exercise their right of self-determination in one of three
ways—by becoming independent, by entering into a ‘free association with an
independent state’ or by integrating with an independent state [78]—United Nations
practice shows a definite preference for independence as the ideal form of
decolonisation. Evidence of independent statehood would present little difficulty, as this
would be provided by admission to the United Nations.
In 1960, the year of Resolution 1514(XV), 17 newly independent states were
admitted to membership. Thereafter, the United Nations experienced a steady growth
in membership, with the size of its new members growing smaller each year. Few
questions were asked about the statehood of the new applicants for membership in the
United Nations. In most instances, the colonial power’s certification of statehood,
demonstrated by the ceremonial act of independence, and the Special Committee on
Decolonization’s assertion of independence, provided sufficient ‘evidence’ of statehood
for the purposes of admission to the United Nations.
The increase in United Nations membership has largely satisfied the decolonisers
and, at the same time, achieved a measure of universality of which the world body may
be justifiably proud. But the speed with which this has been accomplished inevitably
raises doubts as to whether the traditional criteria for statehood have been observed
and still remain intact. [79]
The conclusion seems unavoidable that the self-determination of peoples, which has
become a primary value in the contemporary international order, has led to a relaxation
of some of the requirements
Page 144
of statehood for the purposes of admission to the United Nations. While the
requirements of permanent population and defined territory remain intact, it seems that
others, such as effective government and independence, are no longer strictly insisted
on where they run counter to developments in international law regarding the right of
self-determination. [80] Once a state is admitted to the United Nations, its acceptance as
a state for all purposes is assured. This explains the alacrity with which claimant states
seek admission to the United Nations. This is well illustrated by the case of South
Sudan.
On 9 July 2011 South Sudan declared its independence with the consent of its
‘parent’ state, Sudan. On 14 July it was admitted to the United Nations. Most states did
not have the opportunity to formally declare that they recognised South Sudan. Nor did
they bother to do this as their vote for the admission of South Sudan to the United
Nations was taken as acceptance of the fact that South Sudan enjoyed international
legal personality. It was clear that the admission of South Sudan to the United Nations
established its statehood.
Today, apart from Israel, whose statehood is still denied by many Arab states, all
members of the United Nations are accepted as states despite the fact that several
probably would not have received widespread recognition by individual states had they
been left to make a determination of statehood in accordance with the traditional
criteria. Thus, it is fair to conclude that many states have achieved statehood by
admission to the United Nations and that this procedure for recognition co-exists
alongside the traditional method of unilateral recognition. [81] Any description of the law
of recognition that fails to take account of this development cannot lay claim to be an
accurate reflection of state practice.
The claim that admission to the United Nations constitutes or confirms the existence
of a state has important implications for the debate between ‘constitutivists’ and
‘declaratorists’. The main criticisms directed at the constitutive school are, first, the
anomalous situation that arises where a state is recognised by state A but not by state
B and is therefore both an international person and not an international person at the
same time, and, secondly, the fact that the constitutive doctrine gives individual states
the arbitrary power to
Page 145
recognise an entity as a state or to withhold recognition. Both these weaknesses in the
constitutivist position are remedied by the collective recognition of states through the
United Nations. If all member states within the United Nations recognise each other’s
existence as states—with the possible exception of Israel by some of the Arab states—it
follows logically that the ‘grotesque spectacle’ [82] of an entity being a state for some
states but not for others is no longer a practical possibility. Furthermore, the arbitrary
and subjective individual state decision is replaced by a collective decision of the United
Nations.
In 1951, in his International Law of Recognition, Chen predicted that when the
United Nations had attained complete universality, the notion of recognition would
‘wither away’ and membership of the United Nations would become the sole standard of
relations between states. [83] Unfortunately, this state of affairs has not yet been
achieved. Palestine, recognised by 138 states, remains outside of the United Nations
largely because of the opposition of the United States, while Kosovo, recognised by 117
states, is kept out of the United Nations by the veto of Russia. The fate of such entities,
whether they are described as states or partial states, ensures that unilateral
recognition remains a crucial procedure for establishing or confirming the existence of
states.

Collective non-recognition [84]

The United Nations plays an important role in the admission of new states to the
international community by the process of collective recognition. Conversely, it may
block the acceptance of a state by means of collective non-recognition. The failure of
the Bantustan (TBVC) states and the Turkish Republic of Northern Cyprus to qualify as
states is better explained in terms of non-recognition than by reference to the criteria
for statehood.
The doctrine of non-recognition has its origin in the non-recognition of the puppet
state of Manchukuo. When Japan invaded the Chinese province of Manchuria in 1932
and set up the state of Manchukuo, the Secretary of State of the United States, Mr
Henry Stimson, declared that the United States would not recognise Manchukuo on the
ground that it had been created in violation of the Pact of Paris of 1928, in which states
renounced war. This was followed by a resolution of the Assembly of the League of
Nations calling upon its members not to recognise Manchukuo. Jurisprudentially,
 Page 146
the doctrine of non-recognition is founded on the principle of ex injuria jus non oritur.
According to Hersch Lauterpacht:
Non-recognition is based on the view that acts contrary to international law are invalid and
cannot become a source of legal rights for the wrongdoer. That view applies to international
law as one of the ‘general principles of law recognised by civilized nations’. [85]
When this doctrine of non-recognition was first expounded, the idea that there were
peremptory norms or jus cogens was undeveloped. Today it is accepted that there are
certain basic norms upon which the international order is founded and that these are
peremptory and may not be derogated from under any circumstances. [86] The modern
law of non-recognition takes cognisance of this development. An act in violation of a
norm having the character of jus cogens is illegal, and is therefore null and void. This
applies to the creation of states and to the acquisition of territory. States are under a
duty not to recognise such acts under customary international law and in accordance
with the general principles of law. [87] Resolutions of the Security Council and the
General Assembly are, from a jurisprudential perspective, declaratory in the sense that
they confirm an already existing duty on states not to recognise such situations. In
practical terms, such resolutions are essential as they provide certainty by substituting
for the decision of an individual state a collective determination of illegality and nullity.
The above doctrine of non-recognition is endorsed by the International Law
Commission’s 2001 draft articles on the Responsibility of States for Internationally
Wrongful Acts. [88] Articles 40 and 41 provide that no state shall recognise as lawful a
situation created by a serious breach of an obligation arising under a peremptory norm
of general international law.
The International Court of Justice has given support to this doctrine in the Kosovo
Opinion by acknowledging that Security Council resolutions on Southern Rhodesia,
Northern Cyprus and the Republic of Srpska, determining that resolutions of
independence were illegal, were ‘connected with the unlawful use of force or other
Page 147
egregious violations of norms of general international law, in particular those of a
peremptory character (jus cogens)’. [89]
The following peremptory norms have been recognised by the United Nations for the
purposes of non-recognition:
(a)
The prohibition on aggression. Following the Iraqi invasion of Kuwait in 1990, the
Security Council adopted a resolution declaring that the annexation of Kuwait had
‘no legal validity’ and that states were required not to recognise that
annexation. [90] Similar action was taken by the Security Council in respect of the
Turkish invasion of Northern Cyprus and the establishment of the Turkish
Republic of Northern Cyprus as a consequence of this invasion. [91]
(b)
The prohibition on the acquisition of territory by means of force. The Security
Council has called on states not to recognise Israel’s forcible annexation of East
Jerusalem and the Golan Heights. [92] The seizure of Palestinian land arising from
the construction of Israel’s ‘security wall’ falls into the same category. The
International Court of Justice has held that ‘all states are under an obligation not
to recognise the illegal situation resulting from the construction of the wall’. [93]
(c)
 The prohibition of systematic racial discrimination and the suppression of human
rights. This norm, which has its source in the United Nations
Charter, [94] convention, [95] and customary international law, and has been
recognised by the International Court of Justice, [96] was invoked as a basis for
non-recognition of Rhodesia and the Bantustan states. [97]
(d)
The prohibition of the denial of self-determination.
Katanga, [98] Rhodesia, [99] South Africa’s administration of Namibia from 1966 to
1990, [100] and the Bantustan states were subjected to non-recognition for
violation of this norm.
Page 148
The non-recognition by the international community of South Africa’s Bantustan states
of Transkei, Bophuthatswana, Venda and Ciskei (the TBVC states) is best explained in
the context of the doctrine of collective non-recognition. [101] When Transkei was
granted independence in 1976, the General Assembly of the United Nations
immediately condemned the establishment of Bantustans ‘as designed to consolidate
the inhuman policies of apartheid’, rejected Transkei’s independence as ‘invalid’, and
called upon all states to ‘deny any form of recognition to the so-called independent
Transkei’. [102] This call for non-recognition of Transkei was subsequently endorsed by
the Security Council. [103] The United Nations responded in similar fashion to the
‘independence’ of Bophuthatswana, [104] Venda [105] and Ciskei. [106] The resolutions of
the Security Council and General Assembly made it clear that the creation of the
Bantustan states violated a number of norms in the field of self-determination and
human rights and that it was the violation of these norms that prompted the calls for
non-recognition. These norms were the unlawfulness of apartheid, systematic racial
discrimination and the denial of self-determination. [107]
The collective non-recognition of the Bantustan states on the ground that their
creation and continued existence violated peremptory norms of international law
resulted in their invalidity. The easy re-absorption of Transkei, Bophuthatswana, Venda
and Ciskei into the geographical body of South Africa in 1994, without any transfer
agreements, [108] served to confirm the international community’s position that they had
never become states at all. This experience in failed statehood may rightly be construed
as a victory for the United Nations’ policy of non-recognition and the principle of ex
injuria jus non oritur.
Page 149

Self-determination, territorial integrity and secession

The right of self-determination features prominently in contemporary international

law. [109] This chapter has shown the important role played by the right of self-
determination in the process of decolonisation and the creation of states. Before leaving
the topic of statehood, the question must be asked: does the right of self-determination
have a role to play in the creation of new states in a post-colonial world? [110] Or, put
more bluntly, does the right of self-determination give a politically disaffected ethnic
minority within a fully independent state the right to determine its destiny by seceding
from that state and creating a new state? The problem is illustrated by the claims of
Kosovo and Somaliland to separate statehood and by the unsuccessful attempt of
Catalonia to secede from Spain in 2017.
 The question whether international law recognises secession involves a consideration
of the competing claims of two legal titans, self-determination and territorial integrity.
Inevitably, any attempt on the part of the population of a state to secede in the
exercise of the right of self-determination undermines the territorial integrity of the
state in question. The uncertain legal rules and principles governing secession are
designed to seek reconciliation between these two principles.

1 Self-determination

That the right of self-determination is a legal right under international law is no longer
seriously challenged. It is affirmed in the Charter of the United Nations [111] and given
content in Resolution 1514 (XV) of 1960—the Declaration on the Granting of
Independence to Colonial Countries and Peoples—and subsequent resolutions of the
General Assembly. [112] It has been acknowledged as a ‘norm of international law’ in the
context of decolonisation [113] by the International Court of Justice in the Namibia
Opinion [114] and the Western
Page 150
Sahara Case, [115] confirmed outside the context of decolonisation by the International
Covenant on Civil and Political Rights, [116] and recognised as a legal right by many
authors. [117] In 1995, in the East Timor Case, the International Court of Justice
accepted that it has a special status as ‘one of the essential principles of contemporary
international law’ and enjoys ‘an erga omnes character’. [118] This was confirmed in the
advisory opinion on the Construction of a Wall in the Occupied Palestinian
Territory. [119] In 2017, in Saharawi Arab Democratic Republic v Owners and Charterers
of the Cherry Blossom, it was recognised as a legal right by a South African court. [120]
Article 1(1) of the International Covenant on Civil and Political Rights contains a
broad definition of the right:
All peoples have the right of self-determination. By virtue of that right they freely determine
their political status and freely pursue their economic, social and cultural development.
Article 20(1) of the African Charter on Human and Peoples’ Rights likewise declares:
All peoples shall have the right to existence. They shall have the unquestionable and
inalienable right to self-determination. They shall freely determine that political status and
shall pursue their economic and social development according to the policy they have freely
chosen.
The commitment of these conventions and declarations to self-determination is not
matched by clarity. Serious questions remain about whether self-determination is a
right that belongs to colonial peoples only, the meaning and content of the right and
the identity of the people upon whom the right is conferred.

(a) Self-determination as the right of colonial peoples only

The full implications of the right of self-determination recognised by the UN Charter
were unknown but feared by the international community in 1945. Taken to its logical
extreme, self-determination threatened the sovereignty and territorial integrity of all
states with dissatisfied minorities. It therefore came as a relief to states when self-
determination was invoked to further the cause of decolonisation. [121] General
Assembly resolutions on the Granting of Independence
Page 151
to Colonial Countries and Peoples [122] and the Declaration on Friendly Relations among
States [123] asserted the right of self-determination in the context of decolonisation and
decisions of the International Court of Justice gave support to this limitation. [124] The
International Covenants on Civil and Political Rights and Economic, Social and Cultural
Rights that came into force in 1976, however, recognised the right of self-determination
with no restriction to colonial situations [125] and since then, General Assembly
resolutions [126] and judicial decisions [127] have acknowledged that the right is not
confined to decolonisation. The question arose in the proceedings relating to the Kosovo
Opinion of 2010, and although the Court declined to pronounce on this matter, most
states assumed that the right of self-determination applies outside the colonial
context. [128]

(b) External or internal self-determination

Before 1976, when the two International Human Rights Covenants came into force and
self-determination was invoked mainly in the context of decolonisation, the right was
seen to involve secession or external self-determination. After 1976, however, a
distinction was drawn between the right of external self-determination that belonged to
colonial peoples and the right of internal self-determination which comprised the right
of a population or people within a non-colonial state to choose the government of the
state, to participate in a government that represented the whole people of that state
and to have the human rights of all persons within the state respected. Although human
rights instruments draw no distinction between external and internal self-determination,
internal self-determination has become the preferred option in the post-colonial world.
This fails to answer the question of whether a right of external self-determination arises
in a non-colonial situation when the people of a territory or a people within a territory
are denied their right of internal self-determination. [129]
Page 152

(c) The ‘self’ in self-determination

The term ‘people’ or ‘peoples’ in the context of self-determination is often interpreted to
mean all the people within a state—that is its population—and not the different ethnic
groups or ‘peoples’ that comprise a heterogeneous population. [130] This interpretation,
which takes little account of the complexity of the term ‘people’ or ‘peoples’, [131] or the
ambiguity of the relevant instruments, would ensure that minorities within a state are
not accorded the right of secession. Rosalyn Higgins claims that:
The emphasis in all the relevant instruments and in state practice . . . on the importance of
territorial integrity, means that ‘peoples’ is to be understood in the sense of all the peoples of a
given territory. Of course, all members of distinct minority groups are part of the peoples of
the territory. In that sense they too, as individuals, are holders of the right of self-
determination. But minorities as such do not have a right of self-determination. That means, in
effect, that they have no right to secession, to independence, or to join with comparable
groups in other states. [132]
This is not, however, the only interpretation of the term. In Reference re Secession of
Quebec the Supreme Court of Canada held that ‘it is clear that a “people” may include
only a portion of the population of an existing state’. [133] Moreover, two decisions of the
African Commission on Human and Peoples’ Rights accept that a group of persons with
a common culture, language and history, occupying a distinct territory, may constitute
a people for the purpose of self-determination. [134] Although the International Court of
Justice made no finding on this subject in the Kosovo Opinion, by necessary implication
the Court held that the Albanian population of Kosovo constitutes a people for the
purpose of self-determination. [135] It may therefore be accepted that the term ‘people’
includes a group of persons living within an established state, occupying a distinct
Page 153
territory within that state, that shares a common identity, language, religion, culture,
history and traditions. [136]

2 Territorial integrity

The principle of territorial integrity equals self-determination in terms of its legal

credentials. Article 2(4) of the UN Charter obliges all member states ‘to refrain in their
international relations from the threat or use of force against the territorial integrity or
political independence of any state’. The Constitutive Act of the African Union proclaims
as one of its objectives the defence of the territorial integrity of member
states. [137] Many General Assembly resolutions of principle expound the principle of
territorial integrity. The Declaration on the Granting of Independence to Colonial
Countries and Peoples, contained in Resolution 1514 (XV) of 1960, proclaims the right
of self-determination but qualifies the right by declaring that:
Any attempt aimed at the partial or total disruption of the national unity and territorial integrity
of a country is incompatible with the Purposes and Principles of the Charter of the United
Nations (para 6).
The Declaration on Principles of International Law Concerning Friendly Relations and
Co-operation among States in Accordance with the Charter of the United Nations of
1970, contained in Resolution 2625 of 1970, likewise expounds the right of self-
determination but adds that nothing in the paragraphs asserting the right of self-
determination
shall be construed as authorizing or encouraging any action which would dismember or impair,
totally or in part, the territorial integrity or political unity of sovereign independent states
conducting themselves in compliance with the principles of equal rights and self-determination
of peoples . . . and thus possessed of a government representing the whole people belonging
to the territory without distinction as to race, creed or colour. [138]
In the World Summit Outcome Resolution of 2005, states rededicated themselves to
the principle of respect for territorial integrity. [139]
The principle of territorial integrity, guaranteeing respect for existing boundaries, has
been invoked by the United Nations to block secession in many situations, including
cases in which a minority group within a state has attempted to secede. [140] The
Security Council
Page 154
has thus reaffirmed the territorial integrity of Somalia and thereby dismissed the
secessionist aspirations of Somaliland and Puntland. [141]

3 State practice and judicial decisions on secession

Secession does not create problems where the parent state consents, however
reluctantly, to the creation of a new state by a people occupying part of its territory. In
2011 a referendum was held in South Sudan to determine whether the region would
secede from Sudan. After South Sudan voted overwhelmingly in favour of
independence, the central government of Sudan accepted the outcome of the
referendum. South Sudan became an independent state, with the support of Sudan,
and was admitted to the United Nations in July 2011.
Problems arise where the government of a state insists on the maintenance of its
territorial integrity and resists, often forcibly, the secession of a region of the state
which is dissatisfied with the central government and prepared to take up arms to
assert its independence. It is in such situations that the principles of self-determination
and territorial integrity come into conflict and the rules of international law for the
reconciliation of this conflict are least clear.
In practice, recognition plays a determining role in deciding on the success of a
particular secession. [142] If a sufficient number of states recognise a seceding region as
an independent state, this will give credibility to its claim to statehood. Admission to
the United Nations will place this claim beyond doubt. As shown above, recognition
remains the right of each state and many states are guided by political considerations
in exercising this right rather than the criteria for statehood of the Montevideo
Convention. Consequently, it is difficult to discern clear principles from recognition
practice in this field.
Some take the position that secession in the non-colonial context is absolutely
prohibited. In 1970, Secretary-General U Thant, speaking at the time of Biafra’s
attempted secession from Nigeria, declared:
As far as the question of secession of a particular section of a state is concerned, the United
Nations attitude is unequivocal. As an international organization, the United Nations has never
accepted and does not accept, and I do not believe will ever accept the principle of secession
of a part of its
Page 155
member state. [143]

Such a statement is correct in respect of purported secessions that have been

denounced by the United Nations as being in violation of peremptory norms of
international law. [144] The United Nations has made such determinations in the cases of
Katanga, [145] Rhodesia, [146] Transkei, Bophuthatswana, Venda and Ciskei, [147] the
Turkish Republic of Northern Cyprus [148] and Republica Srpska [149] on the grounds that
such ‘states’ were created in violation of the prohibitions on denial of the right of self-
determination, racial discrimination and the use of force. The suppression of demands
for secession in the cases of, for example, Biafra (from Nigeria), [150] Chechnya (from
the Russian Federation), the Basque region and Catalonia (from Spain), Kashmir (from
India), the Uighur region (from China), Bougainville (from Papua New Guinea), Aceh
(from Indonesia) and Kurdistan (from Iraq) give additional credence to the statement of
U Thant. This opinion, premised on extreme respect for territorial integrity, is not fully
supported by state practice as there are instances in which secessions have succeeded
with full recognition and admission to the United Nations.
The cases in which secession has been successful are mainly characterised by two
phenomena: first, the separate identity of the seceding region in geographical,
historical or constitutional terms; and, secondly, the denial of the right of internal self-
determination, accompanied by the violation of human rights, and the exhaustion of
attempts to secure internal self-determination. (The secession of Israel from the
mandate territory of Palestine in 1948 is an anomaly.) [151]
Secessionist claims made by peoples occupying a distinct area with an identity of its
own are more favourably considered by the international community as they do not
violate, or violate to a lesser
 Page 156
extent, the principle of territorial integrity. This is illustrated by the examples of
Bangladesh, Slovenia, Croatia, Bosnia-Herzegovina, Macedonia and Eritrea.
Bangladesh, [152] previously East Pakistan, was geographically separated from West
Pakistan by India and did not share a common language, history, culture or economy
with it. The two parts were united only by a common religion, Islam. In 1971
Bangladesh seceded from Pakistan and, having secured a large number of recognitions,
was admitted to the United Nations in 1974. The Socialist Federal Republic of
Yugoslavia, before its dissolution, comprised six federal states—Slovenia, Croatia,
Bosnia-Herzegovina, Macedonia, Serbia and Montenegro—each with a separate
territorial, historical and constitutional identity. Whether Slovenia, Croatia, Bosnia-
Herzegovina and Macedonia seceded from the Federal Republic of Yugoslavia in the
1990s or whether the Federal Republic dissolved is disputed. [153] It is, however,
instructive that the Badinter Arbitration Commission, established by the European
Community to oversee the dissolution of the Federal Republic of Yugoslavia, took the
position that the federal boundaries of the constituent federal states within which new
states were created complied with the principle of uti possidetis [154]—that is, were
existing boundaries that might be recognised without offending the territorial integrity
of the former Yugoslavia. The secession of Eritrea from Ethiopia in 1993 likewise took
place without violation of the principle of uti possidetis. Eritrea was an Italian colony
from 1890 to 1941, after which it was a trust territory administered by Great Britain
until 1952—when it became part of a federation with Ethiopia, and later a province of
Ethiopia. [155] Thus, its own separate colonial boundaries were clear before its
incorporation into Ethiopia. There is, however, no consistency in respect of secessionist
claims made by entities with previously identifiable boundaries. Somaliland, which
declared its independence from Somalia in 1991, was previously British Somaliland,
and, for a few days in 1960, the independent state of Somaliland. Despite its pre-
existing colonial boundaries, no state has yet recognised Somaliland.
Secessionist claims may also be favourably considered when a people in a distinct
part of a territory have been denied the right to internal self-determination, subjected
to human rights violations, and exhausted all avenues to secure internal self-
determination. Such
Page 157
claims are premised on the paragraph on self-determination in Resolution 2625 (XXV),
which declares that the right of self-determination shall not authorise action which
would impair the territorial integrity of an independent state conducting itself in
compliance with the principles of equal rights and self-determination of peoples ‘and
thus possessed of a government representing the whole people belonging to the
territory without distinction as to race, creed or colour’. Although there is disagreement
about the meaning to be given to this paragraph, [156] there is strong support for the
view that it recognises that a state loses its right to respect for its territorial integrity if
it denies internal self-determination to a people with the result that it is not ‘possessed
of a government representing the whole people belonging to the territory’. The
oppressed people obtain a right to ‘remedial secession’ in such circumstances, provided
they have exhausted all avenues to secure internal self-determination. The secessions
of Bangladesh, Bosnia-Herzegovina and Eritrea can also be justified on this ground.
In 1998 the Supreme Court of Canada gave support to remedial secession
in Reference re Secession of Quebec [157] in which it gave an opinion on the legal right
of Quebec to secede from Canada. The Court found that ‘international law contains
neither a right of unilateral secession nor the explicit denial of such a right’, [158] that
‘the right to self-determination of a people is normally fulfilled through internal self-
determination’, and that the right to external self-determination (secession) ‘arises only
in the most extreme cases’. [159] Applying this to the case of Quebec, the Court held:
[T]he international law right to self-determination only generates, at best, a right to external
self-determination in situations of former colonies; where a people is oppressed, as for
example under foreign military occupation; or where a definable group is denied meaningful
access to government to pursue their political, economic, social and cultural development. In
all three situations, the people in question are entitled to a right to external self-determination
because they have been denied the ability to exert internally their right to self-determination.
Such circumstances are manifestly inapplicable to Quebec under existing conditions.
Accordingly, neither the population
Page 158
of the province of Quebec, even if characterized in terms of ‘people’ or ‘peoples’, nor its
representative institutions, the National Assembly, the legislature or government of Quebec,
possess (sic) a right, under international law, to secede unilaterally from Canada. [160]
The Supreme Court of Canada inferred a right to external self-determination—remedial
secession—in extreme cases of oppression of a ‘people’ from the paragraph on self-
determination in Resolution 2625 (XXV)—the Declaration on Principles of International
Law Concerning Friendly Relations and Co-operation among States in Accordance with
the Charter of the United Nations of 1970—which protects the territorial integrity of a
state only while it conducts itself ‘in compliance with the principles of equal rights and
self-determination of peoples . . . and [is] thus possessed of a government representing
the whole people belonging to the territory without distinction as to race, creed and
colour’.
In the Katangese Peoples’ Congress v Zaire [161] and Gunme v Cameroon [162] the
African Commission on Human and Peoples’ Rights acknowledged the existence of a
right of remedial secession but found that the circumstances did not warrant the
exercise of this right in the two cases in question. Support for remedial secession
premised on Resolution 2625 (XXV) is also to be found in separate opinions in
the Kosovo Advisory Opinion. [163]

(a) Kosovo opinion

The question of secession came before the International Court of Justice in Accordance
with International Law of the Unilateral Declaration of Independence in Respect of
Kosovo. [164] While expectations that the conundrums of secession might be answered
by the Court were not fulfilled, the proceedings before the Court, in which 43 states
participated, made it clear that states accept that secession is not simply a political act
but rather an act governed by international law.
Kosovo, an autonomous province (but not republic) in the Socialist Federal Republic
of Yugoslavia, within the Republic of Serbia, unsuccessfully attempted to assert its
independence from Serbia after the dissolution of the Socialist Federal Republic of
Yugoslavia in the early 1990s. This resulted in widespread human rights violations and
Page 159
ethnic cleansing by the Serbian government in 1998–1999 and culminated in NATO
intervention. After the surrender of Serbia, the United Nations Security Council in
Resolution 1244 (1999) established a civil and military presence in Kosovo, known as
UNMIK, [165] which governed the territory for 10 years. Attempts to find an acceptable
solution failed as Serbia insisted on Kosovo remaining part of Serbia, and Kosovo was
not prepared to compromise on independence. In 2008 Kosovo declared its
independence from Serbia and to date has been recognised by 117 states, including the
United States and most of the member states of the European Union. Membership of
the United Nations is impossible as both the Russian Federation and China oppose
Kosovo’s statehood. (In this respect it resembles Palestine, which has been recognised
by over 138 states but cannot hope to be admitted to the United Nations without the
approval of the United States.)
Later in 2008 the General Assembly, at the instigation of Serbia, requested the
International Court of Justice to give an advisory opinion on the question: ‘Is the
unilateral declaration of independence by the Provisional Institutions of Self-
Government of Kosovo in accordance with international law?’
The question put to the Court was narrow. Unlike the questions put to the Court in
the Namibia [166] and Wall [167] advisory opinions, this question did not ask the Court for
an opinion on the legal consequences of Kosovo’s unilateral declaration of independence
and this allowed the Court to keep its answers within very limited parameters. [168] The
Court found that the unilateral declaration of independence issued by the
representatives of the people of Kosovo did not violate any general rule of international
law [169] or the lex specialis created by Security Council resolution 1244 (1999) [170] and
expressly declared that its Opinion did not address the issues of Kosovo’s
Page 160
statehood or secession, including the notion of remedial secession. [171] The Court took
the position that Kosovo’s declaration of independence was not illegal as it violated no
prohibitory rule of international law and was therefore permitted. [172] (This, despite the
fact that Kosovo’s declaration was not an abstract statement but the beginning of the
process of secession.) The Court did, however, confirm that a unilateral declaration of
independence was illegal if it violated a peremptory norm, as had been the case with
Rhodesia, the Turkish Republic of Northern Cyprus and the Republic of Srpska. [173]
The most significant feature of the Court’s Opinion related to the question of
territorial integrity. State practice, reflected in resolutions of the Security
Council, [174] had hitherto interpreted respect for territorial integrity broadly to prohibit
secessionist actions on the part of disaffected minorities within a state. With scant
examination of the subject, the Court held that ‘the scope of the principle of territorial
integrity is confined to the sphere of relations between states’. [175] This pronouncement
takes no account of the fact that the principle of territorial integrity, as pointed out by
Judge Koroma in his dissenting opinion, [176] has generally been interpreted as
prohibiting non-state actors—that is, would-be secessionist entities—from impairing the
territorial integrity of a state, even if this is done in the exercise of the right of self-
determination. The Court’s dictum may be interpreted as meaning that states alone are
prohibited from engaging in action that may dismember or impair the territorial
integrity of a state and that no such prohibition extends to non-state actors. Such an
interpretation has far-reaching consequences for the law of secession as it suggests
that the principle of territorial integrity is no obstacle to the exercise of self-
determination. At the very least, this matter warranted further consideration on the
part of the Court.

(b) Post-Kosovo developments

The failure of the International Court to provide guidance on the law governing
secession and the relative success of Kosovo in its bid for independent statehood, with
recognition by 117 states despite its non-admission to the United Nations, has
inevitably resulted in demands for statehood by other entities.
 Page 161
In August 2008, Russia and Georgia fought a brief war over South Ossetia which
spread to Abkhazia. Both South Ossetia and Abkhazia had previously formed
autonomous regions in the Soviet Republic of Georgia and after the dissolution of the
Soviet Union had contested Georgian rule. Following the 2008 war, Russia recognised
the two entities as independent states, and this was followed by recognition by a
handful of states, but with the overwhelming majority of states in the international
community withholding recognition. [177]
The will of a people to exercise self-determination is most clearly expressed in a
referendum. In recent times referenda have been held to test public opinion on
secession in Scotland, the Kurdish autonomous region of Iraq, Catalonia and Crimea. In
2014 the referendum on Scottish independence resulted in 55 per cent of the Scottish
electorate rejecting independence, which put an end to demands for Scottish
independence. In the Kurdish autonomous region of Iraq, a referendum held in
September 2017 revealed tremendous support for independence but the intervention of
Iraqi forces shortly afterwards ended this attempt at secession. Catalonian demands for
independence drew support from a 90 per cent vote in favour of independence in a low
turn-out (43 per cent) referendum in October 2017. A later unilateral declaration of
independence was declared illegal by the Spanish central government, which then
detained the leaders of the secession movement. All these attempted secessions have
involved historical and cultural groups with a distinct identity living in a separate
territory claiming a right to self-determination. In none of the societies in question,
however, has the central government denied internal self-determination or suppressed
human rights to such an extent that the right of remedial secession has become a
possibility. This explains why there has been little international sympathy for these
secessionist demands.
The secession of Crimea from Ukraine in 2014 and its subsequent annexation by
Russia is more complicated. [178] Historically, Crimea was part of Russia from 1783 to
1954 when the General Secretary of the Soviet Communist Party, Nikita Krushchev,
awarded it to the Ukraine SSR as a token of friendship. After the dissolution of the
USSR, Crimea became an autonomous region within Ukraine. In 2014, 65 per cent of
the population of Crimea was Russian. In 2014,
Page 162
after Ukraine broke its ties with Russia and became pro-European, Russian troops were
deployed in Crimea, arguably to protect the Russian Black Sea Fleet at Sevastopol. A
referendum was then held in Crimea in which 95 per cent of the voters (with an 83 per
cent turn-out) gave their support to independence from Ukraine and to the annexation
of Crimea by Russia (which followed shortly after). It is clear that the presence of
Russian troops was a factor in this process but whether this interference amounted to
an unlawful use of force is debatable. The Security Council was unable to pronounce on
this annexation by reason of the Russian veto but the General Assembly did condemn it
as a violation of the territorial integrity of Ukraine. [179]
The decision to secede from Ukraine and to seek incorporation within Russia was the
result of a referendum in a territory historically closer to Russia than Ukraine, in which
a majority, consisting mainly of ethnic Russians, voted in favour of secession. On the
other hand, the circumstances did not warrant remedial secession and the Russian
military deployment in Crimea constituted a violation of the territorial integrity of the
Ukraine within the meaning of the concept as defined by the International Court of
Justice in the Kosovo Opinion. [180] It has been argued that Russia’s intervention was
forcible, which means that the annexation of Crimea violated a peremptory norm and is
therefore invalid. However, unlike Israel’s annexation of East Jerusalem, which has
been condemned as illegal by the Security Council, [181] the annexation of Crimea has
been condemned only by the General Assembly. Inevitably, this weakens the case
against Russia.

4 Conclusion

Although, in the words of the Canadian Supreme Court, ‘international law contains
neither a right of unilateral secession nor the explicit denial of such a right’, [182] it is
clear that secession is governed by rules of international law and that these rules guide
states in their decision whether to recognise an entity claiming to be a state. [183] The
right of self-determination and the principle of territorial integrity remain the most
significant principles but both are subject to qualification. It is the task of states and
international institutions to balance these rights, rules and principles wisely to ensure
fairness and stability in the international legal order.
Page 163

5 A South African postscript

Section 235 of the 1996 Constitution provides:

The right of the South African people as a whole to self-determination, as manifested in this
Constitution, does not preclude, within the framework of this right, recognition of the notion of
the right of self-determination of any community sharing a common cultural and language
heritage, within a territorial entity in the Republic or in any other way, determined by national
legislation.
This section should be interpreted in the context of the above rules, practices and
principles. The people of South Africa ‘as a whole’ enjoy the right to self-determination.
Communities within the Republic ‘sharing a common cultural and language heritage’
enjoy a right of internal self-determination, which may by agreement, endorsed by
‘national legislation’, or take the form of some kind of autonomy ‘within a territorial
entity in the Republic’. The phrase ‘or in any other way’ is of uncertain content. Does it
mean that all options—including secession—remain open, provided they are determined
by ‘national legislation’? [184]

Failed states

According to Sir Robert Jennings, a former President of the International Court of

Justice, there is a tendency for international lawyers to ‘keep away from . . . “black
holes” in the system where the basic assumptions, and even the policies of the law,
have parted from reality’. [185] The phenomenon of the juridical state or failed state is
such a black hole.
An existing, recognised state may descend into anarchy and lawlessness to such an
extent that it ceases to meet the requirements of statehood expounded in the
Montevideo Convention. It retains its territory and population but lacks an effective,
central governmental authority. Warlords control parts of the state. But the state itself,
without a central government, is unable to maintain order or to provide the most basic
services for its people. Such a state may be described as a ‘juridical state’ [186] in the
sense that it exists only as an international legal person with no substance to back its
claim to statehood. Alternatively, it may be described as a ‘failed state’.
Page 164
Yemen, at present, provides an example of a failed state but there are others that come
perilously close.
International law cannot be blamed for the breakdown of governmental authority in
a state but it can be blamed for maintaining the appearance of statehood by continuing
to accept such an entity as a state, and for allowing such an entity to continue to
function at the international level through membership in international governmental
organisations. Moreover, international law can be blamed for accommodating the failed
state by relaxing the requirement of effectiveness—effective government—as a
requirement for statehood, and for allowing legality to replace effectiveness. According
to Kreijen:
The failed State merely exists as a normative construction within the texture of international
relations—as a separate legal unit without real substance. The failed State’s independence
exclusively rests on constitutional international principles with a formal-legal character, such as
the right to self-determination and the uti possidetis juris doctrine, as well as on the
recognition of these principles by the international community. As such the failed State’s
existence rests on comity rather than self-achievement. [187]
The failed state poses a challenge to international law. The withdrawal of recognition,
although a logical consequence and a theoretical possibility, is without precedent and is
politically unacceptable. Some form of temporary international trusteeship is another
avenue that deserves consideration. [188] Unfortunately, the notion of trusteeship is
tainted with colonialism and therefore seems an unlikely solution. This leaves informal
‘state-building’ by the international community as the most viable remedy. [189] Neglect
is not a solution as it leaves millions of people without hope for the future and brings
international law itself into disrepute.

[1] Chapter 1.
[2] R Jennings and A Watts (eds) Oppenheim’s International Law 9 ed (1992) vol 1 at 16.
[3] See, in particular, on this subject, J Crawford The Creation of States in International Law 2 ed
(2006).
[4] Oppenheim (n 2) 87–91.
[5] J Dugard Recognition and the United Nations (1987); J Dugard ‘The secession of states and their
recognition in the wake of Kosovo’ (2013) 357 Recueil des Cours 10 at 57–69. (This lecture series is also
published as a pocketbook by the Hague Academy of International Law (2013). References are, however,
to the pagination of the Hague Recueil.)
[6] J Dugard ‘South Africa’s independent homelands: An exercise in denationalization’ (1980)
10 Denver Journal of International Law and Policy 11; GE Norman ‘The Transkei: South Africa’s
illegitimate child’ (1977) 12 New England Law Review 585; MP Vorster, M Wiechers and DJ van Vuuren
(eds) The Constitutions of Transkei, Bophuthatswana, Venda and Ciskei (1985).
[7] The requirements expounded in the Montevideo Convention receive support from Opinion 1 of the
Arbitration Commission established in 1991, under the chairmanship of Mr Badinter of France, to advise
the European Community on legal problems arising from the dissolution of Yugoslavia. The Commission
stated ‘that the state is commonly defined as a community which consists of a territory and a population
subject to an organized political authority; that such a state is characterized by sovereignty’: 92 ILR 162.
The Montevideo requirements for statehood were accepted in Abdi v Minister of Home Affairs 2011 (3) SA
37 (SCA) 51 para 29.
[8] C Ryngaert and S Sobrie ‘Recognition of states: International law or realpolitik? The practice of
recognition in the wake of Kosovo, South Ossetia and Abkhazia’ (2011) 24 Leiden Journal of International
Law 467 at 472.
[9] (1991) 62 BYIL 559; (1992) 41 ICLQ 477.
[10] See J Duursma Fragmentation and the International Relations of Micro-States (1996).
[11] In the North Sea Continental Shelf Cases, the International Court of Justice declared that ‘there is
. . . no rule that the land frontiers of a State must be fully delimited and defined’ (1969 ICJ Reports 3
para 46).
[12] S v Banda 1989 (4) SA 519 (B) at 540E–F.
[13] Ibid at 540G–H.
[14] Hansard, HC, vol 105, col 100 (12 November 1986); vol 126, cols 760–761 (3 February 1988).
[15] C Warbrick ‘Recognition of states’ (1993) 42 ICLQ 433; M Weller ‘The international response to the
dissolution of the Socialist Federal Republic of Yugoslavia’ (1992) 86 AJIL 569.
[16] This explains why the definition of a state, adopted by the Badinter Arbitration Commission (n 7),
emphasises ‘sovereignty’ (independence) rather than ‘capacity to enter into relations with other states’
(Montevideo) as the fourth requirement of statehood.
[17] The Status of the Transkei Act 100 of 1976 declared that ‘the territory known as Transkei . . . is
hereby declared to be a sovereign and independent state and shall cease to be part of the Republic of
South Africa’ (s 1). See above (n 7). Similar statutes were enacted in respect of the other TBVC states.
[18] H Booysen ‘The South African homelands and their capacity to conclude treaties’ (1982)
8 SAYIL 58.
[19] 1989 (4) SA 519 (B) 543C–D. For criticisms of this decision, see R Thomas ‘“hrough the Looking
Glass”—The status of Bophuthatswana in international law’ (1990) 6 SAJHR 65; DJ Devine ‘Banda’s case
1989: International law implications’ (1990) 107 SALJ 434; HA Strydom ‘Vrae rondom die erkenning van
state’ (1992) 3 Stellenbosch LR 67.
[20] S v Banda (n 19) 543G.
[21] The Anarchical Society (1977) 8–9. See also G Erasmus ‘Criteria for determining statehood’ (1988)
4 SAJHR 207 at 212, 220.
[22] 108 ILR 510, 545 (para 182). The decision is discussed at some length by C Warbrick
‘Unrecognized states and liability for income tax’ (1996) 45 ICLQ 954. The Commissioners incorrectly
attributed these views to J Dugard Recognition and the United Nations instead of the present work.
[23] See LM Thompson The Unification of South Africa 1902–1910 (1960).
[24] Although Britain allowed South Africa some autonomy in the negotiation of commercial and
technical treaties, it retained exclusive control over treaties of a political nature. See MM Lewis ‘The
international status of the British self-governing dominions’ (1922–1923) 3 BYIL 21 at 23; RP Schaffer A
Critical Analysis of the Treaty-making Powers of the Union of South Africa and the Republic of South
Africa (PhD thesis, University of the Witwatersrand 1978) 4–8; RB Stewart ‘Treaty-making procedure in
the British dominions’ (1938) 32 AJIL 467.
[25] South Africa did not send or receive diplomatic agents before World War I: Schaffer (n 24) 9–12.
[26] RP Schaffer ‘The prerogative of war and peace: Its development in South Africa’ (1978)
4 SAYIL 29.
[27] 28 & 29 Vict c 63.
[28] LM Friedlander ‘The admission of states to the League of Nations’ (1928) 9 BYIL 84 at 85.
[29] S Pienaar South Africa and International Relations between the Two World Wars: The League of
Nations Dimension (1987).
[30] J Dugard ‘South West Africa and the supremacy of the South African Parliament’ (1969)
86 SALJ 194.
[31] G Carpenter Introduction to South African Constitutional Law (1987) 205.
[32] Pienaar (n 29) 6.
[33] Oppenheim states: ‘In accordance with the Resolutions of 1926, “full powers” to sign the treaty
are issued on the advice of the Dominion Government concerned, not upon the advice of the Government
of the United Kingdom in London. Thus, the exercise of the treaty-making power of the Dominions cannot
now be regarded as a delegation from any central government; it is derived from their own status’ (L
Oppenheim International Law 8 ed (H Lauterpacht (editor)) 1955 vol 1 at 886).
[34] 22 & 23 Geo v c 4.
[35] Act 69 of 1934.
[36] Act 70 of 1934.
[37] Schaffer (n 26).
[38] See Harksen v President of the Republic of South Africa 1998 (2) SA 1011 (C) at 1026–7. Sed
contra, see Z Motala ‘Under international law, does the new order in South Africa assume the obligations
and responsibilities of the apartheid order? An argument for realism over formalism’ (1997)
30 CILSA 287. Motala argues that South Africa only exercised true self-determination in 1994 and thus
achieved statehood in that year.
[39] There were exceptional cases in which human rights issues were considered. In the early 19th
century Britain made it clear to Brazil and Mexico that it viewed the abolition of the slave trade as a
precondition for recognition: HA Smith Great Britain and the Law of Nations (1932) 129–30, 185–7.
Britain’s recognition of Brazil in January 1826 was followed by a treaty providing for the abolition of the
slave trade in November 1826 (197).
[40] On the withdrawal of recognition, see Oppenheim (n 2) 176; H Kelsen ‘Recognition in international
law’ (1941) 35 AJIL 605 at 611.
[41] (1992) 41 ICLQ 477; (1991) 62 BYIL 559.
[42] See M Weller ‘The international response to the dissolution of the Socialist Federal Republic of
Yugoslavia’ (1992) 86 AJIL 569 at 586; Warbrick (n 15).
[43] JES Fawcett The Law of Nations (1968) 38–9.
[44] ‘Security Council resolutions on Rhodesia’ (1964–1965) 41 BYIL 102 at 112. See the response to
DJ Devine’s article on ‘The requirements of statehood re-examined’ by Fawcett in (1971) 34 MLR 417.
[45] TM Franck Fairness in International Law and Institutions (1995); JR Crawford ‘Democracy and
international law’ (1993) 64 BYIL 113; S Murphy ‘Democratic legitimacy and the recognition of states and
governments’ (1999) 48 ICLQ 545.
[46] CN Okeke Controversial Subjects of Contemporary International Law (1974) 88.
[47] See DJ Devine ‘The requirements of statehood re-examined’ (1971) 34 MLR 410; ‘The status of
Rhodesia in international law’ 1973 Acta Juridica 1 at 84–6.
[48] The Opinions of the Arbitration Commission, under the chairmanship of Mr Badinter, may be found
in 92 ILR 162–206.
[49] See Warbrick (n 15); Weller (n 42).
[50] See Accordance with International Law of the Unilateral Declaration of Independence in Respect of
Kosovo 2010 ICJ Reports 403, 437–8 (para 81). See further Dugard Recognition and the United
Nations (n 5) 86–111.
[51] The two leading treatises on this subject are H Lauterpacht Recognition in International
Law (1947), re-issued in the Grotius Classic Reprint Series with foreword by James Crawford (2012); T
Chen The International Law of Recognition (1951). See further Murphy (n 45); TD Grant The Recognition
of States: Law and Practice in Debate and Evolution (1999); S Talmon Recognition in International Law: A
Bibliography (1998); M Fabry Recognizing States: International Society and the Establishment of New
States since 1776 (2010); Ryngaert and Sobrie (n 8).
[52] For accounts of this debate and the allegiances of different jurists, see Crawford (n 3) 19–28,
Lauterpacht (n 51) ch 4; Chen (n 51) chs 1 and 2; Devine ‘The status of Rhodesia in international law’ (n
47) 90–145; D Raic Statehood and the Law of Self-Determination (2002) 29–38.
[53] For a strong exposition of this view, see H Kelsen ‘Recognition in international law: Theoretical
observations’ (1941) 35 AJIL 605.
[54] JL Brierly Law of Nations 6 ed (ed H Waldock, 1963) 139.
[55] Lauterpacht (n 51) 78.
[56] Lauterpacht (n 51) 6.
[57] In Madzimbamuto v Lardner-Burke NO 1968 (2) SA 284 (RA), Beadle CJ stated (319F): ‘Few
nations (and certainly not the United Kingdom) apply the idealistic “Lauterpacht theory” of recognition, a
theory which presupposes that recognition must always depend on an objective legal appraisal of the true
facts. Political considerations are frequently the overriding ones and they, too often, depend on no
principle other than political expediency.’
[58] The United States did, however, later acknowledge that it had recognised Panama prematurely
and paid $25 million to Colombia as reparation.
[59] See Weller (n 42).
[60] The Badinter Arbitration Commission, charged with the task of monitoring compliance with the
European Community’s guidelines for the recognition of states following the dissolution of Yugoslavia,
found that ‘the existence or disappearance of the state is a question of fact; that the effects of recognition
by other states are purely declaratory’: Opinion 1 in 92 ILR 162.
[61] In S v Oosthuizen 1977 (1) SA 823 (N) the Court found that the fact that Rhodesia had not been
recognised internationally did not mean it was not a state (825A). For a criticism of this decision, see J
Dugard ‘Rhodesia: Does South Africa recognise it as an independent state?’ (1977) 94 SALJ 127.
[62] 1989 (4) SA 519 (B) 531–9.
[63] At 531E, 533–7.
[64] At 531F, 538–9. Although Friedman J is correct in this conclusion, the evidence he adduces does
not support his conclusion. Five of the six examples he cites relate to the recognition of governments and
not states (532–3).
[65] DJ Devine was prepared to accept that as partially recognised states, the TBVC states had some
international personality, although it was ‘relative and weak’: ‘Recognition, newly independent states and
general international law’ (1984) 10 SAYIL 18.
[66] See Caglar v Billingham above (n 22). Sed contra, see MN Shaw International Law 8 ed (2017),
who states that if an entity were totally unrecognised, ‘this would undoubtedly hamper the exercise of its
rights and duties . . . but it would not seem in law to amount to a decisive argument against statehood
itself’ (332).
[67] See Crawford (n 3) at 93.
[68] Oppenheim (n 2) 169.
[69] J Dugard (1969) 86 SALJ 113–14; AJGM Sanders ‘Die erkenning van state en regerings’ (1970)
33 THRHR 259 at 264.
[70] See Dugard (n 61); Devine ‘The status of Rhodesia in international law’ (n 47) and ‘Does South
Africa recognise Rhodesian independence?’ (1969) 86 SALJ 438.
[71] See, further, on this subject Dugard Recognition and the United Nations (n 5) and ‘The secession
of states and their recognition’ (n 5) 57–69.
[72] For example, in the cases of Croatia, Slovenia, Bosnia and Herzegovina and Macedonia. See Weller
(n 42); Warbrick (n 15).
[73] Articles 2(4)–(7), 11(2), 32, 35(2), 43(3), 50, 52(3), 53(1)–(2), 59, 79, 80(1), 81, 93(2), 107,
110(1)–(4).
[74] Article 73(e).
[75] Article 76(b).
[76] Articles 1(2) and 55.
[77] The nine abstaining states were Australia, Belgium, the Dominican Republic, France, Portugal,
South Africa, Spain, the United Kingdom and the United States.
[78] The same options are recognised by Resolution 2625 (XXV) of 1970: the Declaration on Principles
of International Law Concerning Friendly Relations and Co-operation among States in Accordance with the
Charter of the United Nations.
[79] See WV O’Brien and UH Goebel ‘United States recognition policy toward the new nations’ in WV
O’Brien (ed) The New Nations in International Law and Diplomacy (1965) 98 at 212, 223; DP Myers
‘Contemporary practice of the United States relating to international law’ (1961) 55 AJIL 697 at 717; G
Kreijen State Failure, Sovereignty and Effectiveness: Legal Lessons from the Decolonization of Sub-
Saharan Africa (2004), particularly at 49–52.
[80] For example, in 1991, the Federated States of Micronesia and the Republic of the Marshall Islands
were admitted to the United Nations despite their strong dependency on the United States.
[81] Christian Hillgruber states that ‘[o]n admission as a member of the United Nations, the new state
then becomes part of the globally organized community of states by way of co-optation’ (’The admission
of new states to the international community’ (1998) 9 European Journal of International Law 491 at
492).
[82] Lauterpacht (n 51) 78.
[83] Note 51 at 222.
[84] See Dugard Recognition and the United Nations (n 5); ‘The secession of states and their
recognition’ (n 5) 69–72.
[85] Op cit (n 51) 420.
[86] Although the concept of jus cogens is expressly recognised by the Vienna Convention on the Law
of Treaties, which provides that a treaty that violates jus cogens is void (article 53), it is not one that
should be confined to treaties. See Friedman J in S v Banda 1989 (4) SA 519 (B) 544F.
[87] See the Namibia Opinion 1971 ICJ Reports 16 at 54.
[88] Report of the International Law Commission, General Assembly Official Records, 56th Session,
supplement 10 (A/56/10) (2001).
[89] 2010 ICJ Reports at 438 (para 81).
[90] Resolution 662 (1990).
[91] Resolutions 541 (1983) and 550 (1984).
[92] Security Council resolutions 478 (1980) and 497 (1981).
[93] Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory 2004 ICJ
Reports para 163D.
[94] Articles 55 and 56.
[95] International Convention on the Elimination of All Forms of Racial Discrimination (1966), and the
International Covenant on Civil and Political Rights (1966) (article 2).
[96] Barcelona Traction, Light and Power Company Case 1970 ICJ Reports 3 at 32.
[97] Dugard Recognition and the United Nations (n 5) 96–7.
[98] Ibid 86–90.
[99] Ibid 90–8.
[100] Ibid 121.
[101] See the commentary on draft article 41 in the ILC Report on the Responsibility of States (n 88)
289 (para 8).
[102] Resolution 31/6A (1976).
[103] Resolution 402 (1976) and 407 (1997).
[104] General Assembly Resolution 32/105 N (1977).
[105] Statement by Security Council contained in S/13549, 21 September 1979; General Assembly
Resolution 34/93 G (1979).
[106] Statement by Security Council S/14794, 15 December 1981.
[107] General Assembly Resolution 1514 (XV) of 1960.
[108] The need for such agreements to return to South Africa was initially advocated by the National
Party government. See M Wiechers ‘Re-incorporation of the TBVC Countries—International law, practice
and constitutional implications’ (1990–91) 16 SAYIL 119. The present writer advocated the simple repeal
of the four South African statutes conferring ‘independence’ on the Bantustan states: ‘Failure of the TBVC
states’ (1992) 8 SAJHR at v (editorial comment). This was the course followed by the Interim Constitution
of 1993. See sched 7 of Act 200 of 1993.
[109] R McCorquodale (ed) Self-Determination in International Law (2003); K Knop Diversity and Self-
Determination in International Law (2002); A Cassese Self-Determination of Peoples: A Legal
Appraisal (1995); C Tomuschat (ed) Modern Law of Self-Determination (1993); D Raic (n 52); A
Buchanan Justice, Legitimacy and Self-Determination (2004).
[110] C Walter, A von Ungern-Sternberg and K Abushov Self-Determination and Secession in
International Law (2014); M Kohen (ed) Secession: International Law Perspectives (2006); Crawford (n
3) ch 9.
[111] Articles 1, 55.
[112] See in particular Resolution 2625 (XXV) of 1970.
[113] Separate opinion of Judge Dillard in the Western Sahara Case 1975 ICJ Reports 12 at 121–2.
[114] 1971 ICJ Reports 16, 31.
[115] 1975 ICJ Reports 12, 31–3.
[116] Article 1(1).
[117] Crawford (n 3) 108–28; Cassese (n 109); R McCorquodale ‘South Africa and the right of self-
determination’ (1994) 10 SAJHR 4.
[118] 1995 ICJ Reports 90 at 102.
[119] 2004 ICJ Reports 136 para 155.
[120] 2017 (5) SA 105 (ECP) 116 B–C.
[121] See LC Buchheit Secession: The Legitimacy of Self-Determination (1978) 7.
[122] Resolution 1514 (XV).
[123] Resolution 2625 (XXV).
[124] See the Namibia Opinion 1971 ICJ Reports 31; and the Western Sahara Opinion 1975 ICJ Reports
31–3 and 121 (separate opinion of Judge Dillard).
[125] Article 1(1) in both Covenants.
[126] Vienna Declaration on Human Rights (1993); General Assembly resolution 48/121 of 20
December 1993.
[127] Reference re Secession of Quebec (1998) 2 SCR 217 para 138, (1998) 37 ILM 1373.
[128] Dugard ‘The secession of states and their recognition’ (n 5) 84–5.
[129] Ibid 85–8.
[130] R Higgins Problems and Process: International Law and How We Use It (1994) 124.
[131] See J Crawford (ed) The Rights of Peoples (1988); R Kiwanuka ‘The meaning of “people” in the
African Charter of Human and Peoples Rights’ (1988) 82 AJIL 80.
[132] Higgins (n 130) 124.
[133] (1998) 37 ILM 1340 at 1370 (para 124).
[134] Katanganese Peoples’ Congress v Zaire Communication 75/92, African Commission on Human
and Peoples’ Rights (1995) 8th Annual Activity Report; (2000) African Human Rights Law Reports
72. Kevin Mgwanga Gunme v Cameroon Communication 266/2003, African Commission on Human and
Peoples’ Rights (2008–2009) 26th Annual Activity Report. F Viljoen International Human Rights Law in
Africa (2007) 245; D Shelton ‘Self-determination in regional human rights law: From Kosovo to
Cameroon’ (2011) 105 AJIL 60 at 67, 69.
[135] Dugard ‘The secession of states and their recognition’ (n 5) 97.
[136] Dugard ‘The secession of states and their recognition’ (n5) 98.
[137] Article 3(b).
[138] This provision is endorsed by the Vienna Declaration and Programme of Action (para 2) adopted
by the UN World Conference on Human Rights of 1993: (1993) 32 ILM 1661, 1663.
[139] General Assembly resolution 60/L.1 of 15 September 2005 para 1(5).
[140] In 1960–1961 the Security Council invoked this principle to block the secession of Katanga from
the newly independent Congo: Security Council resolution S/5002 of 24 November 1961. See, further,
Dugard ‘The secession of states and their recognition’ (n 5) 110.
[141] Security Council resolutions 1766 (2007) and 1772 (2007).
[142] J Dugard and D Raic ‘The role of recognition in the law and practice of secession’ in M Kohen
(ed) Secession: International Law Perspectives (2006) 94; Ryngaert and Sobrie (n 8).
[143] (1979) UN Monthly Chronicle (February) 40.
[144] See advisory opinion of the International Court of Justice in Accordance with International Law of
the Unilateral Declaration of Independence in Respect of Kosovo 2010 ICJ Reports para 81.
[145] Security Council resolution S/5002 (1961).
[146] Security Council resolution 277 (1970).
[147] General Assembly resolutions 31/6A (Transkei), 32/105N (Bophuthatswana), 34/93G (Venda)
and 35/69A (Ciskei); Security Council resolutions 402 (1976), 407 (1977).
[148] Security Council resolutions 541 (1983) and 550 (1984).
[149] Security Council resolution 787 (1992).
[150] DA Ijalaye ‘Was Biafra at any time a state in international law?’ (1971) 65 AJIL 551.
[151] Crawford The Creation of States in International Law 2 ed (2006) 434; Dugard ‘The secession of
states and their recognition’ (n 5) 142.
[152] See Dugard and Raic (n 142) 120–3.
[153] The author prefers to see the dismemberment of Yugoslavia as a case of secession: ibid 123–32.
[154] Opinion 3, 92 ILR 170. See further M Craven ‘The EC Arbitration Commission on Yugoslavia’
(1995) 66 BYIL 333. See further on uti possidetis, Chapter 8.
[155] See Cassese (n 109) 218–22.
[156] See the disagreement on this issue among contributors to M Kohen (ed) Secession: International
Law Perspectives (2006) 10. See further, Cassese (n 109) 118–20; J Dugard ‘Secession: Is the case of
Yugoslavia a precedent for Africa?’ (1993) 5 African Journal of International and Comparative Law 163; L
Brilmayer ‘Secession and self-determination: A territorial interpretation’ (1991) 16 Yale Journal of
International Law 177; L Buchheit Secession: The Legitimacy of Self-Determination (1978); Buchanan (n
109) ch 8.
[157] (1998) 37 ILM 1340.
[158] Ibid 1369 (para 112).
[159] Ibid 1371 (para 126).
[160] Secession of Quebec (n 157) 1373 (para 138).
[161] Note 134; F Viljoen International Human Rights Law in Africa (2007) 244–5.
[162] Note 134.
[163] Separate opinions of Judges Concado (paras 177–81) and Yusuf (paras 9–17) Accordance with
International Law of the Unilateral Declaration of Independence in Respect of Kosovo 2010 ICJ Reports
403.
[164] 2010 ICJ Reports 403. There is a wealth of literature on this opinion which can be found in the
bibliography attached to Dugard ‘The secession of states and their recognition’ (n 5).
[165] United Nations Interim Administration Mission in Kosovo.
[166] Legal Consequences for States of the Continued Presence of South Africa in Namibia (South West
Africa) notwithstanding Security Council Resolution 276 (1970) 1971 ICJ Reports 16.
[167] Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory 2004 ICJ
Reports 136.
[168] Accordance with International Law of the Unilateral Declaration of Independence in Respect of
Kosovo 2010 ICJ Reports 403, 423 para 51. Cf the criticism of the Court for its failure to adopt a broader
approach by Judge Simma in his separate opinion: 2010 ICJ Reports 480. See further on the Court’s
opinion, R Falk ‘Agora: The ICJ’s Kosovo Advisory Opinion’ (2011) 105 AJIL 50; ‘Kosovo symposium’
(2011) 24 Leiden Journal of International Law 71–161, 331–83, 467–90.
[169] Accordance with International Law of the Unilateral Declaration of Independence in Respect of
Kosovo 2010 ICJ Reports 403 at 436–9 paras 79, 84.
[170] Ibid 451–2 paras 118–19.
[171] Ibid 423 para 51; 438–9 paras 82–3.
[172] Ibid paras 56, 79, 84.
[173] Ibid 437–8 para 81.
[174] See above at notes 140, 141.
[175] Kosovo Opinion (n 169) 437 para 80.
[176] Ibid 468, 475–476. See further J Dugard ‘Abdul Koroma, territorial integrity and
the Kosovo Opinion’ in CC Jalloh and O Elias Shielding Humanity: Essays in International Law in Honour of
Judge Abdul G Koroma (2015) 50.
[177] Ryngaert and Sobrie (n 8) 472.
[178] See SF van den Driest ‘Crimea’s separation from Ukraine: An analysis of the right to self-
determination and (remedial) secession in international law’ (2015) 62 Netherlands International Law
Review 329; J Vidmar ‘The annexation of Crimea and the boundaries of the will of the people’ (2015)
16 German Law Journal 365; C Walter ‘Postscript: Self-determination, secession and the Crimean crisis’
in Walter et al (n 110) 293.
[179] Resolution 68/262 of 27 March 2014.
[180] See above at n 175.
[181] See resolution 478 (1980).
[182] Reference Re Secession of Quebec (1998) 37 ILM 1340 at 1369 (para 112).
[183] Dugard ‘The secession of states and their recognition’ (n 5) 33, 203–4, 214.
[184] On the subject of self-determination in the South African context, see McCorquodale (n 117); HA
Strydom ‘Self-determination, and the South African Interim Constitution’ (1993/4) 19 SAYIL 43.
[185] Foreword to G Kreijen State Failure, Sovereignty and Effectiveness (2004) viii.
[186] RH Jackson Quasi-States: Sovereignty, International Relations and the Third World (1990). See
further, G Barrie ‘Failed states: The new challenge to international law’ (2015) 40 SAYIL 103.
[187] Kreijen (n 185) 375.
[188] Ibid 308–29.
[189] Crawford (n 3) 719–23.
 Page 165

Chapter 6
Recognition of Governments

John Dugard
Once an entity becomes a state, it acquires international personality and participates in
the affairs of the international community. This participation is conducted by the
government of the state, which inevitably will change from time to time, either by
democratic means or by revolution. Changes in government do not affect the
personality of the state or its rights and obligations. The new government succeeds to
the rights and obligations of its predecessor, however it came into existence. [1]
When the change in government is constitutional, no problem arises. Other states
continue doing business with the new government and diplomatic relations are
unaffected. When the change is unconstitutional, problems arise. The new revolutionary
government may have uncertain control over its territory, or it may be unwilling to
comply with its international obligations, or it may be challenged by a rival claimant, or
it may be controlled by a foreign power, or it may have a poor human rights record, or
it may adhere to a different ideology. In these circumstances, other states will have to
decide whether the new government really represents the state or whether it is the
type of government with which they wish to do business.
Traditionally, the decision is conveyed to the new government by recognition.
According to Oppenheim:
A government which is in fact in control of the country and which enjoys the habitual
obedience of the bulk of the population with a reasonable expectancy of permanence, can be
said to represent the state in question and as such to be deserving of recognition. [2]
As with the recognition of states, this judgment is left to the governments of other
states, which in many instances will be guided by political considerations. For example,
the United States refused to recognise the Soviet government of Russia from 1920 to
1933 and the communist government of China from 1948 to 1979.
Recognition by the major powers plays an important legitimating role. Recognition
by states such as the United States, the United Kingdom or France will boost the
legitimacy of a revolutionary
Page 166
government, while the withholding of recognition by these powers may have serious
consequences, both political and economic—hence the importance attached to
recognition by the major powers on the part of unconstitutional governments.
The new concern for human rights has created difficulties for the major powers in
respect of recognition. Although such a power may recognise a government purely on
the ground that it has effective control and appears to be permanent, recognition will
inevitably be construed as a form of political approval. Where the newly recognised
government has a poor human rights record, such apparent approval may outrage both
domestic and international opinion. For this reason, France, [3] the United
States, [4] Britain and several other states [5] have officially announced that they will no
longer accord recognition to governments (as opposed to states). The British statement
of 1980 declares:
[W]e have decided we shall no longer accord recognition to governments. . . .
Where an unconstitutional change of regime takes place in a recognised state, governments
of other states must necessarily consider what dealings, if any, they should have with the new
regime, and whether and to what extent it qualifies to be treated as the government of the
state concerned. . . . [T]he policy of successive British governments has been that we should
make and announce a decision formally ‘recognising’ the new government. . . . This practice
has sometimes been misunderstood, and, despite explanations to the contrary, our
‘recognition’ interpreted as implying approval. For example, in circumstances where there
might be legitimate public concern about the violation of human rights by the new regime, or
the manner in which it achieved power, it has not sufficed to say that an announcement of
‘recognition’ is simply a neutral formality.
We have therefore concluded that there are practical advantages in following the policy of
many other countries in not according recognition to governments. Like them, we shall
continue to decide the nature of our dealings with regimes which come to power
unconstitutionally in the light of our assessment of whether they are able of themselves to
exercise effective control of the territory of the state concerned, and seem likely to continue to
do so. [6]
Page 167
Today it seems that the majority of states adopt this approach. [7]

South Africa’s recognition policy is in line with this trend. As a pariah state during the
apartheid years, its recognition was not sought after for purposes of approval in the
way that claimant governments turned to Britain and the United States for recognition.
Only within the southern African region was recognition by South Africa considered to
be important. The consistent policy of the South African government during the
apartheid era was to adopt a low profile towards revolutionary regimes and to continue
diplomatic and trade relations once it was established that the regime had effective
control over the territory and seemed likely to continue to do so with some degree of
permanency. Explicit statements on recognition were avoided and every effort was
made to depoliticise a highly political subject.
The South African approach of this period is illustrated by the response to military
coups in Lesotho, Transkei, Ciskei and Venda.
On 30 January 1970, Chief Leabua Jonathan, the Prime Minister of Lesotho, seized
power unconstitutionally when he declared a parliamentary election to be invalid and
suspended the constitution. [8] Although Britain granted recognition to the new regime
only on 12 June, Prime Minister Vorster indicated South Africa’s decision to continue
normal relations with Lesotho on 6 February in a parliamentary exchange:
The Prime Minister: Now this question has arisen, ie what is our relationship with Lesotho? I
want to tell you, Sir, that our relationship with Lesotho is exactly the same as it is with
Rhodesia. We are simply carrying on as though nothing has happened. Lesotho is a
neighbouring state of ours. We do not interfere with them. We do not dictate to them. We have
no hand in their decisions. But it is a neighbouring state with which we have to negotiate, and
as far as we are concerned, Chief Jonathan is in effective control of that neighbouring state
and we shall continue our negotiations with him as if in fact nothing has happened in Lesotho.
Sir de Villiers Graaff: That is official recognition.
The Prime Minister: Official recognition or not. Official recognition simply plays no part, just
as little as it did in regard to Rhodesia. It is merely a question of our continuing as if nothing
has happened. We have our specific relations with them and we are continuing with those
specific relations. [9]
When Chief Leabua Jonathan was himself overthrown in a bloodless coup by Major-
General JM Lekhanya in 1986, there was no explicit statement on recognition by the
South African government. Instead, the Department of Foreign Affairs issued a
statement to the
Page 168
effect that the Foreign Minister, Mr RF Botha, and a special emissary from the Military
Council of Lesotho had agreed on the ‘need to work actively for the promotion of good
neighbourliness’. [10]
On 30 December 1987, Major-General Bantu Holomisa took over the government of
Transkei in a military coup. On 21 January 1988, State President PW Botha stated that
the South African government was satisfied that the new administration was in effective
control of all functions of government. [11] Later, the State President declared:
The South African government recognises Transkei as a sovereign independent state and, as it
is in the interests of both countries and their peoples to continue the co-operation which has
traditionally existed between them, will in future conduct its bilateral relations with the new
government. [12]
Military coups in Ciskei [13] and Venda [14] in 1990 were similarly treated.
That the need for continuity and stability continues to guide foreign policy on
recognition of governments is clear from South Africa’s early recognition of the Kabila
government of the Democratic Republic of Congo (formerly Zaire) after the overthrow
of the Mobutu regime in 1997. Commenting on this, the Deputy Minister of Foreign
Affairs, Mr Aziz Pahad, stated:
The immediate recognition by South Africa of President Kabila and his government after their
military forces took over control of the country appears to have created the impression—
especially in the news media—that the South African Government is somehow acting as
sponsor of or, ‘godfather’ to, the Kabila government.
That is not the case at all. The early recognition of the Kabila government was based on the
realisation that a power vacuum in Kinshasa would generate chaos and further contribute to
the economic collapse of the country.
It was seen as vital that peace and stability be restored to the former Zaire at the earliest
possible moment in order to commence the enormous task of bringing about the economic
reconstruction of the country. [15]
It would be incorrect, according to Mr Pahad, to infer approval of a government’s
internal policies from recognition. [16]
Page 169
Although the government’s policy is to adopt a low-profile approach towards
recognition of governments, [17] resembling that of Britain and the United States, this
was not possible when South Africa switched recognition from the government of the
Republic of China (Taiwan) to the government of the People’s Republic of China
(Beijing) in 1998. This was, however, an unusual case made necessary by the need for
post-apartheid South Africa to recognise the People’s Republic of China as the lawful
government of China after it had recognised the government of Taiwan as the
government of China for 30 years. Taiwan is not an independent state. However, the
government of Taiwan competes with the government of the People’s Republic of China
in Beijing for recognition as the government of the whole of China. Most states today
recognise the Beijing government as the lawful government of China, but a minority,
mainly in Central America, persist in recognising Taiwan as the government of China.
This was South Africa’s position before it recognised the Beijing government in
1998. [18]
 The policy of the South African government to support continuity and stability in its
recognition of new governments was apparent in its recognition of the new government
of Zimbabwe when Robert Mugabe was replaced by Emmerson Mnangagwa.

De Facto and De Jure recognition

When a new government assumes effective control of a territory with a likely prospect
of permanency, it should be given full recognition as a de jure government. However,
there may be circumstances that warrant some lesser form of recognition; in such a
case, the new government will be recognised as a de facto government. Such a
distinction applies only in the case of recognition of governments. A state cannot be
recognised de facto.
It has been suggested that de facto recognition is appropriate where the government
is not firmly established [19] or fails to show a
Page 170
willingness to comply with its international obligations. [20] As de facto recognition
implies a measure of disapproval, it normally does not include an exchange of
diplomats.
De facto recognition has been accorded, in practice, in two situations:
(a)
It has been given when the recognising state has doubts about both the stability
and the ideology of the new government. Thus, Britain recognised the Soviet
government de facto in 1921 and de jure in 1924.
(b)
It has also been accorded when there are two rival governments competing for
power. In the Spanish Civil War (1936–1939) Britain continued to recognise the
Republican government as the de jure government but granted de
facto recognition to General Franco as he extended his control over the country.
The same practice was followed in respect of Italy’s conquest of Ethiopia: in 1936
Italy was recognised as the de facto government of Ethiopia while Emperor Haile
Selassie remained the de jure government until 1938. [21]
Today, the distinction between de facto and de jure recognition is largely discredited
and does not seem to have any legal significance. [22]
The present practice of states seems to be to withhold recognition completely where
there are doubts about the stability of a new government or where there are two rival
powers competing for the government of a country. For instance, during the civil war in
Angola between 1975 and 1991, South Africa recognised the state of Angola but did not
recognise either the MPLA [23] or UNITA as the de jure or de facto government of
Angola.

[1] DP O’Connell International Law 2 ed (1970) vol 1 at 394.

[2] R Jennings and A Watts (eds) Oppenheim’s International Law 9 ed (1992) vol 1 at 150. See further
S Talmon Recognition of Governments in International Law (1998).
[3] MJ Peterson ‘Recognition of governments should not be abolished’ (1983) 77 AJIL 31 at 42–3.
[4] (1977) 77 US Department of State Bulletin 462; reproduced in DJ Harris and S Sivakumaran Cases
and Materials on International Law 8 ed (2015). See further LT Galloway Recognizing Foreign
Governments: The Practice of the United States (1978). Cf SD Murphy ‘Democratic legitimacy and the
recognition of states and governments’ (1999) 48 ICLQ 544 at 566.
 [5] In 1988, Australia issued a statement similar to that of Britain—see (1992) 12 Australian Year Book
of International Law 357.
[6] Hansard HL vol 408 cols 1121–2. See, too, S Talmon ‘Recognition of governments: An analysis of
the new British policy and practice’ (1992) 63 BYIL 231.
[7] See Galloway (n 4). For a criticism of this trend, see Peterson (n 3).
[8] AJGM Sanders ‘Die erkenning van state en regerings’ (1970) 33 THRHR 259.
[9] House of Assembly Debates vol 26 col 454 (6 February 1970).
[10] ME Muller ‘South Africa’s foreign policy and international practice as reflected in speeches,
parliamentary statements and replies—1986’ (1986–1987) 12 SAYIL 228 (‘South Africa’s Foreign Policy’).
[11] 1988/89 Race Relations Survey 131.
[12] The Star 11 February 1988, quoted in M Beukes ‘The coup d’etat: “An annual African event”—A
South African reading’ (1987–1988) 13 SAYIL 133.
[13] The Star 7 June 1990; H Botha ‘South Africa’s foreign policy—1990’ (1990/91) 16 SAYIL 227.
[14] The Star 7 June 1990.
[15] The Star 19 June 1998 at 9.
[16] Ibid. Mr Pahad stressed that recognition of the Kabila government was ‘not in any way
inconsistent’ with the government’s ‘commitment to the advancement of human rights in our continent’
and that the government would continue to call for the holding of free and fair elections in the Democratic
Republic of Congo. When South Africa entered into diplomatic relations with Iraq, Mr Pahad emphasised
that this did not mean that South Africa agreed with its internal policies (Pretoria News 6 July 1998).
[17] In an interview with the Pretoria News on 6 July 1998, Mr Pahad declared ‘you recognise the
country but not the government’.
[18] For a discussion of this switch in recognition, see R Henwood ‘South Africa’s foreign policy—1995–
6’ (1996) 21 SAYIL 256 and R Henwood ‘South Africa’s foreign policy—1997’ (1997) 22 SAYIL 151.
[19] See the judgment of Beadle CJ in Madzimbamuto v Lardner-Burke NO 1968 (2) SA 284 (RA) 359.
In this case, the terms ‘de facto’ and ‘de jure’ government were invoked in the municipal law context to
assess the legal status of the Rhodesian government after 1965. On appeal, the Privy Council indicated
that they are terms best confined to international law: Madzimbamuto v Lardner-Burke [1968] 3 WLR
1229 (PC) 1249. Cf Matanzima v President of the Republic of Transkei 1989 (4) SA 989 (Tk) 998B.
[20] Oppenheim (n 2) 156.
[21] O’Connell (n 1) 161.
[22] See AJGM Sanders ‘The courts and recognition of foreign states and governments’ (1975)
92 SALJ 165 at 171.
[23] House of Assembly Debates vol 60 cols 112–13 (27 January 1976). In 1985 Mr RF Botha, South
Africa’s Foreign Minister, stated: ‘The government of the Republic of South Africa recognises the People’s
Republic of Angola as a state but does not recognise the government of President Eduardo Dos Santos as
the de jure or de facto government of the People’s Republic of Angola’ (certificate issued on 1 May 1985
in Banco Comercial de Angola v Banco Popular de Angola WLD; unreported, as the matter was settled out
of court).

Page 171

Chapter 7
Recognition in Municipal Law

John Dugard
Only a recognised state or government enjoys locus standi in a South African court;
only a recognised state or government, or its agents, may plead immunity from the
jurisdiction of a South African court; and only the legislative, executive or judicial acts
of a recognised state or government will be given legal effect by a South African
court. [1] Consequently, when such questions arise, it will be necessary for a municipal
court to decide whether the entity has been recognised.
Two questions present problems: first, how is recognition to be proved? Secondly, is
the rule that no effect should be given to the acts of an unrecognised state or
government absolute? On both these issues South African courts are guided by English
law. [2]

Proof of recognition

The recognition of a state or government is a matter that falls within the non-statutory
powers of the executive. [3] Consequently, where there is any doubt about the status of
a state or government, the court should request the Department of International
Relations and Co-operation (DIRCO) for a certificate setting out its views on the
subject. [4] If this course is not followed, a situation may arise in which the courts and
the executive express different views on matters of foreign policy, which may
embarrass the executive in its conduct of foreign affairs.
Our courts have accepted that such an executive certificate is ‘conclusive on the
matter of recognition’, [5] although this may no
Page 172
longer be the position under our new constitutional rules. [6] Similarly, the Foreign
States Immunities Act provides that a certificate from the Minister of International
Relations and Co-operation shall be conclusive evidence on the question ‘whether any
foreign country is a state’ for the purposes of the Act. [7]
Where the question of recognition is ‘a matter of judicial cognizance’, no certificate is
necessary. [8] However, our courts, either out of disregard for the executive’s interests
or out of ignorance, have accepted entities as states or governments in a number of
highly controversial instances. In 1971 a court took judicial notice of the fact that the
Congolese government was the ‘government of a foreign sovereign state recognised as
such by the South African government’, [9] despite the fact that there were no
diplomatic relations between the two countries. In 1975 the Appellate Division [10] gave
effect to the laws of East Germany in a matrimonial dispute without any enquiry into
the question whether South Africa recognised East Germany as an independent state.
As South Africa had no diplomatic relations with East Germany and the two states were
ideologically hostile to each other, it seems more than likely that the executive would
have indicated that it did not recognise East Germany. [11] In 1977 the Natal Provincial
Division went still further when it pronounced on an issue that the executive had
carefully and studiously avoided pronouncing upon for over a decade [12]—Rhodesian
independence. In S v Oosthuizen, [13] Kriek J took judicial notice of the fact that
Rhodesia became a new state in 1965, despite the absence of recognition by states,
including South Africa. He accordingly held that an extradition agreement entered into
between South Africa and Southern Rhodesia in 1963 had lapsed on the ground that
the international personality of Southern Rhodesia had changed. [14]
Page 173
In all of the above cases, particularly S v Oosthuizen, there was sufficient doubt as
to the status of the state or government in question to have warranted a request by the
Court for an executive certificate. If such a certificate was necessary to establish that
South Africa recognised Mozambique as an independent state and Frelimo as the
government of Mozambique in 1979, as held by Margo J in Inter-Science Research v
Republica Popular de Mocambique, [15] a fortiori it was necessary to obtain an executive
certificate on the status of Rhodesia. Judicial notice of recognition should be confined to
states and governments whose status is a matter of common knowledge—such as the
United States, Britain, India, Japan, etc—and not extended to controversial entities, as
occurred in S v Oosthuizen. Pleas for greater judicial discretion on the subject of
recognition are misplaced if they suggest that the judiciary should make findings on
recognition with no attempt to first ascertain the view of the executive. [16] Recognition
is a highly sensitive subject in international relations and must be left, at least in the
first instance, to the body charged with the task of conducting South Africa’s foreign
relations—the executive. As Lord Atkin warned in Government of the Republic of Spain
v SS ‘Arantzazu Mendi’: [17]
Our state cannot speak with two voices on such a matter, the judiciary saying one thing, the
executive another. Our sovereign has to decide whom he will recognise as a fellow sovereign in
the family of states; and the relations of the foreign states with ours in the matter of state
immunities must flow from that decision alone.
A court is not excluded from pronouncing on the recognition of a state or government.
As suggested in Chapter 4, [18] the 1996 Constitution may permit review of the
executive’s decision, as reflected in an executive certificate. However, it is submitted
that, bearing in mind the words of Lord Atkin in SS Arantzazu Mendi, [19] this should be
done sparingly and with great circumspection. A court may also intervene when the
executive submits a certificate that is ambiguous or non-committal. Indeed, it may
treat such a certificate as an invitation to intervene. That unclear certificates are not
unusual is demonstrated by the certificates submitted to court by the British
government since it abandoned its practice of formally recognising
Page 174
governments in 1980. [20] In such circumstances, it is incumbent on a court to exercise
a discretion that reconciles the interests of the executive with common sense and
justice. The Republic of Somalia v Woodhouse Drake & Carey Suisse [21] offers a helpful
guide to courts on how to proceed in such a case. Here, an English court, in the
absence of clear executive certificates, held that in deciding whether a particular
political faction was the government of Somalia in 1991, it should have regard to the
following factors:
(a)
whether it is the constitutional government of the state;
(b)
the degree, nature and stability of the control that it exercises over the territory
of the state;
(c)
whether the British government has any dealing with it and, if so, what is the
nature of those dealings; and
(d)
in marginal cases, the extent of the international recognition it has as
government of the state.
Applying these principles, the Court found that the faction in question did not qualify as
the government of Somalia.

Legal effect of acts of unrecognised states and governments

For the purposes of municipal law, recognition is constitutive: it creates the state or
government. In the absence of recognition, no effect will be given to the legislative,
executive or judicial acts of a foreign government. The orthodox view is well illustrated
by the leading case on the subject, Luther v Sagor.
In 1919, a timber factory owned by the plaintiff was nationalized by the Soviet
government of Russia. The plaintiff fled to England. In 1920, the defendant purchased
timber from the factory and brought it to England. Britain did not recognise the Soviet
government in 1920. The plaintiff sued for the recovery of the timber on the ground
that the defendant claimed title from the act of an unrecognised government. The lower
court upheld this argument. [22] In 1921, Britain recognised the Soviet government as
the de facto government of Russia. When the matter went on appeal thereafter, the
Foreign Office submitted a certificate to the Court to the effect that it recognised the
Soviet government de facto, and that the previous government had been overthrown in
1917. The Court of Appeal therefore found for
Page 175
the defendant [23] on the ground that he had acquired title to the timber as a result of
the decrees of a recognised government. The court furthermore held that, first, there
was no distinction between de facto and de jure governments for the purpose of the
proceedings, and, secondly, the act of recognition had retrospective effect to 1917
when the Soviet government assumed effective control over Russia. [24]
While the principle expounded in Luther v Sagor is justifiable in respect of political
acts, such as the nationalisation of property, it can result in great injustices in matters
affecting the daily lives of individuals. For instance, in Adams v Adams, [25] a divorce
decree made by a Rhodesian judge appointed by the Smith government during the UDI
period was not recognised by an English court because of the non-recognition of
Rhodesia. Hardships of this kind persuaded the International Court of Justice in
the Namibia Opinion to hold that the consequences of non-recognition of South Africa’s
administration of Namibia should not extend ‘to those acts, such as, for instance, the
registration of births, deaths and marriages, the effects of which can be ignored only to
the detriment of the inhabitants of the territory’. [26]
Considerations of the above kind, coupled with the fact that recognition on occasion
loses sight of political reality, have prompted English courts to qualify the rule in Luther
v Sagor, usually by imaginative interpretations of the executive certificate. In Carl Zeiss
Stiftung v Rayner & Keeler (No 2) [27] the House of Lords held that effect could be given
to the laws of East Germany—which was not recognised as a state by Britain—on the
ground that it was an agent or a subordinate of the recognised authority for the
territory—the Soviet Union. It therefore acted for the Soviet Union within the territory
of East Germany. In the same case, Lord Wilberforce stated that the rule laid down
in Luther v Sagor was not absolute and that ‘where private rights, or acts of everyday
occurrence, or perfunctory acts of administration are concerned . . . the courts may, in
the interests of justice and common sense, where no consideration of public policy to
the contrary has to prevail, give recognition to the actual facts or realities found to exist
in the territory in question’. [28]
Page 176
The reasoning of the Carl Zeiss case was subsequently invoked in respect of Ciskei.
In GUR Corporation v Trust Bank of Africa [29] the plaintiff, a Panamanian company,
contracted with Ciskei to build a hospital and two schools. Under the contract with the
plaintiff, a guarantee of 10 per cent of the contract price was required to be made
available to cover claims by Ciskei for the cost of repairing defects in the building.
Accordingly, the sum of US $300 000 was lodged with the defendant bank in a deposit
account as security for a guarantee given by the bank to Ciskei. Ciskei made a claim on
the guarantee, but it was disputed whether the claim had been made in proper form
before the guarantee expired. Nonetheless, the defendant bank refused to return the
$300 000 to the plaintiff until the question of the validity of the claim was settled. The
plaintiff sued the defendant, seeking the return of the deposit; the defendant in turn
served a third-party notice on Ciskei, seeking to have the validity of Ciskei’s claim
determined. Ciskei counter-claimed for a declaration that it was entitled to the deposit.
None of the parties raised the question whether Ciskei, as an unrecognised state,
had standing to sue or be sued in an English court, as it was in their interest for the
matter to be resolved by an English court. The case came before a South African
expatriate, Steyn J (later Lord Steyn), in the Queen’s Bench Division, who raised the
question himself.
The Foreign and Commonwealth Office submitted two certificates on this subject. In
the first, it stated that ‘Her Majesty’s Government does not recognise the “Republic of
Ciskei” as an independent sovereign state, either de jure or de facto’. In the second, in
response to the question ‘[w]hich state, if any, does Her Majesty’s Government
recognise as . . . entitled to exercise . . . governing authority in respect of . . . Ciskei?’
it stated:
Her Majesty’s Government has not taken and does not have a formal position as regards the
exercise of governing authority over the territory of Ciskei. Her Majesty’s Government does not
have any dealings with the ‘Government of the Republic of Ciskei’ . . . . Her Majesty’s
Government has made representations to the South African Government in relation to certain
matters occurring in Ciskei and others of the Homelands to which South Africa has purported
to grant independence, notably on matters relating to individuals, but has not in general
received any positive response from the South African Government.
On the basis of these certificates, Steyn J ruled that Ciskei had no standing before an
English court. In upholding the orthodox position, Steyn J dismissed the applicability of
either of two possible
Page 177
exceptions to the rule. First, he held that the qualification suggested by Lord
Wilberforce in Carl Zeiss was not applicable as this had been a purely commercial
matter and not one in which the Court had been ‘confronted with the necessity of doing
justice to individuals who were caught up in a political situation that was not of their
making’. [30] Secondly, he held that the Carl Zeiss exception was not applicable as it
was not possible to infer from the certificate that Her Majesty’s Government viewed
South Africa as the governing authority of Ciskei.
On appeal, Sir John Donaldson MR engaged in an ingenious interpretation of the
executive certificate, in a manner described by Professor James Crawford as more
becoming to a novelist than a judge, [31] in order to reach a different conclusion. He
held that the Status of Ciskei Act [32] contained a delegation of legislative power to
Ciskei by South Africa which could be revoked subsequently by the South African
Parliament. He continued:
[T]he certified fact that ‘Her Majesty’s Government has made representations to the South
African Government in relation to certain matters occurring in Ciskei and others of the
Homelands to which South Africa has purported to grant independence’ gives rise to a clear
inference that Her Majesty’s Government regards the Republic of South Africa as continuing to
be entitled to exercise sovereign authority over the territory . . . .
 It follows that in my judgment the legal status of the Republic of Ciskei and its government
is indistinguishable from that which obtained in the case of the GDR and its government at the
time with which the Carl-Zeiss case was concerned . . . . I would therefore allow the appeal
and declare that the government of the Republic of Ciskei has locus standi in the courts of this
country as being a subordinate body set up by the Republic of South Africa to act on its
behalf. [33]
The judgment of the Court of Appeal may be ‘strained and artificial’, [34] but it does
illustrate the determination of the English courts to qualify the rule in Luther v
Sagor and to pursue a more flexible approach to the acts of unrecognised states and
governments. Here the English courts have been influenced by American decisions [35]
Page 178
which have refused to deny effect to the acts of unrecognised governments when this
would fly in the face of reality. [36] For instance, in 1933 a United States court gave
legal effect to an oil nationalization decree of the unrecognised Soviet government,
commenting that ‘to refuse to recognise that Soviet Russia is a government regulating
the internal affairs of the country, is to give fictions an air of reality which they do not
deserve’. [37]
The agency argument adopted in Carl Zeiss and GUR Corporation v Trust Bank of
Africa does not hold out much hope for a principled qualification of Luther v Sagor, as it
is dependent on the formulation of the executive certificate in each case. Lord
Wilberforce’s obiter dictum in Carl Zeiss, however, does offer some prospect for such a
qualification. It has received some support in later decisions [38] and was endorsed by
Lord Denning MR in an obiter dictum in Hesperides Hotels v Aegean Holidays Ltd. [39] In
a matter concerning the United Kingdom’s non-recognition of the Turkish administration
of northern Cyprus, he declared:
If it were necessary . . . I would unhesitatingly hold that the courts of this country can
recognise the laws or acts of a body which is in effective control of a territory even though it
has not been recognised by her Majesty’s Government de jure or de facto: at any rate, in
regard to the laws which regulate the day-to-day affairs of the people, such as their marriages,
their divorces, their leases, their occupations, and so forth; and furthermore that the courts
can receive evidence of the state of affairs so as to see whether the body is in effective control
or not. [40]
The effect to be given to the acts of unrecognised entities has not been fully considered
in South Africa. Nevertheless, there is support for the rule in Luther v Sagor. In Inter-
Science Research v Republica Popular de Mocambique [41] Margo J stated obiter that
without recognition
Page 179
of the government of Mozambique, its nationalisation decrees could not be
acknowledged as governmental acts and ‘would have no greater validity than the
seizure of . . . assets by some non-governmental body, having no lawful authority or
rights in the matter’. [42] In Standard Bank of SA Ltd v Ocean Commodities Inc, [43] a
case concerning Rhodesia, King J held, without examination of the English authorities,
that ‘our courts will not recognise the laws of an unrecognised state’. [44]
South African courts have yet to consider whether the rule in Luther v Sagor is
absolute. South African writers have suggested that the test that should be applied in
departing from this rule is whether cognisance can be taken of the acts of unrecognised
governments ‘without material harm being done to the executive’s foreign policy
goals’. [45] This is a sensible suggestion. Where laws which regulate the private, day-to-
day affairs of ordinary people can be recognised without any embarrassment being
caused to the executive’s foreign policy interests, the courts should refuse to
follow Luther v Sagor. In the absence of a clear indication from the executive of its
foreign policy interest in a certificate to the court, [46] a court may assume that the
executive will raise no objection to the recognition of the laws of an unrecognised
government or state affecting the private lives of ordinary people.

[1] Inter-Science Research and Development Services (Pty) Ltd v Republica Popular de
Mocambique 1980 (2) SA 111 (T) (‘Inter-Science Research v Republica Popular de Mocambique’) 116C–E;
AJGM Sanders ‘The courts and recognition of foreign states and governments’ (1975) 92 SALJ 165.
[2] Inter-Science Research v Republica Popular de Mocambique (n 1) 117D–E; Sanders (n 1) 166, 168.
[3] Margo J in Inter-Science Research v Republica Popular de Mocambique (n 1) 116–17; Sanders (n 1)
165. For an examination of the source of these powers, see Chapter 4 at notes 180–5.
[4] Inter-Science Research v Republica Popular de Mocambique (n 1) 118B.
[5] Ibid 118D.
[6] See Chapter 4 for a discussion of this matter.
[7] Section 17 of Act 87 of 1981.
[8] Inter-Science Research v Republica Popular de Mocambique (n 1) 118A; Bell v S [1997] 2 All SA
692 (E) 696a–b; The Akademik Fyodorov: Government of the Russian Federation v Marine Expeditions
Inc 1996 (4) SA 422 (C) 441J.
[9] Parkin v Government of the Republique Democratique du Congo 1971 (1) SA 259 (W) 259E.
[10] Sperling v Sperling 1975 (3) SA 707 (A).
[11] See the valid criticism of this case by H Booysen ‘Does South Africa recognise the German
Democratic Republic?’ (1975) 1 SAYIL 132.
[12] DJ Devine ‘The status of Rhodesia in international law’ 1974 Acta Juridica 115–24, 129–31.
[13] 1977 (1) SA 823 (N). See, too, S v Charalambous 1970 (1) SA 599 (T).
[14] This case is criticised by J Dugard ‘Rhodesia: Does South Africa recognise it as an independent
state?’ (1977) 94 SALJ 127; H Booysen ‘Recognition, treaties and the court’ (1977) 3 SAYIL 179.
[15] Supra (n 1) 116–18.
[16] See WJA Brand ‘Judisiële onafhanklikheid by kwessies van nie-erkenning’ (1974) 37 THRHR 329 at
338.
[17] [1939] AC 256 at 264.
[18] See para 11 of Chapter 4.
[19] Supra (n 17).
[20] See the certificates placed before the court in GUR Corporation v Trust Bank of Africa Ltd
(Government of the Republic of Ciskei, Third Party) [1986] 3 All ER 449 (CA) (‘GUR Corporation v Trust
Bank of Africa’). The texts of the certificates appear below between notes 29 and 30.
[21] [1992] 3 WLR 744 (QB). Cf Secretary of State for the Home Department v CC and CF [2012]
EWHC 2837 (Admin) paras 122 et seq.
[22] [1921] 1 KB 456.
[23] [1921] 3 KB 532.
[24] See further, on the subject of the retrospective effect of recognition, DP O’Connell International
Law 2 ed (1970) vol 1 at 185–92; Sanders (n 1) 172.
[25] [1971] P 188.
[26] 1971 ICJ Reports 16 at 56. In Emin v Yeldag [2002] 1 FLR 956 (Fam Div) an English court applied
this exception in respect of a divorce granted by a court in the Turkish Republic of Northern Cyprus.
[27] [1967] 1 AC 853 (‘Carl Zeiss’). See DW Greig ‘The Carl Zeiss case and the position of an
unrecognised government in English Law’ (1967) 83 LQR 96.
[28] At 954.
[29] Supra (n 20).
[30] GUR Corporation v Trust Bank of Africa (n 20) 454H–I.
[31] ‘Recent cases’ (1986) 57 BYIL 405 at 409.
[32] Section 3(1) of Act 110 of 1981.
 [33] GUR Corporation v Trust Bank of Africa (n 20) 465–6. For comments on this case, see C Warbrick
‘Unrecognised states and domestic law’ (1987) 50 MLR 84; D Lloyd Jones ‘Recognition of states and
governments—Republic of Ciskei’ 1987 Cambridge Law Journal 7; N Botha ‘From Pimlico to Bisho:
Recognition of South African national states in the light of Carl Zeiss’ (1987–1988) 12 SAYIL 156; FA
Mann ‘The judicial recognition of an unrecognised state’ (1987) 36 ICLQ 348; A Beck ‘A South African
homeland appears in the English courts: Legitimation of the illegitimate’ (1987) 36 ICLQ 350.
[34] Per Friedman J in S v Banda 1989 (4) SA 519 (B) 549F.
[35] Carl Zeiss (n 27) 954.
[36] For an examination of these decisions, see O’Connell (n 24) 172–80.
[37] Salimoff v Standard Oil Company of New York 262 NY 220 (1933), (1933–1934) AD 22.
[38] Both Steyn J and Sir John Donaldson MR gave implied approval to Lord Wilberforce’s obiter dictum
in GUR Corporation v Trust Bank of Africa (n 20) 454H, 463H–I.
[39] [1978] QB 205. See JG Merrills ‘Trespass to foreign land’ (1979) 28 ICLQ 523.
[40] At 218. See further RD Leslie ‘Unrecognised governments in the conflict of Laws: Lord Denning’s
contribution’ (1987) 14 CILSA 165. In Caglar v Billingham (Inspector of Taxes) 108 ILR 510 (noted by C
Warbrick in ‘Recent cases on public international law’ (1996) 45 ICLQ 954) it was held that ‘the courts will
not acknowledge the existence of an unrecognised state if to do so would involve them acting
inconsistently with the foreign policy or diplomatic stance of this country’ (534 para 121). See, too, Kirbis
v Türk (2009) EWHC 1918 (Admin) para 89. Cf Emin v Yeldag (n 26).
[41] Supra (n 1).
[42] At 116D.
[43] 1980 (2) SA 175 (T).
[44] At 181C and 183E.
[45] Sanders (n 1) 170; Brand (n 16) 334, 338–9; HA Strydom ‘Vrae rondom die erkenning van state’
(1992) 3 Stellenbosch LR 67 at 73.
[46] The court should, of course, request an executive certificate in any case where the status of the
state or government is in doubt. See Inter-Science Research v Republica Popular de Mocambique (n 1)
118B.

Page 180

Chapter 8
Territory

John Dugard and Rosalind Elphick

Territory occupies an important place in international law. [1] A state will not qualify as a
‘state’ unless it has a defined territory. [2] Moreover, the extent of a state’s
sovereignty [3] or jurisdiction will in most instances be limited to the extent of its
territory. Today the entire land mass of our planet, except Antarctica, where competing
claims to exclusive jurisdiction have been ‘frozen’ by treaty, [4] falls under the exclusive
jurisdiction of the states that make up the community of nations. Other forms of
‘territory’, notably the sea, sea-bed, airspace and outer space, either fall under the
exclusive jurisdiction of states or are governed by an international regime that accords
them the status of res communis, ie territory for communal use.
In previous centuries large portions of the globe were either uninhabited and
unclaimed, or inhabited by political communities characterised as ‘unorganised’,
‘primitive’, or ‘uncivilised’ by European states. Such territory was designated as terra
nullius, land belonging to no one, which could be acquired by means of discovery in the
15th and 16th centuries, and thereafter by means of occupation. Today no terra
nullius remains, and the concept has come under significant criticism for its historical
disregard of the claims of
Page 181
indigenous peoples. [5] Consequently territory may no longer be acquired by original
title, ie by occupation. [6] New states come into existence by recognition, which gives
them jurisdiction over a defined territory that previously belonged to another state, as
illustrated by the process of decolonisation. Existing states may extend their jurisdiction
over territory previously belonging to another state, in which case recognition of the
territorial change by other states is essential for a valid change in title.
The acquisition of territory today is therefore governed largely by recognition. The
rules that determined the acquisition of territory in earlier days are still important in
boundary disputes because such disputes can only be resolved in most instances by an
historical examination of the sources of the competing claims, which may go back
centuries. Boundary disputes are a common feature of modern international society and
are likely to increase. According to Ian Brownlie:
The pressures of national sentiment, new forms of exploitation of barren and inaccessible
areas, the strategic significance of areas previously neglected, and the pressure of population
on resources, give good cause for a belief that territorial disputes will increase in significance.
This is specially so in Africa and Asia, where the removal of foreign political domination has left
the successor states with a long agenda of unsettled problems, legal and political. [7]
In the formative years of international law jurists viewed sovereignty as being similar to
ownership in private law. Consequently the methods of acquiring property under Roman
private law were
Page 182
borrowed and adapted by international law. In classical international law the modes of
acquisition of territory were therefore discovery, occupation, accretion, cession,
conquest, and prescription. Today claims to territory are based on grounds derived
largely from the traditional methods of acquisition, the principle of effectiveness and
new expectations based on the right of self-determination.
There are important differences between the acquisition of territory under
international law and the acquisition of property in municipal law. In the first instance,
changes in territorial title involve a change of sovereignty which usually affects the
nationality and allegiance of people living on the territory in question. This may explain
why states strongly resist any change in territorial title. Secondly, there is often less
clarity about boundaries and the root of title in international law than in municipal law
as international law lacks the strict and detailed rules of transfer and registration of
ownership known to municipal law.

Background concepts

1 Intertemporal law

Discovery and conquest are no longer accepted as modes of acquisition of territory. The
idea that a state might acquire title to territory by discovery alone, without any
subsequent act of effective occupation, was suspect during the 15th and 16th
centuries. [8] Even if it did exist in this form, it has long since been abandoned as a root
of title. Conquest, on the other hand, was an accepted method of acquiring title until
after World War I. In 1928, war was outlawed by the General Treaty for the
Renunciation of War [9] (also known as the Pact of Paris or the Kellogg-Briand
Pact); [10] and in 1945, the Charter of the United Nations prohibited the use of force in
international relations. [11] As no right may arise from a wrong (ex injuria jus non oritur)
it follows that title acquired by the use of force is no longer recognised—as illustrated
by the Security Council’s non-recognition of Israel’s purported annexation of the Golan
Heights and East Jerusalem; [12] and the General Assembly’s resolution calling upon
states not
Page 183
to recognise a change of status on the territory of Crimea following its annexation by
Russia. [13]
Few territorial titles, if any, are today based on discovery alone. On the other hand,
a large number are based on conquest, arising out of war or colonial expansion. These
titles are to be judged by the law in force at the time the title was first asserted and not
by the law of today. This is the principle of intertemporal law. [14]
The leading authority on this subject is the Island of Palmas Case. [15] After the
Spanish American War of 1898, a defeated Spain ceded the Philippines to the United
States by treaty. In 1906 a United States official visited the Island of Palmas, believed
to be part of the territory ceded to the United States, and found that the Netherlands
had an administration on the small island (with a population of less than 1 000). In a
dispute referred to a single arbitrator, Max Huber, the United States claimed that Spain
had acquired title to the territory by discovery in the 16th century and that this had
been lawfully ceded to the United States by treaty. The Netherlands, on the other hand,
based its claim on the continuous and peaceful display of state authority over the island
since 1700.
In this case, Huber expounded two principles of intertemporal law. [16] The first
principle is that ‘a juridical fact must be appreciated in the light of the law
contemporary with it, and not of the law in force at the time when a dispute in regard
to it arises or falls to be settled’. [17] Consequently, the effect of discovery of the island
by Spain was to be determined by the rules of law in force in the 16th century.
The second principle is more controversial. According to Huber:
As regards the question which of different legal systems prevailing at successive periods is to
be applied in a particular case (the so-called intertemporal law), a distinction must be made
between the creation of rights and the existence of rights. The same principle which subjects
the act creative of a right to the law in force at the time the right arises, demands that the
existence of the right, in other words its continued manifestation, shall follow the conditions
required by the evolution of law. [18]
On the basis of this principle, Huber held that Spain had failed to maintain or
consolidate the title it had acquired by discovery through the effective display of state
authority. Conversely, the Netherlands
Page 184
had exercised its authority over the island peacefully and continuously since 1700. He
therefore awarded the island to the Netherlands.
Huber’s second principle of intertemporal law can be interpreted to destroy the first
completely. [19] If it means that every territorial title has to be reasserted in accordance
with every change in the law, it could produce great instability in the world order. The
alarming implications of such an interpretation are illustrated by the disputes over Goa,
the Falkland Islands, and Walvis Bay before its cession to Namibia in 1994.
 In 1961 India invaded and annexed Goa, a Portuguese colonial enclave on the Indian
subcontinent. In justification of its aggression, India argued that Portugal had illegally
occupied Goa by means of conquest in the 16th century and that India was simply
exercising its right of self-defence against the Portuguese aggressor—some 400 years
later. [20] A similar argument was raised by Argentina when it invaded the Falkland
Islands (called the Malvinas by Argentina) in 1982. Britain’s claim to the islands, argued
Argentina, was based on conquest in 1833, which was now unlawful and therefore
might be overthrown by the use of force. [21]
Arguments of the above nature were raised in respect of South Africa’s occupation of
Walvis Bay before 1994. [22] Walvis Bay was proclaimed a British Crown territory in
1878, and annexed to the Cape Colony in 1884, shortly before Germany annexed South
West Africa itself. From 1884 to 1910 Walvis Bay was administered as part of the Cape
Colony, and after the Union of South Africa was formed in 1910, it became a part of the
Union. In 1915 South African forces occupied German South West Africa. After South
Africa was granted a mandate over South West Africa by the League of Nations, Walvis
Bay was ‘administered as if it were part of the mandated territory’. [23] In 1977, as
Namibian independence became a possibility, South Africa reasserted its right to Walvis
Bay and placed it under the administration of the Cape Province. [24]
According to Lynn Berat, in her study on Walvis Bay, [25] Max
Page 185
Huber’s second principle of intertemporal law was applicable to Walvis Bay. She claimed
that ‘the doctrine of intertemporal law invalidates an original title by applying new rules
that affect the root of title’. Consequently, South Africa’s claim to Walvis Bay, which
was based on 19th century colonialist rules governing the acquisition of territory, was
no longer valid as it was in conflict with new peremptory norms of international law
which outlaw colonialism and assert the right of self-determination. ‘Any right,
transaction, or benefit that violates the jus cogens,’ she maintained, ‘is null and
void.’ [26]
Undoubtedly, the doctrine of jus cogens now plays an important part in
contemporary international law, but to suggest that peremptory norms should be
applied with retroactive effect to territorial titles is positively dangerous. Many
boundaries in the modern world have been fixed in violation of what are today regarded
as peremptory norms of jus cogens, namely the prohibition on the use of force and
colonial occupation. To question the validity of these titles on the ground that they
violate today’s norms of jus cogens applied retrospectively is to open Pandora’s box. As
Professor—later Judge—RY Jennings warned in 1963, ‘[u]nder these conditions no title
would be secure and the supposed aim of the law—stability—would be utterly
defeated’. [27]
There are circumstances in which Huber’s second principle of intertemporal law
should be employed to bring the law into line with reality, as, for example, in
the Aegean Sea Continental Shelf Case [28] in which the International Court of Justice in
1978 interpreted a 1931 treaty reference to Greece’s ‘territorial status’ as including
Greece’s continental shelf, despite the fact that the continental shelf was recognised as
being part of a state’s territory only after 1945. This principle, however, should be used
with great caution. [29] As Jennings suggests, [30] it should be invoked only to deny the
claim of a party relying on the first principle of intertemporal law, where that party has
failed to maintain a minimum degree of sovereign activity over the territory and by its
neglect has abandoned its own claim to title or acquiesced in that of the rival claim—as
happened in the Island of
Page 186
Palmas Case. In such cases, courts will have regard to the effectivités—that is, the
evidence of the effective display of state functions by a state.
The International Court of Justice has settled a number of territorial disputes by
applying the doctrine of intertemporal law, [31] particularly in boundary disputes
between African states. [32] The same is true of other tribunals. [33]

2 Uti possidetis

Closely related to the principle of intertemporal law is the principle of uti possidetis,
according to which colonial boundaries, however arbitrarily drawn by the imperial
powers, are to be respected. [34] This principle, which has its origins in Latin America
and has been endorsed by the Organization of African Unity, [35] is designed to prevent
the chaos that inevitably would result from attempts to redraw boundaries to coincide
with ethnic groupings.
In the Frontier Dispute Case [36] between Burkina Faso and Mali, the International
Court of Justice held that this practice was a customary rule of ‘general scope’ [37] which
applied to a new state from the moment it became independent. In effect this ‘freezes
the territorial title’ [38] and confines the right of self-determination of peoples to a
territory defined by the colonial power. In the Frontier Dispute Case the International
Court acknowledged that the principle of uti possidetis conflicts with the right of peoples
to self-determination. [39] However, said the Court:
Page 187
[T]he maintenance of the territorial status quo in Africa is often seen as the wisest course, to
preserve what has been achieved by peoples who have struggled for their independence, and
to avoid a disruption which would deprive the continent of gains achieved by much sacrifice.
The essential requirement of stability in order to survive, to develop and gradually consolidate
their independence in all fields, has induced African states judiciously to consent to the
respecting of colonial frontiers, and to take account of it in the interpretation of the principle of
self-determination of peoples. [40]
The principle of uti possidetis is closely related to that of ‘territorial integrity’, which is
asserted in the Declaration on the Granting of Independence to Colonial Countries and
Peoples [41] and other resolutions of the General Assembly on the subject of self-
determination.
The principle of uti possidetis in the non-colonial context was applied to justify the
dissolution of Yugoslavia in 1991–92. Here, the Arbitration Commission—established by
the European Community under the chairmanship of Mr Badinter to oversee the
guidelines contained in the European Community’s Declaration on Yugoslavia—held that
the internal federal boundaries of Slovenia, Serbia, Croatia, Bosnia and Herzegovina,
Montenegro and Macedonia, which together comprised the Socialist Federal Republic of
Yugoslavia, became frontiers protected by international law. ‘This conclusion follows,’
said the Commission, ‘from the principle of respect for the territorial status quo and, in
particular, from the principle of uti possidetis . . . today recognised as a general
principle.’ [42] This is dangerous reasoning as it undermines the principle of territorial
integrity in respect of federal states. [43] Conversely, it is unfair to territorial units such
as Kosovo [44] that enjoyed a lower status in the former Yugoslavia but were
nevertheless clearly demarcated and granted a certain degree of autonomy.
 Although a force for stability in international relations, the principle of uti
possidetis is highly controversial. It often internationalises as boundaries lines that were
simply administrative delimitations in the colonial era and not intended as
frontiers. [45] This was acknowledged
Page 188
by the International Court of Justice in the Land, Island and Maritime Frontier Dispute
(El Salvador/Honduras) Case when it stated that ‘uti possidetis juris is essentially a
retrospective principle, investing as international boundaries administrative units
intended originally for quite other purposes’. [46] The principle is, however, mainly
criticized on the ground that it serves the interests of the dominant societal groups at
the expense of the right of peoples to self-determination. [47] Despite these criticisms,
the principle enjoys continued application. It was recently applied by the International
Court of Justice to resolve a boundary dispute in the African context in Frontier Dispute
(Burkina Faso/Niger). [48]

3 Original versus derivative title

A traditional distinction is drawn between original and derivative titles of territorial

sovereignty. The former, on the one hand, relates to titles that attach to land that has
not previously been the subject of any other states title, as in the case of occupation of
a terra nullius or accretion. Derivative titles, on the other hand, are the result of a
transfer of territory from a previous sovereign, as in the case of cession and conquest.

The modes of acquisition of territory

1 Discovery

Claims that discovery gave a complete title to territory during the 15th and 16th
centuries are today dismissed. The prevailing view is that at that time, discovery
created an inchoate title only and that it was necessary for the discoverer to perfect its
title by the effective occupation of the territory within a reasonable time. [49]

2 Occupation

Occupation is the method of acquiring territory not belonging to any other state,
ie terra nullius. Although there is no terra nullius left in the modern world, occupation
still features prominently in boundary disputes. To succeed in a claim based on
occupation, a claimant must prove that it had ‘the intention and will to act as sovereign’
(animus occupandi), that it exercised actual authority over the territory
Page 189
(corpus) [50] and that the territory was terra nullius. Animus occupandi may be proved
by the hoisting of a flag or the placing of a plaque on the territory [51] or by publication
of the occupation in a widely distributed newspaper. [52] Corpus requires proof of the
continuous and effective display of authority over the territory. The degree of authority
required varies according to the nature of the territory. Tribunals have been satisfied
with very little in the actual exercise of sovereign rights in thinly populated or
uninhabitable territories. [53] In Nicaragua v Colombia, the International Court held that
sovereignty over ‘minor’ islands could succeed based on very little proof of effective
occupation, provided that the opposing claimant was unable to present greater
proof. [54]
South Africa’s occupation of Marion Island and Prince Edward Island, in order to set
up a meteorological station, provides an example of a recent assertion of
sovereignty. [55] These islands were first discovered in 1772 by a French navigator,
Marion du Fresne, who made no claim to the islands on behalf of France. In 1776 James
Cook sailed between the islands and gave them their present names. [56] However,
Britain took no steps to establish effective control over the islands. When the South
African government decided to claim these islands in 1947 and 1948, it first consulted
the British government, which gave an assurance that it laid no claim to the
islands. [57] Thereafter, South Africa established its authority over the islands. In 1948
Prime Minister Dr DF Malan made it clear that South Africa had acquired title by
occupation. He told Parliament:
These two islands, though they were discovered so long ago, were practically never occupied.
Therefore, because there was no occupation, sovereignty over these islands was really never
established by any particular country. The
Page 190
nearest to occupation came from the side of England. An application was made to the British
Government on one occasion by a company to lease the island to work the guano deposits
which they thought were on the island, on another occasion by a whaling company. But these
rights which were granted by the British Government were never exercised, and in any case
the islands were never occupied. In annexing these two islands I do not think that the Union
can in any way be accused or suspected of harbouring aggressive imperialistic designs. These
islands practically belong to nobody, so we do not deprive anybody else of his rights. [58]
The Act annexing the islands likewise employs the language of international law in
providing in its preamble:
Whereas effective occupation and administration of Marion Island and Prince Edward Island
were established on the twenty-ninth day of December, 1947 and the fourth day of January,
1948, respectively, and such occupation and administration will continue permanently. [59]
The concept of terra nullius presents difficulties for contemporary international law as it
has undergone important changes. During the formative years of international law
opinion was divided as to whether international law applied to indigenous peoples: the
naturalists argued that all the peoples of the world enjoyed certain inalienable rights,
while the positivists denied such rights to indigenous peoples and claimed that
international law applied to Christian, civilised nations only. [60] During the 19th century
the positivist view prevailed, with the result that indigenous, non-European peoples in
loosely organised societies were viewed as having no rights under international law.
Consequently, their territory was viewed as terra nullius—a designation that gave legal
backing to the colonial expansion of that century. Modern international law, determined
to erase this mark of imperialist paternalism from the historical record, has sought to
minimise the 19th century positivist position. In the Western Sahara Opinion of 1975,
the International Court of Justice found that the Western Sahara was inhabited by
nomadic but socially and politically organised tribes in 1884, at the time of Spanish
colonisation of the territory. It held:
Whatever differences of opinion there may have been among jurists, the state practice of the
relevant period indicates that territories inhabited by tribes or peoples having a social and
political organization were not regarded as terrae nullius. It shows that in the case of such
territories the acquisition of territory was not generally considered as effected unilaterally
through ‘occupation’ of
Page 191
 terra nullius by original title but through agreements concluded with local rulers . . . [S]uch
agreements with local rulers, whether or not considered as an actual ‘cession’ of the territory,
were regarded as derivative roots of title, and not original titles obtained by occupation
of terrae nullius. [61]
When the first British settlement was established in Australia in 1788, it forcibly
displaced a nomadic aboriginal people with a complex social and political organisation.
Conquest, rather than occupation, might therefore be seen to be the method by which
title was acquired, in which case, under English constitutional law, [62] the laws of the
aboriginal people would have remained in force. Judicial decisions held otherwise. [63] In
1889 the Privy Council held that the colony of New South Wales had not been acquired
by conquest but had ‘consisted of a tract of territory, practically unoccupied, without
settled inhabitants or settled law, at the time when it was peacefully annexed to the
British Dominions’. [64] Subsequent Australian decisions confirmed this view, [65] but in
1992 it was repudiated by the High Court in Mabo v State of Queensland, [66] which
approved the approach adopted by the International Court of Justice in the Western
Sahara Opinion.
This is not an issue in South Africa as the indigenous people constitute the majority
and today exercise political power. Despite this, it is interesting to consider the legal
basis for Dutch, British and
Page 192
Boer settlements in South Africa at the expense of the indigenous population.
When the Dutch East India Company first established an outpost at the Cape of Good
Hope in 1652, it was not envisaged as a settlement or colony. Over the years, however,
this outpost evolved from a refreshment station to a settlement as employees of the
company were released to become settlers and began the inevitable process of
expansion in search of new lands. In the process, the indigenous inhabitants of the
Cape, the San and Khoikhoi, were subjugated. While the socio-political fabric of the San
was rudimentary, the Khoikhoi had a developed social organisation. Was the Cape
therefore acquired by the Dutch East India Company on behalf of the Republic of United
Netherlands by occupation of a terra nullius or by conquest? [67]
The African tribes to the east and north were accepted as political societies by the
Dutch, [68] the British [69] and the Boers, [70] who all at some stage or another entered
into treaties with tribal leaders. While the status of these treaties under international
law was uncertain, they did at least make it clear that the African-occupied territories
were not viewed as terrae nullius. The extension of white settlement into the Eastern
Cape, [71]
Page 193
Natal, [72] Orange Free State and Transvaal [73] was therefore based on conquest or
cession and not occupation.
South Africa’s claim to the enclave of Walvis Bay before 1994 was based on a blend
of cession and occupation. In 1876 an agreement was entered into between the special
Cape Commissioner, William Coates Palgrave, and Chief Kamaherero of the Herero,
establishing a protectorate over a large tract of land that included Walvis Bay. [74] This
was not viewed as a treaty of cession but rather as an administrative act over a terra
nullius by Britain that served notice on European rivals that a particular territory was
being occupied. [75] In 1878 Commander Richard Cossantine Dyer RN proclaimed Walvis
Bay a British Crown territory and in 1884 [76] it was annexed to the Cape
Colony. [77] After Dyer’s symbolic act of annexation, a resident magistrate was installed
and an administration established appropriate to such a small, isolated territory with
only a transient indigenous population. Thus, an effective occupation was established
over an area viewed as terra nullius in accordance with international law of the time. [78]

3 Accretion

A state may acquire sovereignty over territory as a result of natural forces, as occurs
when a volcanic island rises within a state’s territorial sea or the delta of a river mouth
expands. The natural changes in the geomorphology of a river mouth has been the
source
Page 194
of a recent dispute settled by the International Court of Justice between Costa Rica and
Nicaragua. In that case, historical agreement between the parties set the boundary
between them as running along the course of the San Juan river to its outlet to the sea
at ‘the extremity of Punta de Castilla’. Over the course of time, however, the mouth of
the river drifted north of Punta de Castilla. The Court determined that the boundary
between the parties had followed the course of the river as it altered. [79]

4 Cession

Cession is the transfer by treaty of sovereignty over the territory of one state to
another state. Before 1945 territory was frequently ceded after a war by the
vanquished state to the victorious state. Thus, in 1814, the Cape of Good Hope was
ceded by the Netherlands to Britain in the Convention of London at the conclusion of
the Napoleonic wars. [80] Cession of territory has also resulted from a gift or sale. In
1866 Austria ceded Venice to France as a gift and shortly thereafter France ceded
Venice to Italy. Russia’s sale of Alaska to the United States in 1867 for $7.2 million was
effected by cession. [81]
South Africa’s cession of Walvis Bay and the Off-Shore Islands (better known as the
Penguin Islands) to Namibia on 1 March 1994 is a recent example of the transfer of
territory from one state to another. [82]
Today, restraints are placed on the cession of territory by the prohibition on the use
of force and the right of self-determination. If the acquisition of territory by conquest is
no longer permitted, it follows logically that the transfer of territory after a war by
cession from the vanquished to the victorious state is also prohibited—in
Page 195
accordance with the principle ex injuria jus non oritur. This is confirmed by article 52 of
the Vienna Convention on the Law of Treaties, which provides that ‘[a] treaty is void if
its conclusion has been procured by the threat or use of force in violation of the
principles of international law embodied in the Charter of the United Nations’.
However, by virtue of the principle of intertemporal law, cessions which took place
following the use of force at a time when the use of force was not unlawful, remain
effective. This is borne out by the submissions of the parties in the case of Obligation to
Negotiate Access to the Pacific Ocean (Bolivia v Chile) in which Bolivia conceded the
validity of the 1904 Peace Treaty between itself and Chile, in terms of which it ceded its
coastal territory to Chile (which had conquered that territory) leaving Bolivia land-
locked. [83]
 The transfer of territory by means of cession without the consent of the people of the
ceded territory is today difficult to reconcile with the right of self-
determination. [84] When the South African government attempted to cede KaNgwane
and the Ingwavuma district of KwaZulu to Swaziland in 1982, in pursuance of the policy
of apartheid, [85] it was argued that this proposed cession, on which the affected people
were not consulted, violated the right (enshrined in the Declaration on the Granting of
Independence to Colonial Countries and Peoples of 1960) of a people to determine their
own future. At the same time, it was argued that the proposed cession violated the
principle of uti possidetis as it sought to tamper with colonial boundaries in Africa.
Political opposition to this proposal compelled the South African government to abandon
this scheme. [86] Arguments based on international law contributed to this decision, as it
was
Page 196
made clear to the government that the validity of any agreement of cession in violation
of the right of self-determination and uti possidetis would be questionable [87] and that
the transfer of sovereignty would be subjected to collective non-recognition by the
international community along the same lines as the non-recognition of the TBVC
states. [88]

5 Conquest

Before the prohibition on the use of force, conquest, annexation and subjugation were
accepted methods of acquiring territory. As in the case of occupation, it was necessary
to demonstrate both animus—the intention to annex the territory—and corpus—the
physical control of the territory. Consequently, a state did not acquire territory by
conquest if it purported to annex territory during the course of a war.
In May and September 1900, Britain purported to annex the Orange Free
State [89] and Transvaal [90] respectively, despite the fact that the war, albeit in the form
of guerrilla operations, continued until 1902. It was generally accepted that Britain’s
annexation was premature and that Britain did not obtain title to the Boer Republics
under international law until the termination of hostilities. [91]
The lawfulness of the British annexation of the South African Republic came before
the Transvaal Supreme Court after the war, in Van Deventer v Hancke & Mossop, [92] in
which the validity of a Republican decree, issued after the British proclamation of
annexation of the Transvaal, was raised. Ironically, counsel for the plaintiff, who had
secured title to certain bales of wool as a result of the Republican decree, was JC
Smuts, one of the Boer generals who had continued to wage war until 1902. [93] Both
Innes CJ and Mason J held that, as judges sitting in a British colonial court, they were
obliged to give effect to the British annexation. At the same time, they cast doubt
Page 197
on the validity of the annexation under international law. According to Innes CJ:
It was argued for the plaintiff that the Annexation Proclamation was premature; that at the
time when this wool was confiscated the district of Vryheid was subject to the de facto control
and administration of the Boer forces; that although the Proclamation purported to annex the
territory of the Transvaal to the empire, there had, at the time of the annexation, been no
effectual occupation of it as a country, and no subjugation of its people; and that therefore the
Republic continued to exist as a state, and its government was entitled to exercise legislative
and administrative functions. It is no doubt correct as a general rule of international law that
two circumstances are necessary to create a complete title by conquest: the conqueror must
 express in some clear manner his intention of adding the territory in question to his dominions,
and he must by the exercise of military force demonstrate his power to hold it as part of his
own possessions. It is also true that in March 1901, large portions of the Transvaal, including
the district of Vryheid, were neither occupied nor dominated by British troops; but on the
contrary were under the de facto control of the Boer forces. And if this were a foreign court
engaged in trying a cause in regard to which the question of when the conquest of the
Transvaal was complete became relevant to the inquiry, it is possible that points of
considerable intricacy and difficulty would present themselves. But those considerations are
not present here. This is a court constituted by the British Crown, exercising powers and
discharging functions derived from the Crown. In its dealings with other states the Crown acts
for the whole nation, and such dealings cannot be questioned or set aside by its courts. They
are acts of state into the validity or invalidity, the wisdom or unwisdom, of which domestic
courts of law have no jurisdiction to inquire. [94]
In 1928 war was outlawed by the Kellogg-Briand Pact and, in 1945, the Charter of the
United Nations prohibited the use of force in international relations against the
territorial integrity or political independence of any state. [95] Modern customary
international law recognises a similar prohibition. [96] In these circumstances, it is clear
that territory can no longer be acquired by the use of force, ie by conquest. On the
other hand, it is equally clear that titles acquired by conquest before 1928 must be
recognised as lawful in accordance with the principle of intertemporal law. [97] Some
have argued that the prohibition on the acquisition of territory by force applies only in
the case of an aggressive, unlawful war and that a state may lawfully
Page 198
obtain title to territory acquired in self-defence. [98] This view is untenable [99] and is
rejected by both state practice and resolutions of the United Nations. The General
Assembly Declaration on Principles of International Law Concerning Friendly Relations
and Co-operation among States in Accordance with the Charter of the United Nations of
1970 [100] draws no distinction between the lawful and unlawful use of force in
providing:
The territory of a state shall not be the object of acquisition by another state resulting from the
threat or use of force. No territorial acquisition resulting from the threat or use of force shall be
recognised as legal.
There have been only a handful of claims of forceful annexation made during the United
Nations era. Moreover, the international reaction to these cases has reinforced the
customary international law prohibition on annexation. Two examples come to mind.
Although no United Nations resolution has branded Israel as the aggressor in the
Six-Day War of 1967, resolutions of both the Security Council [101] and the General
Assembly [102] have condemned Israel’s purported annexation of East Jerusalem and the
Golan Heights on the ground that ‘the acquisition of territory by force is inadmissible’—
with no distinction drawn between the lawful and the unlawful use of force. The non-
recognition of Israel’s annexation of East Jerusalem and the Golan Heights [103] supports
the view that territory may not be acquired in a war of self-defence. South Africa has
supported UN General Assembly condemnations of the Israeli occupation. [104] A more
recent case of annexation has been undertaken by a permanent member of the
Security Council against a United Nations member. In 2014 the Russian Federation
annexed Crimea, part of the territory of Ukraine. This act followed intervention by
forces of the Russian Federation, a referendum, and a declaration of independence in
Crimea. [105] The United Nations General Assembly rejected the validity of the
referendum and does not acknowledge
Page 199
any change in status for the territory. [106] South Africa has not voted in favour of any of
the resolutions passed in condemnation of Russia’s occupation of Crimea: it abstained
from the original resolution condemning the annexation in 2014 and voted against the
subsequent two resolutions, which characterised Crimea as being subject to ‘temporary
occupation’.

6 Prescription

Prescription is a concept of uncertain content in international law. Although judicial

decisions generally avoid the use of the term, they nevertheless support the conclusion
that, by a process analogous to prescription in municipal law, long possession and the
exercise of effective control may give rise to a valid title. In the Kasikili/Sedudu
Island Case, [107] between Botswana and Namibia, both parties agreed that acquisitive
prescription was recognised in international law and that the criteria to be satisfied
were possession à titre de Souverain and peaceful, public and uninterrupted possession
that endured for a certain length of time. Although the International Court was not
required to apply this test, it did not disagree with the test formulated by the parties.
In theory, the distinction between occupation and prescription is clear. Occupation
applies only to a territory that is terra nullius. The original occupation must be lawful,
and passage of time is irrelevant. Prescription, on the other hand, is ‘a portmanteau
concept that comprehends both a possession of which the origin is unclear or disputed,
and an adverse possession which is in origin demonstrably unlawful’. [108] However, they
have much in common for both are rationalisations of effective possession and control.
For this reason, tribunals have frequently failed to draw any distinction between the
two. In the Island of Palmas Case arbitrator Max Huber stressed the importance of
sovereign activities (effectivités) on the territory of the state on the part of a successful
claimant state when he stated that ‘the continuous and peaceful display of territorial
sovereignty (peaceful in relation to other states) is as good as title’. [109] Similarly, in
Page 200
the Legal Status of Eastern Greenland Case, often seen as the leading case on
occupation, the Permanent Court of International Justice emphasised that Denmark’s
claim to Eastern Greenland did not rest on any ‘particular act of occupation’ but on the
peaceful and continuous display of state authority. [110] The International Court of
Justice relied on effectivités in finding for Malaysia in its dispute with Indonesia
in Sovereignty over Pulau Ligitan and Pulau Sipadan. [111] Most recently, in Territorial
and Maritime Dispute (Nicaragua v Colombia), the Court described relevant acts as ‘in
particular, but not limited to, legislative acts or acts of administrative control, acts
relating to the application and enforcement of criminal or civil law, acts regulating
immigration, acts regulating fishing and other economic activities, naval patrols as well
as search and rescue operations’. [112]
As a result of this blurring of the distinction between occupation and prescription,
some writers have suggested that the two modes of territorial acquisition should be
merged into a common source of title [113] or be replaced by a new mode known as the
historical consolidation of title. [114]
Questions of source of title commonly arise in disputes in which the original claim to
title is lost in an uncertain historical record. In such cases, the court is compelled to
decide which of the claimants has effectively exercised sovereignty peacefully and
continuously over the disputed territory in recent times. [115] Here, acquiescence plays a
major role in the identification of title. [116] Where a state fails to protest over the
assertion of sovereignty by another state in a territory that it claims to be its own, it is
unlikely to succeed in persuading a court that it has not abandoned its claim to the
disputed territory, for, as the International Court of Justice stated
in Malaysia/Singapore, ‘failure to respond to conduct à titre de Souverain amounts to
acquiescence’. [117] It was essentially on this basis that Huber found in favour of the
Netherlands in the Island of Palmas Case. Spain failed to protest at the Netherlands’
assertion of authority over the island
Page 201
and was thereby found to have acquiesced in the rival claim. [118] A state that fails to
protest over the exercise of sovereignty by its rival remains silent at its peril. This is an
issue that has featured in the Falklands/Malvinas dispute where Argentina has protested
sporadically and irregularly over Britain’s occupation of the islands. [119]
In the southern African context, the issue of protest has been raised in respect of
Lesotho’s irredentist claims to the eastern part of the Orange Free State (the so-called
‘conquered territories’). South African claims to the territory, based on conquest,
cession and possibly prescription, appear to be sound. [120] Despite this, Lesotho, since
it became independent in 1966, has protested that the territory was unlawfully taken
from it by the Orange Free State. [121] While Lesotho may argue that it was unable to
protest against the incorporation of the territories during the period that it was a British
colony—from 1868 to 1966—it is difficult to reconcile its claims with the principle of uti
possidetis, which seeks to ensure respect for colonial boundaries.

7 Estoppel

The principle of estoppel or preclusion is part of international law. According to Lord

McNair:
[I]t is reasonable to expect that any legal system should possess a rule designed to prevent a
person who makes or concurs in a statement upon which another person in privity with him
relies to the extent of changing his position, from later asserting a different state of affairs. [122]
In territorial disputes, the acceptance by one state of another’s claim to the territory
will preclude it from reasserting its claim. In the Temple of Preah Vihear
Case, [123] concerning a border dispute between Cambodia and Thailand, the evidence
showed that Thailand in its past conduct had accepted that the Temple fell within
Cambodia, with the result that it was estopped from later reasserting its claim to that
area.
In the late 1980s, as Namibian independence drew near and South Africa showed no
sign of including Walvis Bay in an independent
Page 202
Namibia, it was argued that South Africa, by administering Walvis Bay as part of the
mandated territory of South West Africa from 1922 to 1977, had led the international
community to believe that it had integrated Walvis Bay into South West Africa/Namibia,
with the result that it was estopped from later asserting that the port remained part of
South Africa. According to Berat, estoppel had ‘shifted’ title from South Africa to
Namibia. [124] This claim was questionable for two reasons.
First, it is doubtful whether estoppel can operate to ‘shift’ a title. Commenting on the
judgment of the International Court of Justice in the Temple of Preah Vihear Case, RY
Jennings states:
 [A]lthough the case confirms that estoppel may assist, and even assist with decisive effect, in
the interpretation of facts, and instruments and acknowledgements relative to the vesting of
title, it still remains true to say that estoppel is not itself a root of title. [125]
Secondly, it is by no means certain that South Africa’s conduct in respect of Walvis Bay
met the requirements for the operation of estoppel, which are:
(a)
The statement of fact must be clear and unambiguous.
(b)
The statement of fact must be made voluntarily, unconditionally, and must be
authorised.
(c)
There must be reliance in good faith upon the statement either to the detriment
of the party so relying on the statement or to the advantage of the party making
the statement. [126]

8 Territorial claims based on the right of self-determination

In modern international law, a number of claims to territory are made that are based
on the right of self-determination and decolonisation.

(a) Legal ties of a historical nature

In its advisory opinion on Western Sahara, [127] the International Court of Justice
considered the question whether the Spanish colony of Western Sahara had an identity
of its own in 1884, when it was colonised by Spain, which would entitle it to exercise its
right to self-determination as a separate entity, or whether its historical ‘legal ties’ with
either of its neighbours, Morocco and Mauritania, were so close that it should be
reintegrated with either of these territories. Although the International Court found in
favour of the separate identity of Western Sahara, the majority judgment and several
of the
Page 203
separate opinions elaborated on the notion of ‘legal ties’ in the context of
decolonisation.
The principle of ‘legal ties’ has its basis in General Assembly Resolution 3292 (XXIX),
in which the General Assembly asked the International Court, ‘without prejudice to the
application of the principles embodied in General Assembly Resolution 1514 (XV)’, to
give an advisory opinion on, inter alia, the ‘legal ties’ between Western Sahara and
Morocco and Mauritania. That the term ‘legal ties’ has its origin in a resolution of special
application to Western Sahara was stressed by the Court in its opinion when it stated
that the term lacked ‘a very precise meaning’, with the result that its meaning ‘has to
be found rather in the object and purpose of General Assembly Resolution 3292
(XXIX)’. [128] The concept of ‘legal ties’ therefore should not be too hastily invoked in
other situations, [129] particularly when it is to be used to force integration of one
territory into another without first permitting the inhabitants of the disputed territory to
express their views on integration by means of a referendum.
That the doctrine of legal or historical ties is not to be considered as a source of
territorial title, is evidenced by the clear rejection of Iraq’s historical claim to
Kuwait, [130] which constituted the basis for its invasion and annexation of Kuwait in
1990.
(b) Colonial enclaves
An enclave is an area totally surrounded by the territory of one state or a ‘relatively
small area totally surrounded on the landward side by the territory of one other
state’, [131] a concept that clearly encompassed Walvis Bay before 1994. Contemporary
international law contemplates that self-determination, in the case of an enclave
classified as non-self-governing, will take the form of automatic reintegration or
retrocession into the territory of which the enclave forms a part or the exercise of a free
choice by the people of that territory in a plebiscite or referendum.
Where the enclave is small and forms an integral part of the claimant state, the
United Nations has dispensed with the need for a
Page 204
free choice on the part of the inhabitants of the enclave. In this case, priority is given to
para 6 of Resolution 1514 (XV), which insists on the preservation of ‘the national unity
and territorial integrity’ of a decolonised territory. [132] The return of the Spanish
enclave of Ifni to Morocco in 1969 provides an example of retrocession without a
plebiscite or referendum. This enclave, covering some 1 500 km and with a population
2

of 53 000, was completely surrounded by Morocco on its landward side. Walvis Bay was
likewise treated as a colonial enclave to which para 6 of Resolution 1514 (XV) was
applicable. [133] This seems to have been the basis of both General Assembly and
Security Council resolutions, which stressed that Walvis Bay formed an integral part of
Namibia. [134]

9 Concluding comments

International courts do not scrupulously apply the traditional rules relating to the
acquisition of territory. In this respect, international law differs from municipal law in
which clear rules of acquisition of ownership are strictly applied. Occupation and
prescription both require manifestations of effective control with the result that
evidence of sovereign activities on a territory (effectivités) has come to be regarded as
a source of title. This accords with Huber’s dictum in the Island of Palmas Case that ‘the
continuous and peaceful display of territorial sovereignty (peaceful in relation to other
states) is as good as title’. [135] This view has recently been endorsed by the decision in
the Eritrea/Yemen arbitration in which the tribunal [136] reaffirmed:
The modern international law of the acquisition (or attribution) of territory generally requires
that there be: an intentional display of power and authority over the territory, by the exercise
of jurisdiction and state functions on a continuous and peaceful basis. [137]
In territorial disputes, a court will often not look for absolute title on the part of one
state but will be satisfied that one party has, on balance, a relatively stronger title than
the other. Where one state is able to show some legal title to the territory,
the effectivités will determine title. Notions of recognition, acquiescence and estoppel
Page 205
will play a role in this process. In Malaysia/Singapore [138] the International Court relied
heavily on acquiescence and tacit agreement to find that Singapore had a stronger
claim to a disputed island than Malaysia and thus had acquired sovereignty over the
island. In a dissenting opinion, the present writer, sitting as judge ad hoc, [139] stressed
that acquiescence should, as shown in the Eritrea/Yemen case, be accompanied by an
‘intentional display of power and authority over the territory, by the exercise of
jurisdiction and state functions on a continuous and peaceful basis’. [140]
 In the case of former colonies, the principle of uti possidetis will play an important
role in determining title, but where the administrative boundary of the former colony is
unclear, recourse will be had to the effectivités. [141]

Territorial disputes in Africa [142]

Before colonial rule, Africa was not delimited into states. This is not to say that there
were no boundaries at issue; in fact there were many localised, traditional boundaries
established to account for different African peoples, their cultures and traditions.
However, these were never formalised. [143]
As a result of colonisation, the continent was demarcated by foreign powers without
regard to the peoples’ traditional boundaries. [144] These borders divided ethnic groups
and the lands to which they were attached, causing significant ethnic tensions. Despite
this, most colonially created boundaries have remained intact as a result of the
application of the principle of uti possidetis which has been endorsed by African
leaders. [145]
Ethno-political tensions caused by colonial boundaries continue to beset post-colonial
Africa. Many of these disputes have come before the International Court of Justice
which in most cases has turned to colonial boundaries to resolve territorial disputes.
This is illustrated
Page 206
by the following cases which have featured prominently in the litigation of the Court.
The Frontier Dispute case [146] concerned a boundary dispute between Burkino Faso
and Mali arising out the process of decolonisation of French West Africa. In 1960, the
colony of French Sudan became the independent state of Mali and the colony of Upper
Volta became the independent state of Upper Volta (later renamed Burkina Faso). The
Court resolved the dispute by applying the uti possidetis juris principle, holding that,
upon independence, the two states had succeeded to the territorial boundaries which
had delimited the former colonies of Upper Volta and French Sudan in 1932.
In Territorial Dispute, [147] involving a dispute between Libya and Chad, the Court
rejected Libya’s argument that there was no border between them and upheld Chad’s
submission that a 1955 Treaty of Friendship and Good Neighbourliness concluded
between France (which at the time controlled the territory of Chad, known then as
French Equatorial Africa) and Libya had created a boundary line between the parties.
In Land and Maritime Boundary between Cameroon and Nigeria [148] Cameroon
requested the Court to determine the course of the boundary between itself and Nigeria
from Lake Chad to the sea. In so doing the Court established the boundary between
Lake Chad and the Bakassi Peninsula by examining 17 separate sectors of the
boundary, each corresponding to a different historical instrument. In Bakassi, the Court
decided that the Anglo-German Agreement of 11 March 1913 applied to delimit the
area, and that sovereignty over the peninsula accordingly lay with Cameroon.
In Kasikili/Sedudu Island, [149] the Court resolved a dispute between Botswana and
Namibia by interpreting and applying a delimitation treaty concluded in 1890 between
Great Britain and Germany, in terms of which the boundary between these states’
respective spheres of influence in the area of the island was set to run along the ‘main
channel’ of the River Chobe. The Court determined the course of the river’s ‘main
channel’ by reference to studies of the depth and width of the channel and its
navigability and on this basis concluded that the island formed part of the territory of
Botswana.
In Frontier Dispute (Benin/Niger), [150] a Chamber of the International Court resolved
a dispute between Benin and Niger on the basis of a boundary inherited from the
French administration at the time when
Page 207
both Benin and Niger were colonies of France, forming part of French West Africa. The
Chamber determined that for one sector of the frontier area there had been a modus
vivendi between the local authorities of the region during the colonial administration in
terms of which the main navigable channel of the River Niger was regarded as
constituting the intercolonial boundary. In relation to a second sector of the frontier
area the Chamber found that the administrative authorities during the colonial period
had considered that the course of the River Mekrou formed the intercolonial boundary.
The Chamber accordingly held that the boundary between Benin and Niger followed the
median line of that river.
In Frontier Dispute (Burkina Faso/Republic of Niger), [151] the Court again determined
the boundary between the two states by reference to the acts and delimitation
decisions of the French colonial administration prior to their independence.
At present there are boundary disputes involving Namibia arising out of the colonial
era. A secessionist movement in the Caprivi Strip—the Caprivi Liberation Army—argues
that the Caprivi Strip, whose people are ethnically different from the people of Namibia,
did not become part of the German Protectorate of South West Africa as a result of the
1890 Anglo-German Agreement, which simply provided that the territory would fall
within the German sphere of influence. Consequently, the argument goes, the Caprivi
Strip did not become part of Namibia on independence in 1990. This dispute is central
to the trial of members of the Caprivi Liberation Army for treason at present underway
in Namibia. [152] Then there is the dispute with South Africa relating to the boundary
along the Orange River. Whereas the Namibian constitution establishes the boundary
between Namibia and South Africa as running along the middle of the Orange
River, [153] South Africa claims that the boundary runs along the high-water line on the
northern bank in terms of the Anglo-German treaty of 1890. [154]
Page 208

Antarctica

A number of states [155] have made claims to exclusive sovereignty over different parts
of Antarctica. However, these claims, [156] based on discovery, a minimal degree of
occupation, and geographic contiguity, are disputed by some of the rival contenders
and are not recognised by either the United States or Russia. These facts, coupled with
the demand from many quarters that Antarctica should be used for peaceful scientific
research and subjected to rigorous conservation measures, resulted in the 1959
Antarctic Treaty, [157] which, in article 4, ‘freezes’ all territorial claims. It provides that
while nothing in the treaty shall be interpreted as prejudicing the prior claims of states
to territorial sovereignty:
[N]o acts or activities taking place while the present treaty is in force shall constitute a basis
for asserting, supporting or denying a claim to territorial sovereignty in Antarctica or create
any rights of sovereignty in Antarctica. No new claim, or enlargement of an existing claim, to
territorial sovereignty in Antarctica shall be asserted while the present treaty is in force. [158]
The treaty provides for the demilitarisation of Antarctica and envisages that the
continent will be used for peaceful scientific research, carried out in a spirit of
international co-operation with strict regard to ecological protection.
Although South Africa has made no territorial claim to any part of Antarctica, it was
one of the original 12 signatory states to the Treaty [159] because of its special interest
in the region arising from its geographical situation. South Africa maintains a number of
bases in Antarctica for the purpose of scientific research. [160]
It is unlikely that Antarctica will ever become the subject of territorial sovereignty by
states as there is strong resistance, particularly among developing nations, to any
suggestion that the continent should be made subject to national appropriation. Instead
it is suggested either that Antarctica should become a wilderness park,
Page 209
with natural resource exploitation prohibited and with special environmental protection,
or that it should be designated, like the deep seabed, as being the common heritage of
mankind, with the consequence that any revenues obtained from the exploitation of the
region should be shared with developing countries. In 1991 the Antarctic Treaty
members adopted a treaty which prohibits ‘[a]ny activity relating to mineral resources,
other than scientific research’. [161]
The Antarctic Treaties Act of 1996 [162] incorporates the Antarctic Treaty, the Protocol
on Environmental Protection to the Antarctic Treaty (1991), the Convention for the
Conservation of Antarctic Seals (1972) and the Convention on the Conservation of
Antarctic Marine Living Resources (1980) into South African law, [163] and confers
jurisdiction upon the South African executive and judiciary to exercise jurisdiction in
Antarctica over South African citizens and residents and any person ‘who is a member
of or is responsible for organizing an expedition which has been organized in the
Republic to visit Antarctica, but not an expedition organized by the government of
another Contracting Party’, [164] particularly for the purpose of preventing damage to the
Antarctic environment.

[1] In Malaysia/Singapore 2008 ICJ Reports (para 122) the International Court of Justice stressed ‘the
central importance in international law and relations of state sovereignty over territory and of the stability
and certainty of that sovereignty’. See, too, MN Shaw Title to Territory in Africa: International Legal
Issues (1986); M Kohen (ed) Territoriality and International Law (2016); RY Jennings (with a new
introduction by M Kohen) The Acquisition of Territory in International Law (2017) (first published in
1963); MN Shaw The International Law of Territory (2018).
[2] See Chapter 5.
[3] The term ‘sovereignty’ is avoided wherever possible because its meaning varies according to the
discipline and context in which it is used. The meaning of the term in international law was accurately
described by arbitrator Max Huber in the Island of Palmas Case: ‘Sovereignty in the relations between
states signifies independence. Independence in regard to a portion of the globe is a right to exercise
therein, to the exclusion of any other state, the function of a state’ (2 RIAA 829 (1928) at 838). See
further R Jennings ‘Sovereignty and international law’ in G Kreijen (ed) State, Sovereignty, and
International Governance (2002) 27.
[4] See below, notes 164–71.
[5] TW Bennett and CH Powell ‘Aboriginal title in South Africa revisited’ (1999) 15 SAJHR 449; R
Choudree and G McIntyre ‘Resititution of land in two jurisdictions: Australia and South Africa’ (1997)
114 SALJ 187; U Ozlem ‘Developing the doctrine of aboriginal title in South Africa: Source and content’
(2002) 46 Journal of African Law 131.
[6] The concepts of terra nullius and occupation have gained recent attention in the context of the polar
regions: PE Steinberg, J Tasch and H Gerhardt Contesting the Arctic: Politics and Imaginaries in the
Circumpolar North (2015) 18–43; K Dodds, AD Hemmings and P Roberts Handbook on the Politics of
Antarctica (2017) 54–6; C Collins Territories beyond Possession? Antarctica and Outer Space (2017) 7;
CR Rossi ‘A unique international problem: The Svalbard Treaty, equal enjoyment, and terra nullius:
Lessons of territorial temptation from history’ (2016) 15 Washington University Global Studies Law
Review 93; W Erlank ‘Rethinking terra nullius and property law in space’ (2015) 18 Potchefstroom
Electronic Law Journal 2503.
[7] Principles of Public International Law 7 ed (2008) 123. See, too, C Anyangwe ‘African border
disputes and their settlement by international judicial process’ (2003) 28 SAYIL 29; KH
Kaikobad Interpretation and Revision of International Boundary Decisions (2012); J Levitt (ed) Africa:
Mapping New Boundaries in International Law (2008); F Nguendi Ikome ‘Africa’s international borders as
potential sources of conflict and future threats to peace and security’ (2012) 223 Institute for Security
Studies Papers 16.
[8] J Crawford Brownlie’s Principles of Public International Law 8 ed (2012) 223.
[9] See Chapter 24.
[10] Named after the United States Secretary of State and the French Foreign Minister respectively.
[11] Article 2(4).
[12] See below (n 101).
[13] Resolution 68/262 (27 March 2014); TD Grant Aggression against Ukraine: Territory,
Responsibility and International Law (2015); J Bering ‘The prohibition of annexation: Lessons from
Crimea’ (2017) 49 NYUJ Int’l L&Pol 747.
[14] On the application of this rule in Africa, see M van der Linden The Acquisition of Africa (1870–
1914): The Nature of International Law (2017) 246 et seq.
[15] (United States v Netherlands) 2 RIAA 829 (1928).
[16] TO Elias ‘The doctrine of intertemporal law’ (1980) 74 AJIL 285.
[17] Island of Palmas Case (n 15) 845.
[18] Ibid.
[19] P Jessup ‘The Palmas Island arbitration’ (1928) 22 AJIL 735; Jennings (n 1) 42–5.
[20] SCOR, 16th Year, meetings 987 and 988 (18 December 1961). See further DJ Harris Cases and
Materials on International Law 8 ed (2015) 355–7.
[21] Ibid 213–17.
[22] For expositions of South Africa’s claims to Walvis Bay, see PEJ Brooks ‘The legal status of Walvis
Bay’ (1976) 2 SAYIL 187; NJ Botha ‘Walvis Bay: Miscellany’ (1979) 12 CILSA 255; AJ Faris ‘The
administration of Walvis Bay’ (1979) 5 SAYIL 63.
[23] Section 1 of the South West Africa Affairs Act 24 of 1922.
[24] Proclamation R202 GG 5731 of 31 August 1977 (Reg Gaz 2525) issued in terms of the South West
Africa Constitution Amendment Act 95 of 1977.
[25] Walvis Bay: Decolonization and International Law (1990) 161–7. For a critical review of this study,
see J Dugard ‘Walvis Bay and international law: Reflections on a recent study’ (1991) 108 SALJ 82. See
further GP Goeckner and IR Gunning ‘Namibia, South Africa and the Walvis Bay dispute’ (1980) 89 Yale
LJ 903.
[26] Berat (n 25) 164.
[27] Jennings (n 1) 45.
[28] 1978 ICJ Reports 3 at 33–4.
[29] Elias (n 16) 305–6.
[30] Jennings (n 1) 44–5.
[31] Minquiers and Ecrehos (France v United Kingdom) 1953 ICJ Reports 47; Right of Passage over
Indian Territory (Portugal v India) 1960 ICJ Reports 6; Temple of Preah Vihear (Cambodia v
Thailand) 1962 ICJ Reports 6; Western Sahara, Advisory Opinion 1975 ICJ Reports 12; Land, Island and
Maritime Frontier Dispute (El Salvador v Honduras; Nicaragua intervening) 1992 ICJ Reports
351; Maritime Delimitation and Territorial Questions between Qatar and Bahrain 2001 ICJ Reports 40.
[32] Kasikili/Sedudu Island (Botswana v Namibia) 1999 ICJ Reports 1045; Land and Maritime Boundary
between Cameroon and Nigeria (Cameroon v Nigeria: Equatorial Guinea intervening) 2002 ICJ Reports
303.
[33] Dubai-Sharjah Land Border Arbitration (1981) 91 ILR 543; Eritrea and Yemen (1998) 114 ILR 1 at
32–4.
[34] For a challenge to this principle, see DM Ahmed Boundaries and Secession in Africa and
International Law: Challenging Uti Possidetis (2015).
[35] In 1964, at the first session of the Conference of African Heads of State and Government, it was
resolved that all member states of the OAU ‘pledge themselves to respect the borders existing on their
achievement of national independence’: AGH/Res 16(1).
[36] 1986 ICJ Reports 554.
[37] At 565.
[38] At 568.
[39] See further on this conflict GN Barrie ‘Uti possidetis versus self-determination and modern
international law: In Africa the chickens are coming home to roost’ 1988 TSAR 451; FD Mnyongani
‘Between a rock and a hard place: The right of self-determination versus uti possidetis in Africa’ (2008)
41 CILSA 463.
[40] 1986 ICJ Reports 567.
[41] See Chapter 5.
[42] Opinion 3, 92 ILR 170. See, too, Opinion 2, 92 ILR 967. See further M Craven ‘The EC Arbitration
Commission on Yugoslavia’ (1995) 66 BYIL 333.
[43] See J Dugard ‘Secession: Is the case of Yugoslavia a precedent for Africa?’ (1993) 5 African
Journal of International and Comparative Law 163.
[44] See the Independent International Commission on Kosovo The Kosovo Report: Conflict,
International Response, Lessons Learned (2000).
[45] SN Lalonde Determining Boundaries in a Conflicted World: The Role of Uti Possidetis (2002).
[46] 1992 ICJ Reports 355 at 388 para 43. See, too, Frontier Dispute Case (n 40) 566.
[47] J Castellino and S Allen Title to Territory in International Law: A Temporal Analysis (2003).
[48] 2013 ICJ Reports 44 at 73, 84.
[49] Brownlie (n 8) 223; Island of Palmas Case (n 15) 845.
[50] Legal Status of Eastern Greenland 1933 PCIJ Reports Series A/B no 53 at 45–6; Western Sahara
Case 1975 ICJ Reports 12 at 43.
[51] This was done by the early British settlers of West Falkland.
[52] When Argentina took possession of the Falkland Islands in 1821, it advertised this fact in the The
Times of London (PJ Beck The Falkland Islands as an International Problem (Routledge Revivals) (2014)
89; Harris (n 20) 179.
[53] Eastern Greenland Case (n 50) 46; Island of Palmas Case (n 15) 840; Clipperton Island case
(1932) 26 AJIL 390.
[54] 2012 ICJ Reports 624 para 80. See also Sovereignty over Pedra Branca/Pulau Batu Puteh, Middle
Rocks and South Ledge (Malaysia/Singapore) 2008 ICJ Report 12; M Kohen ‘Original title in the light of
the ICJ judgment on sovereignty over Pedra Branca/Pulau Batu Puteh, Middle Rocks and South Ledge’
(2013) 15 Journal of the History of International Law 151.
[55] See RP Schaffer ‘The extension of South African treaties to the territories of South West Africa and
the Prince Edward Islands’ (1978) 95 SALJ 63.
[56] Prince Edward Island was named after the fourth son of King George III.
[57] This was confirmed by an exchange of notes of February 1949: Union of South Africa Treaty
Series no 11 of 1950.
[58] House of Assembly Debates vol 65 col 3041 (22 September 1948).
[59] The Prince Edward Islands Act 43 of 1948. Section 1(1) of the Act is less elegant as it states that
the islands were ‘annexed’ to South Africa. Clearly, this was a case of occupation and not annexation.
[60] For a description of this jurisprudential debate, see MF Lindley The Acquisition and Government of
Backward Territory in International Law (1926); Berat (n 25) at 104–9.
[61] 1975 ICJ Reports 12 at 39. See further Shaw (n 1) 31–8.
[62] Campbell v Hall [1774] 1 Cowper 204; G Carpenter Introduction to South African Constitutional
Law (1987) 18.
[63] See R Balkin ‘International law and sovereign rights of indigenous peoples’ in B Hocking
(ed) International Law and Aboriginal Rights (1988) 19.
[64] Cooper v Stuart [1889] 14 AC 286 at 291. In 1979 Justice Murphy declared that this statement
‘may be regarded either as having been made in ignorance or as a convenient falsehood to justify the
taking of aborigine’s land’ (Coe v Commonwealth of Australia (1979) 24 ALR 118 at 137–8).
[65] Milirrpum v Nabalco Pty Ltd and the Commonwealth of Australia (1971) 17 FLR 141 at 242; Coe v
Commonwealth of Australia (n 64) 129 (per Gibbs CJ).
[66] (1992) 107 ALR 1 at 26–9 (Brennan J), 82–3 (Deane, Gaudron JJ), 141–2 (Toohey J). See further
A Reilly ‘The Australian experience of aboriginal title: Lessons for South Africa’ (2000) 16 SAJHR 512.
In Alexkor Ltd v Richtersveld Community 2004 (5) SA 460 (CC) the Constitutional Court, after
citing Mabo, stated: ‘Courts in other jurisdictions have in recent times been faced with complex and
difficult problems dealing, after the event, with injustices caused by dispossessions of land, or rights in
land, from indigenous inhabitants by later occupiers of the land in question. These later occupiers claimed
political and legal sovereignty over the land, and such dispossessions invariably took place in a racially
discriminatory manner. They often occurred centuries ago, when the legal norms and principles of later
occupiers differed substantially from those of today. In this regard our situation in this country differs
substantially from that of the jurisdictions referred to above in that both our Interim Constitution and the
Constitution [s 25(7)] have dealt expressly with this problem’ (paras 34–5). See, too, Richtersveld
Community v Alexkor Ltd 2003 (6) SA 104 (SCA).
[67] In 1660 Jan van Riebeeck informed a group of Khoikhoi who had been defeated in a skirmish that
they had lost their land in war. It is unlikely, however, that this attitude could be attributed to the Dutch
East India Company. See SFN Gie Geskiedenis van Suid-Afrika Part I (1928) 77–8; L Marquard The Story
of South Africa (1955) 39.
[68] TRH Davenport South Africa: A Modern History 3 ed (1987) at 32–3.
[69] The British entered into numerous treaties with the African tribes on the eastern frontier. For
references to these treaties, see JA Kalley South Africa by Treaty 1806–1986 (1987). The treaties are
published in Treaties Entered into by the Governors of the Colony of the Cape of Good Hope and Other
British Authorities with Native Chieftains between 1803 and 1854 (1857). According to E Brookes The
History of the Native Policy in South Africa from 1830 to the Present Day 2 ed (1927) the policy of the
British in the 1830s was ‘to enter into treaties with the various chiefs on the basis of their independent
sovereignty’ (14). Dr John Philip, the famous missionary, told a Select Committee of the House of
Commons in 1836 that in his opinion ‘[t]he Caffres were quite capable of understanding a system of
international law, and of appreciating it’ (14). See further TW Bennett ‘Aboriginal title in South Africa’
(1993) 9 SAJHR 443; TW Bennett and CH Powell ‘Aboriginal title in South Africa revisited’ (1999)
15 SAJHR 449.
[70] See S Hofmeyr Die Boere-Republieke en die Volkereg (1933) ch 4.
[71] Territories in the Eastern Cape were first annexed by Britain and then incorporated into the Cape
Colony. By 1829, the eastern boundary of the Colony extended to the Keiskamma. In 1847 the land
between the Keiskamma and the Kei was annexed to the Cape Colony as a separate imperial dependency
under the name of British Kaffraria. In 1860 British Kaffraria became a Crown colony and in 1865 it was
incorporated in the Cape Colony (Act 3 of 1865 (C)). Griqualand East, Fingoland and Griqualand West
were annexed to the Cape Colony in 1877 (Acts 38 and 39 of 1877 (C)). Walvis Bay and certain
territories on the St John’s River were incorporated in the Cape Colony by Act 35 of 1884 (C);
Tembuland, Galekaland and Bomvanaland by Act 3 of 1885 (C); and Pondoland by Act 5 of 1894 (C).
[72] The Boer Republic in Natal, established in 1838, was not recognised by Britain, and in 1842 Britain
occupied Natal. In 1843 Natal was formally annexed by the British government, and in 1856 it became a
separate Crown colony. In 1897 Zululand and Tongaland, annexed by Britain in 1887 and 1895
respectively, were incorporated in Natal.
[73] According to Hofmeyr (n 70) it was clear that ‘beide die Republieke die beginsel gehuldig het . . .
dat agterlike volke, solang daar maar ’n rudimentêre gesag oor hulle uitgeoefen word, ’n reg het op die
soewereiniteit oor hulle gebied, geldig teenoor die meer beskaafde volke. Hierdie beginsel het in die
praktyk van die ander state in die reël toepassing gevind, ten minste as teoretiese uitgangspunt. ’n
Verkryging van die soewereiniteit oor ’n naturelle-gebied word dan ook by gebreke aan toestemming van
die naturelle, deur die state as reël op verowering gegrond’ (145). Cf CC Eloff ‘Lesotho claims to part of
the Orange Free State’ (1978) 4 SAYIL 108, who states that the Boers acquired title to the triangle of
land between the Orange and the Caledon rivers by occupation as it was terra nullius (114). The eastern
part of the Orange Free State, known as the ‘Conquered Territory’, was acquired by conquest and cession
after wars with the Basuto, confirmed by the Treaty of Aliwal North of 1869.
[74] Berat (n 25) 30–5.
[75] Ibid 115.
[76] Ibid 36–7.
[77] Walfish Bay and St Johns River Annexation Act 35 of 1884 (C).
[78] Berat (n 25) 121. H Booysen supports the view that Walvis Bay was acquired by
occupation: Volkereg en sy Verhouding tot die Suid-Afrikaanse Reg (1989) 198–9.
[79] Land Boundary in the Northern Part of Isla Portillos (Costa Rica v Nicaragua) judgment of 2
February 2018.
[80] GW Eybers Select Constitutional Documents Illustrating South African History 1795–1910 (1918)
19; HR Hahlo and E Kahn South Africa: The Development of its Laws and Constitution (1960) 5. Similarly,
Alsace-Lorraine was ceded by Germany to France in the Treaty of Versailles (1919). The Island of Palmas
Case (n 15) concerned a post-war treaty of cession of territory from Spain to the USA.
[81] For further historical examples of cession, see R Jennings and A Watts (eds) Oppenheim’s
International Law 9 ed (2008) vol 1 at 680 et seq.
[82] The text of the treaty is published in GN R951 GG 1574 of 20 May 1994 (Reg Gaz 5338) and
(1994) 33 ILM 1526. See, too, Transfer of Walvis Bay to Namibia Act 203 of 1993. See further J Dugard
‘Public international law’ in 1993 Annual Survey 61 at 72 and 1994 Annual Survey 98 at 99. The history
of Walvis Bay is described above at notes 22 to 25. The Off-Shore islands of Hollams Bird, Mercury,
Ichaboe, Long, Seal, Penguin, Halifax, Possession, Albatross Rock, Pomona, Plum-pudding and Roast Beef
(or Sinclair’s Island) were annexed in 1874 to the Cape, which administered them until 1994. See further
Berat (n 25) at 195–208.
[83] Written statement of Bolivia on the preliminary objection of Chile (7 November 2014) para 27. See
judgment of 1 October 2018.
[84] L Buchheit Secession: The Legitimacy of Self-Determination (1978). This view finds support in
Grotius De Jure Belli ac Pacis 2.6.4.
[85] The obvious goal of this proposed cession was to transfer a substantial portion of the South African
black population to a foreign state and thereby reduce the number of black people within South Africa’s
borders. The Minister of Foreign Affairs, Mr RF Botha, sought, however, to justify it as an historical border
readjustment that would consolidate the Swazi people within one territory: House of Assembly
Debates cols 6252–8 (6 May 1982). See further N Marais ‘The Swaziland border dispute’ (1982)
8 SAYIL 248–59.
[86] When it became clear that both KaNgwane and KwaZulu were opposed to the cession, the matter
was referred to two judicial commissions under the chairmanship of Mr Justice FLH Rumpff. The
commissions were discontinued in 1984 on the ground that the people in the territories in question were
either opposed to the cession or could not freely express their views in the prevailing political climate
(1984 Race Relations Survey 505–8). The proposal was then abandoned. See further on this: House of
Assembly Debates cols 145–8 (1 February 1983); M Beukes ‘Oor die kwessie van grensaanpassings met
Swaziland’ (1983) 24 Codicillus (May) 19.
[87] On the ground that it would violate article 53 of the Vienna Convention on the Law of Treaties,
which renders a treaty void if it conflicts with a peremptory norm of international law—a status for which
the right of self-determination certainly qualifies. See Chapter 5.
[88] AJGM Sanders ‘The “Swaziland Deal” and the international law principle of the self-determination
of peoples’ (1983) 24 Codicillus (Oct) 34.
[89] The text of the annexation proclamation appears in Eybers (n 80) 344.
[90] Ibid 514.
[91] Oppenheim (n 81) 700, fn 2; T Baty International Law in South Africa (1900) 90–3; Hofmeyr (n
70) 62–8.
[92] 1903 TS 401.
[93] The exchange between Smuts and the Court is fascinating as Smuts carefully refrained from
asserting his own personal knowledge of the history of the war (403–7).
[94] At 409–10. See, too, the judgments of Mason and Bristowe JJ at 419 and 424, respectively.
[95] Article 2(4).
[96] Military and Paramilitary Activities in and against Nicaragua (Nicaragua v USA) 1986 ICJ Reports
14 at 99–100.
[97] See above. Jennings (n 1) 69, 72. See, too, Bolivia v Chile (n 83).
[98] S Schwebel ‘What weight to conquest?’ (1970) 64 AJIL 344; M Halberstam ‘Recognition, use of
force and the legal effect of UN resolutions under the revised restatement of the foreign relations law of
the United States’ (1984) 19 Israel Law Review 495 at 503–8. Cf Oppenheim (n 81) 703.
[99] Jennings (n 1) 70–2; DW Bowett ‘International law relating to occupied territory: A rejoinder’
(1971) 87 LQR 473.
[100] Resolution 2625 (XXV).
[101] Resolutions 242 (1967), 252 (1968), 298 (1971), 476 (1980), 478 (1980), 497 (1981).
[102] For example, Resolutions 34/70 (1979), ES/72 (1980), 37/123A (1982), 39/146A (1984).
[103] See J Dugard Recognition and the United Nations (1987) 111–15, 155–6.
[104] A/RES/71/99 (6 December 2016).
[105] See TD Grant ‘Annexation of Crimea’ (2015) 109 AJIL 68; J Bering ‘The prohibition on
annexation: Lessons from Crimea’ (2017) 49 NYU JIL&Pol 747.
[106] GA Res 68/262 (27 March 2014). See also GA Res 71/205 (19 December 2016) and GA Res
72/190 (19 December 2017).
[107] 1999 ICJ Reports 1045 at 1103–4. In this case, Namibia argued that the Masubia tribe of
Namibia had continuously and exclusively occupied the island with the full knowledge of Botswana. The
Court found it unnecessary to pronounce on the subject of prescription as it had not been established that
the members of the tribe had exercised functions of state authority on the island on behalf of Namibia
(South West Africa): 1999 ICJ Reports 1101–6.
[108] Jennings (n 1) 36.
[109] Island of Palmas Case (n 15) 839.
[110] 1933 PCIJ Reports Series A/B no 53 at 45.
[111] 2002 ICJ Reports 625 paras 134–9.
[112] 2012 ICJ Reports 624 at 655.
[113] MN Shaw treats the two modes of acquisition under the heading ‘the exercise of effective
control’: International Law 8 ed (2017) 379.
[114] C de Visscher Theory and Reality in Public International Law (1968) 209. See the criticism of this
suggestion by Jennings (n 1) 24–8. The International Court of Justice expressed doubts about historical
consolidation as a basis of title in territorial disputes in Land and Maritime Boundary between Cameroon
and Nigeria 2002 ICJ Reports 303 at 352.
[115] Minquiers and Ecrehos Case 1953 ICJ Reports 47. Brownlie (n 8) 223–6.
[116] Brownlie (n 8) 228.
[117] 2008 ICJ Reports 12 paras 120–1.
[118] Jennings (n 1) 30.
[119] Harris (n 20) 179–82; PJ Beck The Falkland Islands as an International Problem (Routledge
Revivals) (2014) 61–84 (detailing the Argentinian claim).
[120] See CC Eloff ‘Lesotho claims to part of the Orange Free State’ (1978) 4 SAYIL 109; I
Brownlie African Boundaries: A Legal and Diplomatic Encyclopaedia (1979) 1108.
[121] Eloff (n 120) 109–10.
[122] The Law of Treaties (1961) 485. See further I Sinclair ‘Estoppel and acquiescence’ in AV Lowe
and M Fitzmaurice (eds) Fifty Years of the International Court of Justice (1996) 104.
[123] 1962 ICJ Reports 6 at 30–2.
[124] Op cit (n 25) 167–73.
[125] Op cit (n 1) 67.
[126] DW Bowett ‘Estoppel before international tribunals and its relation to acquiescence’ (1957)
33 BYIL 176 at 202.
[127] 1975 ICJ Reports 12.
[128] 1975 ICJ Reports 12 at 40. In his separate opinion in the Western Sahara Case, Judge Petrén
stated: ‘The question of the extent to which, and under what conditions, past legal ties may influence the
decolonization of a territory seems to me to fall within an as yet inadequately explored area of
contemporary international law’ (112).
[129] See the attempts to apply this argument to Walvis Bay, Berat (n 25) 155–60, Goeckner and
Gunning (n 25).
[130] See, on these claims, RV Pillai and M Kumar ‘The political and legal status of Kuwait’ (1962)
11 ICLQ 108.
[131] MN Shaw Title to Territory in Africa (1986) 134.
[132] J Crawford The Creation of States in International Law 2 ed (2006) 624, 647.
[133] Shaw (n 131) 137–40.
[134] Resolution 432 (1978) of the Security Council declared that ‘the territorial integrity and unity of
Namibia must be assured through the reintegration of Walvis Bay within its territory’.
[135] Island of Palmas Case (n 15) 839.
[136] This tribunal of five arbitrators included three judges who served as President of the International
Court of Justice: RY Jennings, S Schwebel and R Higgins.
[137] 22 RIAA 209 (1998) 239, 114 ILR 1 at 69.
[138] 2008 ICJ Reports 133 paras 120–1, 273–6.
[139] Ibid paras 36–43 (dissenting opinion of J Dugard).
[140] Supra (n 137).
[141] Land, Island and Maritime Frontier Dispute (El Salvador)/Honduras; Nicaragua Intervening 1992
ICJ Reports 351 at 389.
[142] I Brownlie African Boundaries: A Legal and Diplomatic Encyclopedia (1979); N Ikome, ‘Africa’s
international borders as potential sources of conflict and future threats to peace and security’ (2012)
223 Institute for Security Studies Papers 16.
[143] M van der Linden The Acquisition of Africa (1870–1914): The Nature of International Law (2017).
[144] G Oduntan International Law and Boundary Disputes in Africa (2015).
[145] OAU Resolution AHG/Res 16 (I) (July 1964); Constitutive Act of the African Union article 4(b)
(entered into force on 26 May 2001).
[146] 1986 ICJ Reports 554.
[147] 1994 ICJ Reports 6.
[148] 2002 ICJ Reports 303.
[149] 1999 ICJ Reports 1045.
[150] 2005 ICJ Reports 90.
[151] 2013 ICJ Reports 44.
[152] Munuma v S (CC 03/2004) [2018] NAHCMD 142 (29 May 2018).
[153] Article 1(4) of the Constitution of the Republic of Namibia.
[154] See G Erasmus and D Hamman ‘Where is the Orange River Mouth? The demarcation of the South
African/Namibian maritime boundary’ (1987–88) 13 SAYIL 49.
[155] Argentina, Australia, Chile, France, New Zealand, Norway and the United Kingdom.
[156] See MT De Quintal ‘Sovereignty disputes in the Antarctic’ (1984) 10 SAYIL 161; DR Rothwell
‘Polar territorial and maritime sovereignty in the twenty-first century’ in C Chinkin and F
Baetens Sovereignty, Statehood and State Responsibility (2017) 110.
[157] See GN Barrie ‘The Antarctic Treaty: Example of law and its sociological infrastructure’ (1975)
8 CILSA 212; GN Barrie ‘The Antarctic Treaty forty years on’ (1999) 116 SALJ 173.
[158] The text of the treaty appears in 402 UNTS 71.
[159] The seven states listed in n 155 and Belgium, Japan, South Africa, the USSR and the USA.
[160] See Barrie ‘The Antarctic Treaty forty years on’ (n 157). IJ Ansorge et al ‘Exploring South Africa’s
southern frontier: a twenty year vision for polar research’ (2017) 113 South African Journal of Science 2.
[161] Protocol on Environmental Protection to the Antarctic Treaty (1991) 30 ILM 1455 (article 7).
[162] Act 60 of 1996.
[163] Section 3.
[164] Section 2.

Page 210
 Chapter 9
Jurisdiction and International Crimes

John Dugard, Max du Plessis and Eshed Cohen

Jurisdiction

Jurisdiction is an important aspect of sovereignty. [1] Sovereignty empowers a state to

exercise the functions of a state within a particular territory to the exclusion of other
states. [2] Jurisdiction is the branch of law that defines these functions. The term
therefore refers to the authority that a state has to exercise its governmental functions
by legislation, executive and enforcement action, and judicial decrees over persons and
property. In most circumstances the exercise of the functions of a state is limited to the
territory of the state. [3]
South Africa, like other states, zealously guards against any attempt on the part of
other states to exercise their governmental functions within its territorial limits. Foreign
police officers may not make arrests in South Africa and foreign governments may not
enforce their sovereign acts through South African courts. [4] Any intervention in the
domestic affairs of South Africa by other states or international organisations will be
resisted as a violation of the prohibition on foreign intervention that receives
recognition in article 2(7) of the Charter of the United Nations. [5]
If all states confined the exercise of their governmental functions to their own
territories, and jurisdiction was entirely territorial, this topic would require little
examination. This, however, is not the case. International trade, migration, travel and
crime ensure that states will have an interest in extending their jurisdiction beyond
their
Page 211
territorial limits to cover persons and property in other countries. The exercise of civil
jurisdiction by one state over persons or property in another state, in accordance with
the rules of private international law, [6] seldom elicits protest from the state in which
the person or property is situated. [7] The same cannot be said about the application of
the criminal law to persons, property and events in other countries. This frequently
becomes a matter of public debate and the territorial state will be compelled to protest
against the exercise of criminal jurisdiction over matters that fall within its exclusive
territorial jurisdiction. For this reason, most of the rules relating to the exercise of
domestic jurisdiction in international law relate to criminal offences.
The starting point for any discussion of jurisdiction is to distinguish between the two
aspects of a state’s domestic criminal jurisdiction. First, there is prescriptive
jurisdiction. [8] Prescriptive jurisdiction refers to a state’s authority to criminalise certain
conduct. [9] A state exercises prescriptive jurisdiction when it renders applicable—
though it does not necessarily apply—its criminal law to certain conduct by criminalising
the conduct concerned. So when state A, in law X, says that murder is a crime, it
exercises prescriptive jurisdiction. [10] Secondly, there is enforcement
jurisdiction. [11] Enforcement jurisdiction is a state’s ability to apply and enforce its
criminal laws on certain conduct. This enforcement can occur by (inter alia)
Page 212
arresting, trying, sentencing and/or punishing a person for engaging in certain
conduct. [12] So when state A, by law X, arrests, detains and (assuming conviction)
incarcerates person P for murder, then state A exercises enforcement jurisdiction.
These two aspects of jurisdiction are independent. [13] A state can exercise prescriptive
jurisdiction without exercising enforcement jurisdiction over certain conduct. [14]
The Lotus Case is the seminal case on the international law regulating the
extraterritorial exercise of prescriptive and enforcement jurisdiction. [15] In this case, a
French ship, the Lotus, collided with a Turkish ship, the Boz-Kourt, on the high seas.
The latter ship sank and a number of crew members and passengers lost their lives.
The Lotus picked up the survivors and put into port in Turkey. Here, the officer of the
watch on board the Lotus at the time of the collision was arrested, tried and convicted
of culpable homicide. France objected to Turkey’s exercise of jurisdiction and the
dispute was referred to the Permanent Court of International Justice. Before the Court,
France argued that only the flag-state had jurisdiction over acts committed on board a
vessel on the high seas, while Turkey claimed that it had jurisdiction by reason of the
fact that the effects of the collision had been felt on a Turkish ship, which was to be
viewed as part of Turkish territory.
In its judgment, the Court expounded the following principles of jurisdiction:
(a)
A state ‘may not exercise its power in any form in the territory of another state’—
unless there is a permissive rule to the contrary. [16]
(b)
International law does not prohibit a state ‘from exercising jurisdiction in its own
territory, in respect of any case which relates to acts which have taken place
abroad’. States have ‘a wide measure of discretion’ to extend the application of
their
Page 213
laws and the jurisdiction of their courts to persons, property and acts outside
their territory, ‘which is only limited in certain cases by prohibitive rules’. [17]
(c)
’The territoriality of criminal law, therefore, is not an absolute principle of
international law.’ [18]
The Court held, by the casting vote of the President, that as no rule of international law
prohibited Turkey from trying a person for an offence that had produced effects on a
Turkish vessel, and hence within Turkey itself, Turkey had not violated international
law. In other words, a state can only exercise enforcement jurisdiction extraterritorially
if there is an international law allowing it to do so. The default position is that
enforcement jurisdiction is territorial only. Hence, the Court explained:
The first and foremost restriction imposed by international law upon a State is that—failing the
existence of a permissive rule to the contrary—it may not exercise its power in any form in the
territory of another State. In this sense jurisdiction is certainly territorial; it cannot be
exercised by a State outside its territory. [19]
On the other hand, prescriptive jurisdiction can be exercised extraterritorially, unless
there is a rule in international law prohibiting a state from doing so.
The Lotus decision has been widely criticised, [20] particularly because it empowers
states to exercise prescriptive jurisdiction over acts occurring outside their territory—
except where there is proof of a rule of international law prohibiting such prescription.
 Although the principle of the Lotus Case that a state may exercise prescriptive
jurisdiction over acts occurring abroad in the absence of a prohibitory rule remains
unchanged, in practice states have sought to limit the exercise of extraterritorial
prescriptive jurisdiction to cases in which there is a direct and substantial connection
between the state exercising jurisdiction and the matter in question. [21] Such
connections have been called ‘heads’ or ‘principles’ on which a state can extend
prescriptive jurisdiction extraterritorially. [22] Presumably,
Page 214
states have relied on these heads to confine the exercise of their extraterritorial
prescriptive jurisdiction within reasonable limits. So states generally restrict the
exercise of jurisdiction to matters committed within their territories or having an effect
within their territories, to matters affecting their nationals, or to acts threatening their
security. This list is not exhaustive.
So, for example, state A may make it a crime to murder any of its citizens anywhere
in the world. This would permissibly extend prescriptive jurisdiction beyond the territory
of state A on the basis of passive personality (discussed below). However, should state
A proscribe smoking anywhere in the world, it would not be extending its prescriptive
jurisdiction in tune with state practice. It would not be relying on any of the principles
of extraterritoriality discussed below, although this is, according to the Lotus Case, still
legal.
In S v Basson [23] the Constitutional Court held that a South African court might
exercise criminal jurisdiction over a conspiracy entered into in South Africa by members
of the South African Defence Force to murder opponents of the South African
administration in Namibia during South Africa’s occupation of that territory. This was on
the ground that there was a ‘real and substantial link’ between South Africa and the
crime—though, this comment was obiter because the Court ultimately found that South
Africa had jurisdiction on the ground of territoriality because at all relevant times
Namibia was under South African occupation. [24]
The various heads on which prescriptive jurisdiction can be extended
extraterritorially are discussed below. It must be emphasised that enforcement
jurisdiction cannot be extended on any of these grounds extraterritorially. So, for
example, even though state A can criminalise murder committed against its nationals
anywhere in the world, [25] it cannot unilaterally enforce that law (for example, by
attempting to detain the accused) when the accused is in another state’s territory.
Instead, the state seeking to prosecute the accused should resort to extradition
proceedings. [26]

1 Territoriality

‘From the standpoint of international law, the jurisdictional competence of a state is

primarily territorial,’ said the European Court of
Page 215
Human Rights in Bankovic v Belgium. [27] From this it follows that a state may assert
both aspects of its jurisdiction over all criminal acts that occur within its territory and
over all persons responsible for such criminal acts, whatever their nationality. [28] In
South Africa, as in other countries influenced by the Anglo-American common law, this
is the principal basis for the exercise of prescriptive and enforcement jurisdiction. This
is reflected in the presumption against the extraterritorial operation of criminal laws. [29]
 South African ‘territory’, for the purposes of criminal jurisdiction, includes South
African territorial waters [30] and airspace. Furthermore, any offence committed on
board a South African ship [31] on the high seas or in a South African aircraft [32] above
the high seas or foreign territory may be tried in a South African court. As theft is a
continuing offence under South African law, a court will exercise jurisdiction over a
person who possesses property stolen in another state. [33] In S v Basson, the
Constitutional Court suggested that a territory occupied by South Africa and deemed to
be part of South Africa by an Act of Parliament will give South Africa territorial
jurisdiction in respect of crimes committed there. [34] In certain circumstances, a South
African court will refuse to exercise jurisdiction over a crime committed within South
African territory. Foreign diplomats
Page 216
are granted immunity from the jurisdiction of municipal courts; [35] a court will not try a
person who has been brought before it as a result of an unlawful abduction from
another state; [36] and a court will probably refuse jurisdiction over a person wanted for
a non-political crime committed in South Africa who is on board a ship or aircraft forced
to enter South Africa in distress. [37]

2 Subjective and objective territoriality

A state may exercise jurisdiction where the crime is commenced within its territory and
completed in another state (subjective territoriality) or where the crime is commenced
within a foreign state and completed within its territory (objective
territoriality). [38] Thus, if a gunman standing in state A shoots and kills his victim in
state B, both state A (under the principle of subjective territoriality) and state B (under
the principle of objective territoriality) will have jurisdiction over the gunman. An
extension of objective territoriality is to be found in the ‘effects’ principle, according to
which the state in which the effect or impact of the crime is felt may exercise
jurisdiction. This was the basis upon which Turkey exercised jurisdiction in the Lotus
Case. The effect of the collision was felt on the Turkish ship, which the Court held to be
assimilated to Turkish territory.
In Zimbabwe the ‘effects’ principle was invoked in the bizarre case of S v
Mharapara [39] in which an ex-Zimbabwean diplomat was convicted of theft from the
Zimbabwe government committed while he was in the Zimbabwe diplomatic mission in
Belgium. In dismissing the accused’s appeal, Gubbay JA said:
[A] strict interpretation of the principle of territoriality could create injustice where the
constituent elements of the crime occur in more than one state or where the locus commissi is
fortuitous as far as the harm flowing from the crime is concerned . . . A more flexible and
realistic approach based on the place of impact, or of intended impact, of the crime must be
favoured. [40]
Although this is a necessary extension of the territoriality principle, it may lend itself to
abuse—as illustrated by the manner in which
Page 217
the United States has sought to implement its anti-trust legislation, which has both civil
and criminal law features. [41]
The United States anti-trust laws, notably the Sherman Act of 1890, seek to ensure
competition in commerce by prohibiting monopolistic practices. Relying on the objective
territoriality or ‘effects’ principle, United States courts have given these laws
extraterritorial effect [42] to extend to alleged monopolistic agreements governing trade
and commerce abroad whose effect, however remote, is felt in the United States. Most
countries object strongly to this application of the ‘effects’ principle to anti-competitive
conduct that is not seen as punishable outside the United States. Moreover, it is argued
that the United States practice violates international law on the ground that it exceeds
the permissible limits of extraterritorial jurisdiction. [43] Although more recent judicial
decisions in the United States have sought to modify the impact of the ‘effects’ doctrine
by requiring United States courts to consider the interests of foreign nations, [44] the
anti-trust laws continue to intrude on the commercial activities of foreign
corporations. [45] In the 1970s the Westinghouse Electric Corporation of the United
States filed suit in a United States court against a number of foreign uranium producers
in which it alleged that there was price-fixing in violation of the Sherman Act. Included
among the defendants were Anglo-American Corporation and Nuclear Fuel Corporation,
a subsidiary of Anglo-American. In response to these proceedings, the United
Kingdom, [46] Canada, Australia and South Africa enacted legislation aimed at
frustrating or blocking the enforcement of the anti-trust laws by prohibiting compliance
with
Page 218
United States judicial decrees requesting inspection of documents or evidence located
within their territories.
The South African legislation is contained in the Protection of Businesses
Act. [47] Section 1 provides that, except with the permission of the Minister of Economic
Affairs, ‘no judgment, order, direction, arbitration award, interrogatory, commission
rogatoire, letters of request or any other request delivered, given or issued or
emanating from outside the Republic’ shall be enforced in South Africa if it arises from
an act ‘connected with the mining, production, importation, exportation, refinement,
possession, use or sale of or ownership to [sic] any matter or material, of whatever
nature, whether within, outside, into or from the Republic’. Section 1A prohibits the
recognition and enforcement of judgments that provide for multiple or punitive
damages—a clear reference to the United States anti-trust legislation, which permits
treble damages in anti-trust proceedings. Section 1B provides for the recovery within
South Africa by South African residents of the punitive or multiple damages award of a
foreign court already paid outside South Africa.
South African law on price fixing extends to acts that have an effect within the
Republic. In American Natural Soda Ash Corporation v Competition Commission [48] the
Supreme Court of Appeal interpreted s 3(1) of the Competition Act, [49] which extends
price fixing ‘to all economic activity within, or having an effect within, the Republic’ to
mean both benign and malign effects. [50] It rejected the argument that it extends only
to acts with negative or deleterious effects within the Republic. [51]

3 Protection of the state

A state may exercise prescriptive jurisdiction over aliens who have committed acts
abroad that are considered prejudicial to its safety and security. The most famous
example of a state doing this was in Attorney-General of the Government of Israel
v Eichmann. There, the District Court of Jerusalem held that Israel had jurisdiction over
the crimes committed by Adolf Eichmann because his crimes had, by
Page 219
definition, been prejudicial to the safety and security of the Jewish nation. [52] The
House of Lords has also endorsed the protective principle. [53] In R v
Neumann, [54] Murray J stated, in the trial of an alien resident who had committed acts
of treason against South Africa abroad, that as South Africa was a sovereign state, it
was ‘automatically entitled to punish crime directed against its independence and
safety’. Aliens tried in this way must have some connection with South Africa, usually in
the form of residence. Were this restraint not adopted, it would give rise to an abuse of
extraterritorial powers and to protest from other states.

4 Nationality

Many countries, particularly those with a civil-law tradition, criminalise certain conduct
committed by their own nationals abroad. Thus, state A may punish its national for the
crime of murder committed in state B, where the victim was a national of state B, once
state A’s national is within state A’s territory. This is known as the exercise of
jurisdiction on the ground of ‘active nationality’. In some civil-law countries, for
example in Germany and Sweden, jurisdiction exists even when the accused became a
national after the alleged crime. [55]
Countries influenced by the Anglo-American common law will not exercise
prescriptive jurisdiction on this ground unless the municipal law clearly confers
jurisdiction. [56] In S v Mharapara, described above, the trial judge exercised
prescriptive jurisdiction on the ground of nationality, holding that ‘a state has
jurisdiction with respect to any crime committed outside its territory by a person or
persons who is or are its nationals at the time when the offence was committed or
Page 220
when he is or they are prosecuted and punished’. [57] On appeal, while upholding the
conviction of the accused in accordance with the ‘effects’ principle, Gubbay JA rejected
nationality as a basis for jurisdiction. He stated:
[T]here is no rule of international law directing or obliging states to exercise criminal
jurisdiction over their nationals for offences committed abroad. International law
merely permits every state to apply its jurisdiction against its own citizens even when they are
situate outside its boundaries. . . . Thus the fact that customary international law is part of the
municipal law of a state does not assist, because there is only a permissive principle involved
and not a mandatory rule. The permissibility under international law for a state to exercise
jurisdiction is not a sufficient basis for the exercise of jurisdiction by a municipal court of that
state. A municipal court must be satisfied in addition that the municipal law itself authorises
the trial of a national for an offence committed abroad which would be punishable if committed
at home. [58]
South Africa, like other common-law countries, treats treason as an exception to the
rule that it will not exercise extraterritorial prescriptive jurisdiction on grounds of
nationality. In R v Holm; R v Pienaar [59] the Appellate Division noted how not all
countries criminalise certain conduct of nationals committed abroad. For example,
England only prosecutes its nationals for crimes committed abroad if there is legislation
specifically providing for this. [60] Watermeyer CJ seemed to incorporate this into South
African law when he held that treason is an exceptional crime in the sense that South
Africa can prosecute its nationals for it wherever that crime occurred:
[S]o far as high treason committed by a subject is concerned, there exists no international
custom or comity which debars a state from trying and punishing the offender no matter where
the offence has been committed. The reason for this is clear, it is because high treason,
committed outside of the territory of the state concerned, is an offence only against such state.
 No other state is interested in punishing the offender and the punishment of the offender by
the state concerned does not encroach upon the rights of other states. [61]
Ordinarily, it would seem, crimes committed abroad by South Africans would not fall
within South Africa’s jurisdiction unless it was treason or authorised by statute.
Page 221
The Constitutional Court in S v Basson apparently endorsed this approach. [62] There,
the Court held that South Africa could exercise its jurisdiction over South Africans who
entered into a conspiracy to commit crimes in Namibia while the latter was under
occupation by South Africa. The court said that jurisdiction could be exercised not
because of the nationality of the accused but because of a ‘real and substantial link’
between the two countries on the basis of territoriality. [63]
The Court’s express reliance on a real and substantial territorial link, [64] and not on
nationality, suggests that South African law does not extend jurisdiction to crimes
committed by nationals abroad. Instead, a statute will have to authorise the exercise of
jurisdiction in respect of nationality—for example, as in the Implementation of the
Rome Statute of the International Criminal Court Act. [65] Hence, the Court accepted
that as a general proposition, South African courts ‘have declined to exercise
jurisdiction over persons who commit crimes in other countries. This, as Dugard points
out, is an aspect of sovereignty which has given rise to a presumption against the
extraterritorial operation of criminal law’. [66]

5 Passive personality

This principle allows a state to exercise prescriptive jurisdiction over a person who
commits an offence abroad that harms one of its own nationals. In the past, Anglo-
American countries objected strongly to this basis of jurisdiction. [67] In recent times,
however, this jurisdictional ground has been invoked in order to suppress international
terrorism. [68] Following the killing of an American national by terrorists on an Italian
ship (the Achille Lauro) on the high seas in 1985, the United States enacted legislation
to give its courts jurisdiction to try anyone
Page 222
who kills or intentionally causes serious bodily injury to a national of the United States
outside the United States where the offence ‘was intended to coerce, intimidate, or
retaliate against a government or a civilian population’. [69] In United States v Yunis (No
2) [70] a United States district court invoked passive personality as a basis for exercising
jurisdiction over a Lebanese national who hijacked a Jordanian aircraft with United
States nationals on board and flew the aircraft over a number of Mediterranean
countries.
Although the Rome Statute of the International Criminal Court does not confer
jurisdiction on the Court on the basis of passive personality vis-à-vis nationals of state
parties, the South African statute implementing this treaty empowers a South African
court to exercise jurisdiction in respect of genocide, crimes against humanity and war
crimes committed outside South Africa where the accused person has committed the
crime ‘against a South African citizen or against a person who is ordinarily resident in
the Republic’. [71]

6 Universal jurisdiction and international crimes [72]

The principles discussed above empower a state to exercise prescriptive jurisdiction
over a crime committed abroad in violation of its own national laws. Some conduct
violates not only the domestic legal order of a state but also the international order.
Such conduct constitutes an international crime. Ideally such crimes should be tried by
an international court, but before 1998, a permanent international criminal court
seemed impossible. In 1998 a diplomatic conference in Rome gave its approval to the
establishment of such a court and in 2002 the International Criminal Court came into
being. [73] But the Rome Statute of the International Criminal Court (its founding
statute) operates on the basis of complementarity and largely leaves it to the national
courts of states to enforce international criminal law, either by trying offenders
themselves or by extraditing them to countries that will do so. The principle of aut
dedere aut judicare (punire)—extradite or try (punish)—is the basis for
Page 223
the enforcement of international criminal law. [74] When a national court exercises
prescriptive jurisdiction in this way over an international crime with which it has no
jurisdictional link of the kind described above (in sections 1 to 5) it is said that it
exercises universal jurisdiction. Here, the national court acts as the agent of the
international community in the prosecution of an enemy of all mankind in whose
punishment all states have an equal interest. [75]
Seen in this way, universal jurisdiction is defined in the negative. It is when a state
exercises prescriptive jurisdiction without there being one of the connections listed
above. Defined positively then, universal jurisdiction is prescriptive jurisdiction over
offences committed abroad by persons who, at the time of commission, are non-
resident aliens, where such offences are not deemed to constitute threats to the
fundamental interests of the prescribing state [76] or give rise to effects within the
prescribing state’s territory. [77]
Some argue that universal jurisdiction cannot exist within a treaty because then
jurisdiction is circumscribed by pacta tertiis and cannot be exercised by ‘any
state’. [78] Instead, true universal jurisdiction applies only in the case of crimes under
customary international law, in respect of which all states have the right to prosecute.
Such crimes are limited to piracy, slave-trading, war crimes, crimes against humanity,
genocide and torture. However, in recent years a number of international crimes have
been created by multilateral treaties, which confer wide jurisdictional powers upon state
parties. These treaties confer universal jurisdiction in that signatory states are required
to prosecute or extradite persons who are accused or convicted of the treaty’s crimes
and who happen to be present in their territory. Because the exercise of prescriptive
jurisdiction in such cases does not invoke one of the jurisdictional connections listed
above, this is universal jurisdiction. [79] The fact that states not party to the relevant
treaty have not proscribed the conduct in question
Page 224
makes no difference to how state parties are criminalising conduct notwithstanding the
absence of a jurisdictional nexus. [80]
It must be emphasised that international law permits states to exercise jurisdiction
over international crimes. It does not compel them to do so in the absence of a treaty
obligation. [81] Moreover, most states, including South Africa, will not try a person for an
international crime unless the conduct has been criminalised under municipal law. [82]
Universal jurisdiction has in recent times become an issue of
controversy. [83] Individual states have enacted legislation to give universal jurisdiction
to their own courts to try international crimes recognised by the Rome Statute—
genocide, crimes against humanity and war crimes. For example, the Implementation of
the Rome Statute of the International Criminal Court Act empowers a South African
court to exercise jurisdiction over a person who has committed such crimes outside
South Africa if ‘that person, after the commission of the crime, is present in the
territory of the Republic’. [84] No jurisdictional nexus is required to proscribe the conduct
in question.
Some national statutes do not limit universal jurisdiction to the exercise of universal
jurisdiction over persons present in their
Page 225
territory and go further by allowing their courts to issue warrants for the arrest of
persons outside their own territory. The question whether international law permits the
exercise of universal jurisdiction by national courts over persons in absentia arose in a
dispute between the Democratic Republic of Congo (DRC) and Belgium in the Arrest
Warrant case.
In the Arrest Warrant case, [85] a Belgian judge issued a warrant in absentia for the
arrest of the Foreign Minister of the DRC, charging him with crimes against humanity
and grave breaches of the Geneva Conventions of 1949 arising out of acts committed in
the DRC. The warrant was issued in terms of a Belgian statute conferring jurisdiction on
Belgian courts to try such crimes ‘wheresoever they may have been committed’. The
DRC challenged the legality of the arrest warrant before the International Court of
Justice on two grounds: first, that the issue of the warrant constituted an ‘exercise of
an excessive universal jurisdiction’ and, secondly, that the Minister was entitled to
immunity from prosecution before a Belgian court.
The International Court upheld the second challenge, which made it unnecessary for
the Court to decide on the question of universal jurisdiction. [86] Several judges did,
however, give separate opinions which dealt with this topic. Judge Guillaume, President
of the Court, found that international law does not recognise universal jurisdiction,
except for the crime of piracy, and that universal jurisdiction is ‘unknown to
international law’. [87] Judges Higgins, Kooijmans and Buergenthal, in a separate but
joint opinion, stated that universal jurisdiction may be exercised ‘only over those crimes
regarded as the most heinous by the international community’ [88]—such as piracy, war
crimes, crimes against humanity and genocide—and that ‘a State may choose to
exercise a universal criminal jurisdiction in absentia’ [89] as there is no rule of
international law that prohibits this. Judge ad hoc Van den Wyngaert likewise held that
‘there is no conventional or customary international law or legal doctrine in support of
the proposition that (universal) jurisdiction for war crimes and crimes
Page 226
against humanity can only be exercised if the defendant is present on the territory of
the prosecuting state’. [90]
The Belgian statute conferring universal jurisdiction has since been substantially
amended in response to political pressure, [91] particularly from the United States after
proceedings had been initiated against Ariel Sharon and former president George HW
Bush. It now allows Belgian courts to hear complaints only when the suspect is Belgian
or lives in Belgium, the complainant is Belgian or a Belgian resident, or if a treaty
requires Belgium to exercise jurisdiction. [92] Other states—such as Spain—retain wide
universal jurisdiction statutes, but in practice judges tend to require some connection
with the forum state as a precondition for the exercise of jurisdiction. [93]
 Universal jurisdiction remains a controversial basis for the exercise of criminal
jurisdiction. The political will to exercise universal jurisdiction is seldom forthcoming
from states, [94] as illustrated by the manner in which the South African government
refused to arrest President al-Bashir, even when South Africa was requested by the
International Criminal Court to do so as a state party to the Court. [95] Although
universal jurisdiction has not often succeeded in judicial proceedings it has, however,
succeeded in deterring persons suspected of international crimes from foreign travel.
Ironically, the United States, which is vehemently opposed to the exercise of
universal criminal jurisdiction, practises broad civil universal jurisdiction over human
rights violations committed by non-nationals abroad under the Alien Torts Statute of
1789, which confers civil jurisdiction on a US federal court where an alien sues for a
tort committed in violation of the law of nations in any part of the world. [96]
Page 227

International crimes [97]

International crimes are crimes that threaten the good order not only of particular
states but of the international community as a whole. They are crimes in whose
suppression all states have an interest as they violate values that constitute the
foundation of the world public order. Some international crimes have their roots in
custom while others are the creations of conventions aimed principally at the
suppression of human rights violations and international terrorism. Historically,
international crimes were largely prosecuted at a domestic level. But since the 1990s,
various international legal tribunals have been established to try accused persons for
international crimes. [98] Below, various crimes (whether customary or conventional) are
explored, with particular emphasis on how (in recent years) states have expanded their
domestic prescriptive jurisdiction over these crimes.

1 Customary international law crimes

The earliest international crime was piracy, which is today codified in both the 1958
Geneva Convention on the High Seas [99] and the 1982 United Nations Convention on
the Law of the Sea. [100] According to these Conventions, piracy is defined as an illegal
act of violence committed for private ends by the crew or passengers of a private ship
or aircraft and directed against another ship or aircraft on the high seas. Another crime
under customary law is slave trading; indeed, it was at one time classified as a species
of piracy. Piracy is a crime under South African law. [101] Piracy is dealt with more fully
in Chapter 17 on the Law of the Sea.
War crimes and crimes against humanity are also crimes under customary
international law. [102] The London Charter of 1945, [103] which
Page 228
established the Nuremberg Tribunal, [104] defined these crimes for the purposes of the
trial held before it; but it is generally accepted that war crimes, and possibly crimes
against humanity, were already part of international customary law before 1945. Both
are recognised as crimes against the peace and security of mankind by the
International Law Commission’s 1996 Draft Code of Crimes against the Peace and
Security of Mankind. [105] In 1998 these concepts were refined, redefined and expanded
to take account of new developments in customary law in the Rome Statute of the
International Criminal Court. [106] These crimes are considered more fully in Chapter 10.
Genocide and torture are today also recognised as crimes under customary
international law, although both have a basis in a multilateral treaty.

2 International treaty crimes

In recent times, international crimes have been created by multilateral treaties,

principally to punish human rights violators or international terrorists. These treaties
generally oblige states to try or to extradite offenders. Few recognise universal
jurisdiction as a basis for prosecution, while many invoke a combination of traditional
bases for the exercise of multiple jurisdiction in order to ensure that all states with
some national interest in the prosecution of the offender will have jurisdiction.

(a) Genocide
The Genocide Convention of 1948 [107] created the crime of genocide—defined as any
act ‘committed with the intent to destroy in whole or in part a national, ethnical, racial
or religious group’. This crime is now part of customary international law, as evidenced
by its inclusion in the International Law Commission’s Draft Code of Crimes against the
Peace and Security of Mankind [108] and the Rome Statute of the International Criminal
Court. South Africa became a party to this Convention on 10 December 1998. Genocide
is considered more fully in Chapter 10.

(b) Apartheid
The international crime of apartheid is obviously of special interest to South Africa.
Page 229
In 1973 the General Assembly sponsored the International Convention on the
Suppression and Punishment of the Crime of Apartheid, [109] which declares that
‘apartheid is a crime against humanity’ [110] and criminalises the principal features of
apartheid, ranging from murder, torture and arbitrary arrests of members of a racial
group to legislative measures calculated to prevent a racial group from participation in
the political, social, economic and cultural life of the country, when committed for the
purpose of establishing and maintaining domination by one racial group over any other
racial group and systematically oppressing its members. [111]
Parties to the Convention undertake to enact legislation to prosecute persons
responsible for this international crime. Persons charged with this crime may be tried by
the courts of any signatory state ‘which may acquire jurisdiction over the person of the
accused’ or by an international penal tribunal. [112] As far as signatories are concerned,
a type of universal jurisdiction is therefore recognised. The offence is also made an
extraditable crime. This Convention was not implemented in practice during the
apartheid era. The symbolic impact of the Convention was, however, considerable as it
portrayed the policies and practices of apartheid as a crime against humanity—a
categorisation later confirmed by the Report of the Truth and Reconciliation
Commission. [113]
Although the Apartheid Convention is primarily aimed at apartheid as practised in
South Africa, it is not limited in time or space. Provisions of the Convention repeatedly
refer to the fact that it is to apply not only to apartheid as practised in South Africa but
to similar policies of racial discrimination wherever practised. Article I of the Convention
declares that ‘apartheid is a crime against humanity and that inhuman acts resulting
from the policies and practices of racial segregation and discrimination . . . are crimes
violating the principles of international law’, and in article IV parties agree to punish the
crime of apartheid and ‘similar segregationist policies or their manifestations’. That
apartheid knows no geographical limitation is
Page 230
confirmed by the 1977 Additional Protocol I to the Geneva Conventions of
1949, [114] which recognises as a grave breach of the Protocol and the Conventions
‘practices of apartheid and other inhuman and degrading practices involving outrages
upon personal dignity, based on racial discrimination’. [115]
While the Apartheid Convention remains binding as a convention on more than 100
states that are parties to the agreement, there has been an attempt in recent years to
subsume it in the crime against humanity. The 1996 Draft Code of Crimes against the
Peace and Security of Mankind recognises institutionalised racial discrimination as a
species of crime against humanity and explains in its commentary on this provision that
‘it is in fact the crime of apartheid under a more general denomination’. [116] The Rome
Statute of the International Criminal Court adopts a similar approach but retains the
term ‘crime of apartheid’ to describe inhumane acts ‘committed in the context of an
institutionalised regime of systematic oppression and domination by one racial group
over any other racial group or groups and committed with the intention of maintaining
that regime’. [117]
The precise status of the crime of apartheid is today uncertain. It is a war crime in
terms of Additional Protocol I. It is an international treaty crime for states parties to the
Apartheid Convention. It is a treaty crime, albeit a species of crimes against humanity,
for states that are parties to the Rome Statute of the International Criminal
Court. [118] It may be a crime against humanity under customary international
law, [119] although an American court has held that it is not a customary law
international wrong for the purposes of the Alien Tort Statute on the ground that the
convention has not been adopted ‘by most world powers’. [120] Although the crime has a
life of its own under contemporary international law, it serves as a constant reminder of
the racial policies once pursued by the South African government.

(c) Torture
The Torture Convention, to which South Africa became a party in 1998, is considered in
Chapter 15. It obliges a state party to make
Page 231
torture punishable under its domestic law and to take measures to establish jurisdiction
over an act of torture committed within its territory when either the offender or victim
is a national. Where the offender is ‘present’ in its territory, a state party is required to
either try or extradite him, which in effect establishes universal jurisdiction for parties
to the Convention. [121] Torture is a crime under customary international law [122] and is
designated as a form of crime against humanity under the Rome Statute of the
International Criminal Court. [123]

(d) Hijacking
Three conventions seek to outlaw hijacking and to facilitate the prosecution of
hijackers.
 The Tokyo Convention of 1963, [124] which applies to acts which jeopardise the safety
of an aircraft in flight outside the territory of any state, confers jurisdiction over
offences committed on board such an aircraft on the state of registration of the
aircraft. [125] A signatory state that is not the state of registration may interfere with an
aircraft in flight only in order to exercise its criminal jurisdiction over an offence
committed on board an aircraft where ‘the offence has effect on the territory of such
state’, the offence has been committed by or against a national, or the offence is
against the security of the state. [126] These clearly represent the jurisdictional
connections discussed earlier in this chapter.
The Hague Convention of 1970 [127] makes it an offence for any person on board an
aircraft unlawfully, by force or intimidation, to
Page 232
seize control of the aircraft while it is in flight. [128] Parties are required to try or to
extradite such an offender. [129] A state is permitted to exercise jurisdiction when the
offence is committed on board an aircraft registered in that state, when the aircraft on
board which the offence is committed lands in the territory of the state with the alleged
offender still on board, or when the offence is committed on board an aircraft leased
without crew to a lessee who has his principal place of business or residence in that
state. In addition, a party to the Convention shall take steps ‘to establish its jurisdiction
over the offence in the case where the alleged offender is present in its territory’ and it
does not extradite him. [130]
The Montreal Convention of 1971 [131] establishes similar jurisdictional rules for
persons who sabotage aircraft on the ground or who place devices on an aircraft in
order to endanger its safety in flight.
South Africa is a party to all three Conventions and has incorporated their main
provisions into municipal law in the Civil Aviation Act of 2009. [132] In terms of s 133 of
this statute, the following acts, inter alia, are criminal offences punishable by a fine or
imprisonment not exceeding 30 years:
(a)
the unlawful seizure of or exercise of control over an aircraft by force or
intimidation;
(b)
any act of violence, including an assault on any person on board the aircraft,
which is likely to endanger the safety of the aircraft;
(c)
the communication of false information that endangers the safety of the aircraft
in service;
(d)
any damage to an aircraft in service, or the placing of a device likely to cause
damage to an aircraft in service, which renders the aircraft incapable of flight or
endangers its safety. [133]
South African courts are given competence over offences under the Act committed
outside South African airspace when the crime takes place on board a South African
aircraft; the aircraft in which the offence is committed lands in the Republic with the
offender on
Page 233
board; the act takes place on board an aircraft leased without crew to a lessee who
resides permanently in the Republic; the offender is ‘present in the Republic’; or the
offender is apprehended in the Republic. [134] These last two grounds reflect universal
jurisdiction. If the South African government elects not to prosecute the offender, it
may extradite him to the state in which the aircraft is registered (or the lessee of the
aircraft is permanently resident) or in which the aircraft landed with the offender on
board, provided that such state is a party to one of the above-mentioned
Conventions. [135]

(e) Offences against the safety of maritime navigation

Piracy, as shown above, involves an act by the crew of one ship against another ship on
the high seas, committed for private profit. It does not cover the situation in which
passengers on board a ship seize control of the ship by unlawful means on the high
seas for a political purpose. This gap in the law was starkly illustrated in 1985 when a
group of Palestinian terrorists, who had boarded the Achille Lauro luxury liner as
passengers, seized control of the ship, held its crew and passengers hostage for the
release of prisoners in Israel, and murdered an American national. [136] The result was
the adoption of the Rome Convention for the Suppression of Unlawful Acts against the
Safety of Maritime Navigation of 1988, [137] which extends the crime of hijacking to
ships beyond the outer limits of the territorial sea and requires signatories to try or
extradite offenders. Jurisdiction to try offenders is conferred on the flag-state of the
ship seized, the state of which the offender is a national, the state whose nationals are
victims, and the state at which the seizure is directed ‘to compel that state to do or
abstain from doing any act’ (a combination of the ‘effects’ and state protection
principles). As in the case of aircraft hijacking, signatory states are required to establish
jurisdiction over offenders ‘present’ in their territory when they choose not to
extradite. [138]
Page 234

(f) Drug-trafficking
The 1988 Convention against Illicit Traffic in Narcotic Drugs and Psychotropic
Substances, [139] to which South Africa is a party, [140] is a treaty of co-operation which
requires states parties to criminalise the production and distribution of certain narcotic
drugs and to exercise jurisdiction over such crimes committed within their territory. It
also provides for the extradition of offenders, the confiscation of prohibited drugs and
the proceeds of crimes, and mutual legal assistance in the combating of drug-
trafficking.

(g) International terrorism

The adoption of a comprehensive agreement to outlaw international terrorism has been
on the agenda of the international community for over 70 years. [141] In 1937 a
Convention for the Prevention and Punishment of Terrorism [142] was drafted in response
to the assassination of King Alexander I of Yugoslavia, which defined acts of terrorism
as ‘criminal acts directed against a state and intended or calculated to create a state of
terror in the minds of particular persons, or a group of persons or the general public’.
The Convention, however, was ratified by one state only (India) and never came into
force. In more recent times, attempts to draft such a treaty have failed. During the
Cold War and, particularly, the period of decolonisation, one state’s ‘terrorist’ was often
another’s ‘freedom fighter’. [143] Wars of national liberation were supported by the
Soviet Bloc and developing states, which argued that all methods employed to
overthrow colonial, racist or alien regimes were permissible. In this climate, states
sought to reach consensus on narrowly defined species of terrorism rather than
terrorism itself. Thus, the hijacking of aircraft
Page 235
and ships was criminalised. The taking of hostages [144] and acts of terrorism aimed at
diplomats [145] were also prohibited by treaty.
The end of the Cold War and decolonisation resulted in a radical change of attitude
towards international terrorism. The international community no longer showed concern
for the causes of terrorism, and the motives of the terrorist, but with the most effective
ways of eliminating terrorism. Resolutions of the General Assembly of the United
Nations contrasted sharply with the deliberate inaction of this body in the early 1970s.
In 1994, in its Declaration on Measures to Eliminate International Terrorism, [146] the
General Assembly declared that member states of the United Nations condemn all acts
of terrorism wherever and by whomever committed, and that
criminal acts intended or calculated to provoke a state of terror in the general public, a group
of persons or particular persons for political purposes are in any circumstances unjustifiable,
whatever the considerations of a political, philosophical, ideological, racial, ethnic, religious or
any other nature that may be invoked to justify them.
This new mood resulted in the adoption, in 1997, of the International Convention for
the Suppression of Terrorist Bombings, [147] which comes close to a general anti-
terrorism convention. This Convention makes it an offence for any person to unlawfully
and intentionally place or detonate an explosive device in a place of public use, a state
or government facility, or a transportation system with intent to cause serious bodily
injury or extensive destruction. [148] A state party is required to exercise criminal
jurisdiction over such crimes under domestic law on grounds of territoriality and active
nationality and may do so on grounds of passive personality or where ‘the offence is
committed in an attempt to compel that state to do or abstain from doing any
act’. [149] As with similar conventions, a state party is obliged to exercise jurisdiction
over an offender ‘present’ in its territory when it fails to extradite such
person. [150] Unlike other conventions, this Convention expressly declares that acts
prohibited by this Convention cannot be justified on political, philosophical, ideological,
racial, ethnic or religious grounds, [151] and
Page 236
that the acts cannot be categorised as political crimes for the purpose of
extradition. [152]
This Convention was followed, in 2000, by the International Convention for the
Suppression of the Financing of Terrorism, [153] which makes it an offence to fund
terrorist offences as defined in the Conventions discussed above, and in 2005 by the
International Convention for the Suppression of Acts of Nuclear Terrorism, which makes
it an offence to possess nuclear material or make a nuclear device with intent to cause
death, bodily injury or damage to the environment. The year 1998 saw the adoption of
the Rome Statute of the International Criminal Court, but, despite proposals that this
Court be given jurisdiction over terrorist offences, the Court’s jurisdiction was confined
to aggression (defined only in 2010), genocide, crimes against humanity and war
crimes.
The Sixth (Legal) Committee of the General Assembly has, since 2000, been
engaged in the task of drafting a comprehensive convention on terrorism which will
embrace all forms of terrorism. The outbreak of violence in the Middle East in 2000 has,
however, resurrected the debate over the legitimacy of measures taken by those
engaged in resistance to military occupation and this has prevented the adoption of
such a treaty. This has not, however, deterred regional organisations from adopting
general conventions aimed at the suppression of terrorism. [154] In 1999 the
Organization of African Unity adopted a Convention of the Prevention and Combating of
Terrorism, which defines terrorism in broad terms but provides that
the struggle waged by peoples in accordance with the principles of international law for their
liberation or self-determination, including armed struggle against colonialism, occupation,
aggression and domination by foreign forces shall not be considered as terrorist acts. [155]
Since 11 September 2001, when massive acts of terrorism were perpetrated in New
York and Washington DC, the combating of international terrorism has become the
principal concern of the international community. The Security Council of the United
Nations has adopted a number of resolutions, notably resolutions 1373 (2001)
Page 237
and 1566 (2004), under Chapter VII of the Charter, calling upon states to prevent and
suppress the financing of terrorist acts, to refrain from providing support to groups
involved in terrorist acts, to prevent the commission of terrorist acts, to become parties
to international treaties on terrorism, and to enact and enforce domestic legislation to
suppress international terrorism. Resolution 1566, adopted on 8 October 2004, is
particularly important. It calls upon states to co-operate fully in the fight against
terrorism ‘on the basis of the principle of extradite or prosecute’ (aut dedere aut
judicare) and attempts a comprehensive definition of terrorism in a paragraph that:
Recalls that criminal acts, including against civilians, committed with the intent to cause death
or serious bodily injury, or taking of hostages, with the purpose to provoke a state of terror in
the general public or in a group of persons or particular persons, intimidate a population or
compel a government or an international organization to do or to abstain from doing any act,
which constitute offences within the scope of and as defined in the international conventions
and protocols relating to terrorism, are under no circumstances justifiable by considerations of
a political, philosophical, ideological, racial, ethnic, religious or other similar nature, and calls
upon all States to prevent such acts and, if not prevented, to ensure that such acts are
punished by penalties consistent with their grave nature.
The Security Council, in Resolution 2253 (2015), recently reaffirmed that member
states must hold accountable those responsible for committing, organising or
supporting terrorist acts, ‘in accordance with their obligations under international law,
in order to find and bring to justice, extradite, or prosecute any person who supports,
facilitates, participates or attempts to participate in the direct or indirect financing of
activities conducted by ISIL, Al-Qaeda and associated individuals, groups, undertakings
and entities’.
It is important to note that the main objective of the UN counter-terrorism
instruments (being to ensure the apprehension, prosecution, or extradition of persons
suspected of committing acts of terrorism) cannot be achieved without the assistance of
domestic courts. That is because there are no international tribunals with jurisdictional
competence to try acts of terrorism as defined in the counter-terrorism
instruments. [156] Accordingly, the duty to bring perpetrators of terrorist acts to justice
falls solely on domestic courts. [157]
Page 238
No doubt for that reason, in the Madan case, the Supreme Court of India was of the
view that terrorist violence ‘affects society as a whole by terrorising and disturbing the
harmony of society’. [158] Similarly, the Argentinian Supreme Court in the Clavel case
defined terrorism as a ‘crime jus gentium, the prosecution of which is not in the
exclusive interest of the state injured by it, but . . . benefits, ultimately, all civilised
nations who are therefore obligated to co-operate in the global fight against
terrorism’. [159]
In the light of the above treaties and state practice, it is difficult to resist the
conclusion that terrorism is today a crime under customary international law. This was
confirmed by an Interlocutory Decision of the Appeals Chamber of the Special Tribunal
for Lebanon on 16 February 2011, which ruled that treaties, UN resolutions, and the
legislative and judicial practice of states
evince the formation of a general opinio juris in the international community, accompanied by
practice consistent with such opinio, to the effect that a customary rule of international law
regarding the international crime of terrorism, at least in time of peace, has indeed emerged.
This customary rule requires the following three key elements: (i) the perpetration of a
criminal act (such as murder, kidnapping, hostage-taking, arson and so on), or threatening
such an act; (ii) the intent to spread fear among the population (which would generally entail
the creation of public danger) or directly or indirectly coerce a national or international
authority to take some action, or to refrain from taking it; (iii) when the act involves a
transnational element. [160]

3 The Protection of Constitutional Democracy against

Terrorist and Related Activities Act 33 of 2004

In 2004 the South African Parliament enacted the Protection of Constitutional

Democracy against Terrorist and Related Activities Act [161] (’the Act’) to give effect to
South Africa’s obligations in respect of the suppression of terrorism under United
Nations conventions, Security Council resolutions and the OAU Convention on the
Prevention and Combating of Terrorism of 1999. [162] The Constitutional Court has
explained that the purposes of the Act are to create an offence of terrorism that gives
effect to international instruments, to extend South African courts’ jurisdiction over
these offences
Page 239
wherever they are committed, to fulfil South Africa’s general duty to combat terrorism,
and to fulfil South Africa’s specific duty to bring to trial perpetrators of terrorism,
wherever perpetrated, whom it does not extradite. In S v Okah, the Constitutional
Court held that ‘[t]he general duty to combat terrorism is broad. It commands a
reading of the Act that enables South Africa to participate, as a member of the
international community, in the fight against an international and transnational
phenomenon’. [163]
The Act distinguishes between offences constituting ‘terrorist activity’ and
‘convention offences’, and confers wide jurisdictional powers on South African courts in
respect of both types of terrorism. Although the Act primarily seeks to give effect to
United Nations conventions and resolutions, its loyalty to the OAU (now the African
Union (AU)) Convention is apparent in its concern to exclude those engaged in wars of
national liberation from the ambit of terrorism.
Section 1(1) provides a broad definition of ‘terrorist activity’, divided into (a) the acts
constituting terrorism, (b) the intent required, and (c) the motivation of the
perpetrator. ‘Terrorist activity’ covers any act committed in or outside South Africa
which involves the ‘systematic repeated or arbitrary use of violence’ by any means;
‘systematic, repeated or arbitrary’ release into the environment of dangerous or
harmful substances; endangering the life of any person or persons; causing serious
bodily injury to or killing any person or persons; causing serious risk to the health or
safety of the public; causing the destruction of or substantial damage to any property,
natural resource or environmental or cultural heritage; causing any major economic
loss; acts designed to disrupt seriously any essential service; or acts that create a
‘serious public emergency situation or a general insurrection in the Republic’. The
activity must be intended, or by its nature and context reasonably be regarded as being
intended, to:
(i)
threaten the unity and territorial integrity of the Republic;
(ii)
intimidate, or to induce or cause feelings of insecurity within, the public, or a segment of the
public, with regard to its security, including its economic security, or to induce, cause or spread
feelings of terror, fear or panic in a civilian population; or
(iii)
unduly compel, intimidate, force, coerce, induce or cause a person, a government, the general
public or a segment of the public, or a domestic or an international organisation or body or
intergovernmental organisation or body, to do or to abstain or refrain from doing any act, or to
adopt or abandon a particular standpoint, or to act in accordance with certain principles,
Page 240
whether the public or the person, government, body, or organisation or institution
referred to in subparagraphs (ii) or (iii), as the case may be, is inside or outside the
Republic . . . .
Finally, the activity must be committed ‘for the purpose of the advancement of an
individual or collective political, religious, ideological or philosophical motive, objective,
cause or undertaking’.
Any person who engages in ‘terrorist activity’ is guilty of the offence of terrorism (s
2), and any person who commits any act that furthers terrorist activities is guilty of the
offence of association with or connection with terrorist activities (s 3).
The offence of terrorism is more widely defined than in any of the international
instruments that it seeks to incorporate. It could, for instance, include political protest
meetings or industrial action. Hence, s 1(3) excludes from the definition of ‘terrorist
activity’ acts that seriously disrupt essential services or cause major economic loss
which are ‘committed in pursuance of any advocacy, protest, dissent or industrial
action’, and which are not intended to cause harm of the kind resulting from any other
act constituting a ‘terrorist activity’.
Sections 4 to 10 incorporate anti-terrorism conventions to which South Africa is, or
intends to become, a party, into municipal law. The International Convention on the
Suppression of the Financing of Terrorism (s 4), the International Convention of the
Suppression of Terrorist Bombings (s 5), the Protocol for the Suppression of Unlawful
Acts against the Safety of Fixed Platforms on the Continental Shelf (s 6), the
International Convention against the Taking of Hostages (s 7), the Convention on the
Prevention and Punishment of Crimes against Internationally Protected Persons
including Diplomatic Agents (s 8), the conventions dealing with the hijacking and
sabotage of aircraft (s 9), and the Convention for the Suppression of Unlawful Acts
against the Safety of Maritime Navigation (s 10) are incorporated into South African
law. Section 9 fails fully to incorporate The Hague [164] and Montreal Conventions [165] on
aviation terrorism, and no attempt is made to incorporate the conventions on nuclear
terrorism. This is because these conventions are incorporated into South African law by
the Civil Aviation Act of 2009 [166] and the Nuclear Energy Act of 1999. [167] This is
confirmed by the definition of ‘convention offence’ in s 1 of the Act.
Page 241
 Wide jurisdictional powers are conferred on South African courts in respect of
offences under the Act. Jurisdiction may be exercised on grounds of territoriality, active
personality, passive personality, protection of the state, or ‘any other basis recognized
by law’. [168] A person who is ‘present’ in South Africa may be arrested and subjected to
criminal proceedings or extradition if a South African court has jurisdiction or ‘any court
in a foreign state may have jurisdiction’. [169]
In S v Okah the Constitutional Court was faced with the issue of whether the Act
confers extraterritorial jurisdiction on South African courts to try alleged offences
(beyond the financing of an offence) that occurred outside of South Africa. [170] Okah
had been convicted by a South African court of planning and financing various
bombings that occurred in Nigeria. In Okah’s appeal, the Supreme Court of Appeal held
that the Act gives South African courts extraterritorial jurisdiction over the financing of
terrorist activities only. Jurisdiction is not extended extraterritorially for the crime of
terrorism itself. [171] It therefore overruled the High Court’s finding that Okah was guilty
of terrorism—he was only found guilty of financing terrorism.
The Constitutional Court rejected the SCA’s finding. It held that the proper meaning
of ‘specified offence’, over which South African courts are given extraterritorial
jurisdiction, includes the terrorism activity itself and is not limited to the financing
thereof. This was because, among other reasons, limiting extraterritorial jurisdiction to
financing alone would create absurd instances in which it would be possible to
‘prosecute the banker, but not the bomber’. [172] The Constitutional Court held:
While it is true that territoriality has been the traditional basis on which courts establish
jurisdiction, international and South African jurisprudence recognise other methods of asserting
jurisdiction. Comity concerns fall away in cases where there is no infringement on the
sovereignty of another state. This is particularly true when the crimes over which a court
asserts jurisdiction have an international dimension. We should not, through a narrow
interpretation of section 15(1), mistakenly perpetuate an historical disinclination to extra-
territoriality. [173]
The President is required to give notice by proclamation in the Gazette of any Security
Council resolution adopted under Chapter VII
Page 242
of the Charter that identifies a terrorist ‘entity’ against which states are obliged to take
specified actions. Such a proclamation must be tabled in Parliament for consideration
and action. [174]
United Nations anti-terrorism conventions and Security Council resolutions make no
allowance by way of exception for the actions of national liberation movements. The
OAU (now AU) Convention on the Prevention and Combating of Terrorism of 1999, to
which South Africa is a party, does, however, exclude from the ambit of terrorism acts
committed in the course of ‘the struggle waged by peoples in accordance with the
principles of international law for their liberation or self-determination, including armed
struggle against colonialism, occupation, aggression and domination by foreign forces’.
This has clearly put the South African law-maker in a quandary as the Act seeks to
accommodate both United Nations conventions and the OAU (now AU) Convention.
First, the Act’s preamble asserts that the activities of national liberation movements
are exempt from the prohibition on terrorism. Then, s 1(4) provides:
Notwithstanding any provision of this Act or any other law, any act committed during a
struggle waged by peoples, including any action during an armed struggle, in the exercise or
furtherance of their legitimate right to national liberation, self-determination and independence
against colonialism, or occupation or aggression or domination by alien or foreign forces, in
 accordance with the principles of international law, especially international humanitarian law,
including the purposes and principles of the Charter of the United Nations and the Declaration
on Principles of International Law concerning Friendly Relations and Cooperation among States
in accordance with the said Charter, shall not, for any reason, including for purposes of
prosecution or extradition, be considered as a terrorist activity, as defined in subsection (1).
That the law-maker was aware that the above exemption is highly controversial is
demonstrated by s 1(5), which provides:
Notwithstanding any provision in any other law, and subject to subsection (4), a political
philosophical, ideological, racial, ethnic, religious or any similar motive, shall not be considered
for any reason, including for purposes of prosecution or extradition, to be a justifiable defence
in respect of an offence of which the definition of terrorist activity forms an integral part
[emphasis added].
The language of the latter provision is taken from provisions in the International
Convention for the Suppression of Terrorist Bombings [175] and the International
Convention for the Suppression of the
Page 243
Financing of Terrorism, [176] which are designed to ensure that a political or ideological
motive—including commitment to national liberation—shall not be a justifiable defence
to terrorism. In Okah, the Constitutional Court explained:
Section 1(4) of the Act exempts certain acts from the definition of terrorism, and, thus,
prosecution under the Act. To fall under section 1(4), the following criteria must all be met: (1)
the act must have taken place within the context of a ‘struggle waged by peoples’; (2) that
struggle must be ‘in the exercise or furtherance of their legitimate right to national liberation,
self-determination and independence against colonialism, or occupation or aggression or
domination by alien or foreign forces’; and (3) the act must be taken ‘in accordance with the
principles of international law, especially international humanitarian law’.
On the facts of Okah, the bombings orchestrated by Okah violated principles of
international humanitarian law. They did not distinguish between combatants and
civilians. So Okah’s conduct did not qualify for an exemption under s 1(4). That is so,
for at least two further reasons (which the Constitutional Court did not consider
expressly in its judgment). First, according to articles 25 and 103 of the UN Charter,
member states are obliged to carry out Security Council decisions even if they conflict
with any ‘other international agreements’. [177] Accordingly, even if s 1(4) of the Act is
to be understood as modelled on the AU Terrorism Convention, [178] the Security Council
resolutions must be understood as imposing overriding obligations on South Africa.
Secondly, the AU Terrorism Convention’s—and s 1(4)’s—internal limitation, which
requires that the struggle ‘must be waged in accordance with the principles of
international law’, means that any acts that are committed contrary to the Security
Council’s resolutions on terrorism are in any event not in accordance with those
principles.
Whether the subjection of sub-s (5) to sub-s (4) of s 1 is effective is placed in doubt
by the definition of ‘terrorist activity’, which in (c) requires that ‘terrorist activity’ be
committed for the advancement of a ‘political, religious, ideological or philosophical
motive, objective, cause or undertaking’. The amendment to the Extradition Act of
1962, [179] contained in the schedule to the Protection of Constitutional
Page 244
Democracy against Terrorist and Related Activities Act 33 of 2004, [180] adds to the
confusion on this subject as it provides that extradition may not be refused ‘on the sole
ground that it concerns a political offence’ in the case of offences under ss 4 and 5 of
Act 33 of 2004—that is, offences under the Conventions on Terrorist Bombings and the
Financing of Terrorism. At least in this case, there is no exemption for the political
offender engaged in a struggle for national liberation.
The exemption of those engaged in a struggle for national liberation from the
criminalisation of terrorism cannot be reconciled with the prevailing international
approach to terrorism reflected in international conventions and Security Council
resolutions. This is made clear, as shown above, in Security Council resolution 1566 of
8 October 2004, which stresses that terrorist acts ‘are under no circumstances
justifiable by considerations of a political, philosophical, ideological, racial, ethnic,
religious or other similar nature’—and which has been reiterated by the Security Council
in Resolution 2322 (2016). In its preamble, it reaffirms ‘that terrorism in all its forms
and manifestations constitutes one of the most serious threats to peace and security’
and that any acts of terrorism are criminal and unjustifiable regardless of the
motivations, whenever, whoever and by whomsoever committed. Section 233 of the
Constitution makes clear that any interpretation of the Act’s s 1(4) exemption must be
consistent with international law, which includes Security Council resolutions. These
resolutions make no allowance by way of exception for the actions of national liberation
movements.
The Protection of Constitutional Democracy against Terrorist and Related Activities
Act contains provisions dealing with the institution of proceedings, the penalties to be
imposed, evidence, investigating powers and preservation orders. These provisions fall
largely within the field of domestic law and are consequently not considered.

[1] C Ryngaert Jurisdiction in International Law 2 ed (2015); Abdi v Minister of Home Affairs 2011 (3)
SA 37 (SCA) 51, fn 8.
[2] Island of Palmas Case (United States v Netherlands) 2 RIAA 829 (1928) 838.
[3] In Kaunda v President of the Republic of South Africa 2005 (4) SA 235 (CC) the Constitutional Court
stated: ‘It is a general rule of international law that the laws of a State ordinarily apply only within its
own territory’ (para 38).
[4] Commissioner of Taxes, Federation of Rhodesia v McFarland 1965 (1) SA 470 (W) 474A–
B; Standard Bank of South Africa Ltd v Ocean Commodities Inc 1980 (2) SA 175 (T) 184G–185D; Abdi v
Minister of Home Affairs (n 1) para 29.
[5] This article provides that ‘[n]othing contained in the present Charter shall authorize the United
Nations to intervene in matters which are essentially within the domestic jurisdiction of any state’. This
provision, which during the apartheid era formed the cornerstone of South Africa’s foreign policy, is
considered in Chapter 15.
[6] CF Forsyth Private International Law 5 ed (2012).
[7] An exception to this rule is to be found in the United States Alien Tort Statute of 1789, which
confers civil jurisdiction on a US federal court where an alien sues for a tort committed in violation of the
law of nations. (See on this statute Sosa v Alvarez-Machain (US Supreme Court) (2004) 43 ILM 1390.) In
2004 a US federal district court considered (and dismissed) claims under this statute by victims of
apartheid against several multinational corporations that had engaged in business in apartheid South
Africa. In finding that doing business in apartheid South Africa had not been a tort committed in violation
of the law of nations, the Court heeded the South African government’s objection to the litigation on the
ground that it would discourage foreign investment in South Africa. In re South African Apartheid
Litigation: Ntsebeza et al v Citigroup et al 346 F Supp 2d 538. Cf Khulumani v Barclay National Bank 504
F 3rd ed 254. Further on this subject, see n 97 below.
[8] It is also known as legislative jurisdiction.
[9] R O’Keefe ‘Universal jurisdiction: Clarifying the basic concept’ (2004) 2 Journal of International
Criminal Justice 735 at 736.
[10] Prescriptive jurisdiction is thus important because conduct must be criminalised before it can be
prosecuted—otherwise the criminal law in question will be retroactive. Ibid 742. The only exception would
be if a state is purely monist and does not need to prohibit an international crime at the domestic level
before enforcing it. See C Kress ‘Universal jurisdiction over international crimes and the Institut de Droit
International’ (2006) 4 JICJ 561 at 564.
[11] This is also known as executive jurisdiction.
[12] O’Keefe (n 9) at 736.
[13] Some writers refer to a third aspect to jurisdiction: adjudicative jurisdiction. This refers to a
national court’s competence to adjudge certain matters. In a criminal law context, adjudicative
jurisdiction overlaps entirely with prescriptive and enforcement jurisdiction. When a court pronounces on
the applicability of the criminal law concerned, that is a manifestation of prescriptive jurisdiction. When a
court tries, convicts and/or sentences a person for committing certain conduct, that is enforcement
jurisdiction. See O’Keefe (n 9) at 737; Kress (n 10) at 564. See also the Canadian Supreme Court’s
decision in R v Hape [2007] 2 SCR 292, 2007 SCC 26.
[14] The converse is also true in principle, although exercising enforcement jurisdiction without first
exercising prescriptive jurisdiction could violate the law of non-retroactivity. O’Keefe (n 9) 741.
[15] 1927 PCIJ Reports Series A no 10.
[16] At 18–19. This principle was referred to with approval by Vieyra J in Commissioner of Taxes,
Federation of Rhodesia v McFarland (n 4) 473G–H. See, too, Kaunda v President of the Republic of South
Africa (n 3) para 38.
[17] At 19 (emphasis added).
[18] At 20.
[19] Lotus Case (n 15) 18.
[20] See H Lauterpacht International Law: Collected Papers (ed E Lauterpacht) (1970) vol 1, 488–9.
See, too, the tentative criticism of this dictum in Kaunda v President of the Republic of South Africa (n 3)
para 39. Also see O’Keefe (n 9) fn 12.
[21] See FA Mann ‘The doctrine of jurisdiction in international law’ (1964) 111 Hague Recueil 1 and
‘The doctrine of international jurisdiction revisited after twenty years’ (1984) 186 Hague Recueil 9.
[22] O’Keefe (n 9) 738; H Woolaver ‘Prosecuting international crimes in South Africa: Interpreting the
requirement of the accused’s presence in South African territory under the implementation of the Rome
Statute of the ICC Act’ (2014) 131 SALJ 253 at 256. The status of these principles in international law is
far from clear. See O’Keefe (n 9) fn 12.
[23] 2007 (3) SA 582 (CC), 2005 (12) BCLR 1192 (CC) paras 226–30. Here, the Court relied on the
decision of the Supreme Court of Canada in Libman v The Queen [1985] 2 SCR 178.
[24] Ibid paras 225–7.
[25] On the basis of passive personality, discussed below.
[26] See Chapter 11.
[27] (2002) 41 ILM 517, 123 ILR 94 para 59. In this case, a claim involving injuries resulting from the
NATO bombing of Belgrade was rejected as inadmissible because the bombing did not occur in the
territories of NATO states.
[28] R v Holm; R v Pienaar 1948 (1) SA 925 (A) 929–30; Commissioner of Taxes, Federation of
Rhodesia v McFarland (n 4) 473F. See in respect of civil jurisdiction Coin Security Group (Pty) Ltd v Smit
NO 1991 (2) SA 315 (T). In Stopforth & Veenendal v Minister of Justice 2000 (1) SA 113 (SCA) the
Supreme Court of Appeal held that the Amnesty Committee, established under the Promotion of National
Unity and Reconciliation Act 34 of 1995, had no power to grant amnesty ‘in respect of offences committed
outside South Africa which are not triable in this country but in another country in which any amnesty
purportedly conferred by the Amnesty Committee would not be recognised’.
[29] S v Makhutla 1968 (2) SA 768 (O); S v Maseki 1981 (4) SA 374 (T); GE Devenish Interpretation
of Statutes (1992) 215. Parliament may expressly provide that a statute is to operate extraterritorially.
For example, see s 2(1) of the Prevention of Organised Crime Act 121 of 1998; and s 4(3) of the
Implementation of the Rome Statute of the International Criminal Court Act 27 of 2002.
[30] South Africa’s territorial waters extend for 12 miles from the low-water line of the coastline. In
addition, South Africa has an exclusive economic zone of 200 miles from its coastline, in which it
exercises jurisdiction over natural resources: ss 4 and 7 of the Maritime Zones Act 15 of 1994. See, too, s
22(2) of the Defence Act 42 of 2002.
[31] Section 327 of the Merchant Shipping Act 57 of 1951.
[32] Section 150 of the Civil Aviation Act 13 of 2009.
[33] S v Kruger 1989 (1) SA 785 (A) 793C–E.
[34] S v Basson (n 23) paras 226–7.
[35] See Chapter 12.
[36] S v Ebrahim 1991 (2) SA 553 (A).
[37] Nkondo v Minister of Police 1980 (2) SA 894 (O) 898–900.
[38] See S v Dersley 1997 (2) SACR 253 (Ck) 255–60, particularly at 260C–E. It is unfortunate that
White J, in an otherwise thorough and long overdue examination of the bases for the exercise of criminal
jurisdiction in South Africa, saw fit to use the terms ‘citizen’ (257I) and ‘domicile’ (258J). The basis for
the exercise of criminal jurisdiction is physical presence.
[39] 1986 (1) SA 556 (ZS).
[40] At 563–4. This dictum was approved in S v Dersley (n 38) 259–60. Cf Martin v Republic of South
Africa 836 F 2d 91 (2d Cir 1987); M Leigh ‘Sovereign immunity’ (1988) 82 AJIL 583.
[41] B Cartoon ‘The Westinghouse Case: Collective response to the extraterritorial enforcement of
United States anti-trust laws’ (1983) 100 SALJ 731. See further DW Bowett ‘Jurisdiction: Changing
patterns of authority over activities and resources’ (1982) 53 BYIL 1; AV Lowe ‘The problems of
extraterritorial jurisdiction: Economic sovereignty and the search for a solution’ (1985) 34 ICLQ 724.
[42] In United States v Aluminum Company of America (Alcoa) 148 F 2d 416 (2d Cir 1945) the Court
declared that ‘any state may impose liabilities, even upon persons not within its allegiance, for conduct
outside its borders that has consequences within its borders which the state reprehends’ (443).
[43] R Jennings and A Watts (eds) Oppenheim’s International Law 9 ed (1992) vol 1 at 476.
[44] Timberlane Lumber Co v Bank of America 549 F 2d 597 (9th Cir 1976).
[45] See, for example, the manner in which the anti-trust laws were invoked to block an attempt by
Minorco SA to obtain control of Consolidated Goldfields: Consolidated Goldfields Plc v Minorco SA 871 F
2d 252 (1989); WH Callcott ‘Application of US law to foreign transactions’ (1989) 87 AJIL 923; FA Mann
‘The extremism of American extraterritorial jurisdiction’ (1990) 39 ICLQ 410.
[46] See AV Lowe ‘Blocking extraterritorial jurisdiction: The British Protection of Trading Interests Act,
1980’ (1981) 75 AJIL 257.
[47] 99 of 1978, as amended.
[48] 2005 (6) SA 158 (SCA).
[49] 89 of 1998.
[50] Paragraph 26.
[51] The Supreme Court of Appeal reached this conclusion without any discussion of international law.
The Competition Appeal Court, in reaching the same conclusion, made a thorough examination of
international law on the subject, citing both the Lotus Case (n 15) and the Barcelona Traction Case 1970
ICJ Reports 3 in the course of its reasoning. See American Soda Ash Corporation v Competition
Commission of South Africa 12/CAC/Dec 01; Oxford Reports on International Law ILDC 493.
[52] Attorney-General of the Government of Israel v Eichmann (1961) 36 ILR 5 para 30. This was
confirmed on appeal by the Supreme Court of Israel in Attorney-General of Israel v Eichmann (1962) 36
ILR 277 at 304 para 12. This is notwithstanding how, at the time of the Second World War and the
Holocaust, Israel did not exist. See D Lasok ‘The Eichmann trial’ (1962) 11 ICLQ 355 at 364.
[53] Joyce v DPP [1946] AC 347.
[54] 1949 (3) SA 1238 (Special Ct) 1250. This echoes the decision of the House of Lords in Joyce v
DPP [1946] AC 347. See, too, R v Holm; R v Pienaar (n 28) 930; Nduli v Minister of Justice 1978 (1) SA
893 (A) 912–13; S v Basson (n 23) para 225.
[55] See the Strafgesetzbuch [StGB] S 4 (FRG) and the Swedish Penal Code Ch 2, s 2. Commentators
have argued that this violates the rule of non-retroactivity. See C Bassiouni International Extradition:
United States Law and Practice 5 ed (2007) 401 and O’Keefe (n 9) 742–3. But see R Cryer An
Introduction to International Criminal Law 2 ed (2010) 48. Cryer explains that such states require the
conduct to be a crime at the locus delicti and then that jurisdiction is ‘borrowed’ by the state exercising
jurisdiction. This is also because such civil states refuse to extradite their own nationals and so they
extend jurisdiction to ensure prosecution.
[56] The United States rarely relies on it. See Bassiouni (n 55) 401.
[57] S v Mharapara (n 39) 47D–E.
[58] S v Mharapara (n 39) 559E–G. See, too, R v Holm; R v Pienaar (n 28) 930; S v Basson (n 23)
para 224 and para 172, fn 147.
[59] Supra (n 28).
[60] See R v Page [1953] 2 All ER 1335 at 1356. Note that s 9 of the British Offences against the
Persons Act 1861 asserts jurisdiction over murders committed by British nationals anywhere in the world.
[61] R v Holm; R v Pienaar (n 28) 931.
[62] S v Basson (n 23).
[63] Paragraphs 225–7.
[64] The Court cited with approval (para 226) La Forest J’s finding for the Supreme Court of Canada
in Libman v The Queen [1985] 2 SCR 178 at 212–3 that ‘all that is necessary to make an offence subject
to the jurisdiction of our courts is that a significant portion of the activities constituting that offence took
place in Canada. As it is put by modern academics, it is sufficient that there be a “real and substantial
link” between an offence and this country, a test well-known in public and private international law’.
[65] 27 of 2002, s 4(3)(a) and (b).
[66] S v Basson (n 23) para 223.
[67] See the separate opinion of Judge Moore of the United States in the Lotus Case (n 15) 89–93.
[68] Rome Convention for the Suppression of Unlawful Acts against the Safety of Maritime Navigation
(1988) 27 ILM 672; International Convention for the Suppression of Terrorist Bombings (1998)
37 ILM 249.
[69] Omnibus Diplomatic Security and Anti-Terrorism Act of 1986, Pub L No 99–399 1202(a), 100 Stat
853, 896 (codified at 18 USCA 2331 (Supp 1989)).
[70] 681 F Supp 896 (1988); 82 ILR 344.
[71] Section 4(3)(d) of the Implementation of the Rome Statute of the International Criminal Court Act
27 of 2002.
[72] L Reydams Universal Jurisdiction: International and Municipal Legal Perspectives (2003); S
Macedo (ed) National Courts and the Prosecution of Serious Crimes under International Law (2004);
‘Editorial comments on universality’ (2003) 1 Journal of International Criminal Justice 580; JD van der
Vyver ‘Universal jurisdiction in international criminal law’ (1999) 24 SAYIL 107.
[73] See Chapter 10.
[74] M Cherif Bassiouni and EM Wise Aut Dedere Aut Judicare: The Duty to Extradite or Prosecute in
International Law (1995). See Questions Relating to the Obligation to Prosecute or Extradite (Belgium v
Senegal) 2009 ICJ Reports at 156 and Questions Relating to the Obligation to Prosecute or Extradite
(Belgium v Senegal) 2012 ICJ Reports 422.
[75] Attorney-General of the Government of Israel v Eichmann (1962) 36 ILR 277 298–304.
[76] However, the authors are of the view that criminalising and prosecuting international crimes, given
their heinous nature and threat to international peace, will always be in the broader interests of a state.
[77] O’Keefe (n 9) 745.
[78] A Cassese ‘Is the bell tolling for universality? A plea for a sensible notion of universal jurisdiction’
(2003) 1 JICJ 589 at 594.
[79] Of course, the enforcement of the relevant criminal laws is contingent on the accused being
present in the state party’s territory.
[80] O’Keefe (n 9) 746–7.
[81] Though some commentators argue that there is a duty in customary international law to extradite
or prosecute those accused or convicted of such crimes. See Bassiouni (n 55) 7, 25.
[82] In S v Basson 2005 (1) SA 171 (CC) the Constitutional Court appeared to be prepared to permit
the extraterritorial prosecution of Dr Basson for crimes under the Riotous Assemblies Act 17 of 1956 on
the ground that the conduct in question constituted crimes against humanity and war crimes (paras 34,
37, 119–26). In a later decision in this case, however, the Constitutional Court found it unnecessary ‘to
consider whether customary international law could be used either as the basis in itself for a prosecution
under common law, or, alternatively, as an aid to the interpretation of s 18(2)(a) of the Riotous
Assemblies Act’ (n 23) para 172, fn 147. See further W Ferdinandusse Direct Application of International
Criminal Law in National Courts (2005). On the other hand, s 232 of the Constitution says that customary
international law is law in the Republic unless contrary to the Constitution or legislation. So customary
law crimes are domestic crimes in South Africa, so long as they are consistent with the Constitution and
legislation.
[83] See H Kissinger ‘The pitfalls of universal jurisdiction: Risking judicial tyranny’ (2001) 80 Foreign
Affairs 86. In S v Basson (n 23) the Constitutional Court stated that for the purposes of this case ‘it is not
necessary to enter into controversies surrounding the existence of universal jurisdiction for crimes against
humanity and war crimes and a concomitant duty to prosecute’ (para 172, fn 147). Most recently, see the
African Union’s criticism of universal jurisdiction in H van der Wilt ‘Universal jurisdiction under attack: An
assessment of African misgivings towards international criminal justice as administered by Western
states’ (2011) 9 JICJ 1043.
[84] Section 4(3)(d) of Act 27 of 2002.
[85] 2002 ICJ Reports 3. See further on this case G Erasmus and G Kemp ‘The application of
international criminal law before domestic courts in the light of recent developments in international and
constitutional law’ (2002) 27 SAYIL 64; M du Plessis and S Bosch ‘Immunities and universal jurisdiction—
The World Court steps in (or on?)’ (2003) 28 SAYIL 346.
[86] See Chapter 12.
[87] Arrest Warrant case (n 85) 42, para 12.
[88] Ibid 81, para 60.
[89] Ibid 80, para 59.
[90] Arrest Warrant case (n 85) 173, para 58.
[91] See S Ratner ‘Belgium’s war crimes statute: A postmortem’ (2003) 97 AJIL 888; N Roht-Arriaza
‘Universal jurisdiction: Steps forward, steps back’ (2004) 17 Leiden Journal of International Law 375.
[92] S Smis and K Van Der Borght ‘Introductory note to Belgium’s amendment to the law of June 16,
1993 (as amended by the law of February 10, 1999) concerning the punishment of grave breaches of
humanitarian law’ (2003) 42 ILM 740.
[93] Roht-Arriaza (n 91); Guatemala Genocide Case (2003) 42 ILM 683; Peruvian Genocide
Case (2003) 42 ILM 1200.
[94] See M Langer ‘The diplomacy of universal jurisdiction: The political branches and the transnational
prosecution of international crimes’ (2011) 105 AJIL 1; J Dugard ‘International criminal law, the
International Criminal Court and civil society’ 2016 Acta Juridica 3 at 7–8.
[95] See Chapter 10.
[96] See Sosa v Alvarez-Machain (n 7). See, too, the joint opinion of Judges Higgins, Kooijmans and
Buergenthal in the Arrest Warrant Case (n 85) 77, para 48. Recently, however, the US Supreme Court
has limited the scope of the Alien Torts Statute. The activities complained of must ‘touch and concern’ the
territory of the United States. See Kiobel v Royal Dutch Petroleum 133 S Ct 1659 (2013) at 1669.
The Kiobel limitation was used to reject claims by South African apartheid victims for damages from the
United States. See Balintulo v Ford Motor Co 796 F 3d 160, 167 (CA2 (NY) 2015). See M Swart ‘Requiem
for a dream? The impact of Kiobel on apartheid reparations in South Africa’ (2015) 13 JICJ 353.
[97] A Cassese and P Gaeta Cassese’s International Criminal Law 3 ed (2013); R Cryer, H Friman, D
Robertson and E Wilmshurst International Criminal Law and Procedure 3 ed (2014); C Stahn and L van
den Herik (eds) Future Perspectives on International Criminal Justice (2010); A Cassese (ed) The Oxford
Companion to International Criminal Justice (2009).
[98] See Chapter 10.
[99] Article 15.
[100] Article 101. See, generally, AP Rubin The Law of Piracy 2 ed (1998).
[101] Section 24 of the Defence Act 42 of 2002.
[102] M Cherif Bassiouni Crimes against Humanity in International Criminal Law (1999); TLH
McCormack and GJ Simpson (eds) The Law of War Crimes (1997).
[103] 5 UNTS 251, (1945) 39 AJIL Suppl 257.
[104] The judgment of this tribunal appears in (1947) 41 AJIL 172.
[105] Report of the International Law Commission on the work of its 48th Session, 1996, GAOR, 51st
Session Suppl No 10 (A/51/10).
[106] (1998) 37 ILM 999, articles 7 and 8. See further Chapter 10.
[107] See WA Schabas Genocide in International Law: The Crime of Crimes 2 ed (2009).
[108] See n 105.
[109] (1974) 13 ILM 50; RS Clark ‘The crime of apartheid’ in International Criminal Law (ed M
Bassiouni) vol I (Crimes) (1987); J Dugard ‘L’apartheid’ in H Ascensio, E Decaux and A Pellet Droit
International Pénal (2000) 349. For critical commentaries on this Convention, see H Booysen ‘Convention
on the crime of apartheid’ (1976) 2 SAYIL 56; GN Barrie ‘The Apartheid Convention after five years’
1981 TSAR 280; JC Heunis United Nations Versus South Africa (1986) 281; P Eden ‘The role of the Rome
Statute in the criminalization of apartheid’ (2014) 12 JICJ 171.
[110] Article I.
[111] Article II.
[112] Article V.
[113] Truth and Reconciliation Commission of South Africa Report (1998) vol 1 at 94, vol 5 at 222.
[114] (1977) 16 ILM 1391.
[115] Articles 85(4)(c) and 85(5).
[116] Supra (n 105) 99.
[117] Articles 7(1)(j) and 7(2)(h).
[118] Eden (n 109).
[119] In S v Basson (n 82) the Constitutional Court stated that it is ‘clear that the practice of apartheid
constituted crimes against humanity’ (para 37).
[120] In re South African Apartheid Litigation: Ntsebeza v Citigroup Inc (n 7).
[121] Convention against Torture and Other Cruel, Inhuman or Degrading Treatment (1984)
23 ILM 1027, article 5. For more on this subject, see R v Bow Street Stipendiary Magistrate, Ex p
Pinochet Ugarte (No 3) [1999] 2 All ER 97 (HL). In 1999, South Africa failed to try or extradite to Ethiopia
Mengistu Haile Mariam, the former dictator of Ethiopia, who was wanted by that country for torture and
other international crimes, despite its obligations under the Torture Convention to which it became a
party in 1998. See further J Dugard and CJ Roederer ‘Public international law’ 1999 Annual Survey 99 at
100.
[122] Furundzija Case IT-95-17/1-T10 (Trial Chamber of ICTY, Judgment, 10 December 1998) 121 ILR
213 at 260; E de Wet ‘The prohibition of torture as an international norm of jus cogens and its implication
for national and customary law’ (2004) 15 European Journal of International Law 97.
[123] Article 7(1)(f).
[124] Convention on Offences and Certain Other Acts Committed on Board Aircraft 704 UNTS 219.
[125] Article 3.
[126] Article 4.
[127] Convention for the Suppression of Unlawful Seizure of Aircraft 860 UNTS 105, (1971)
10 ILM 133.
[128] ‘Flight’ occurs at any time from the moment when all the external doors of an aircraft are closed
following embarkation until the moment when the doors are opened for disembarkation: article 3.
[129] Articles 7 and 8.
[130] Article 4.
[131] Convention for the Suppression of Unlawful Acts against the Safety of Civil Aviation (1971)
10 ILM 1151. In 1988, a Protocol to the Convention was adopted to bring attacks against people in
airports within the scope of the Convention: (1988) 27 ILM 627.
[132] Act 13 of 2009.
[133] For judicial interpretations of the predecessor of this provision in the Civil Aviation Offences Act
10 of 1972, see S v Jeffers 1975 (4) SA 657 (W); S v Jeffers 1976 (2) SA 636 (A); S v Bergman 1984 (1)
SA 182 (C); S v Hoare 1982 (4) SA 865 (N).
[134] Sections 150 and 151.
[135] Section 152.
[136] A Cassese Terrorism, Politics and Law: The Achille Lauro Affair (1989).
[137] (1988) 27 ILM 672. See further M Halberstam ‘Terrorism on the high seas: The Achille
Lauro piracy and the IMO Convention on maritime safety’ (1988) 82 AJIL 269; G Plant ‘The Convention
for the Suppression of Unlawful Acts against the Safety of Maritime Navigation’ (1990) 39 ICLQ 27.
[138] Article 6.
[139] (1989) 28 ILM 493. See further N Boister ‘The historical development of international legal
measures to suppress illicit drug trafficking’ (1997) 30 CILSA 1.
[140] Prince v President, Cape Law Society 2002 (2) SA 794 (CC) 824, 837, 851, 858–9.
[141] J Dugard ‘Terrorism and international law: Consensus at last?’ in E Yakpo and T Boumedra Liber
Amicorum Mohammed Bedjaoui (1999) 159; Symposium ‘A war against terrorism: What role for
international law? US and European perspectives’ (2003) 14 European Journal of International Law 209; J
Dugard ‘The problem of the definition of terrorism in international law’ in P Eden and T O’Donnell
(eds) September 11, 2001: A Turning Point in International and Domestic Law? (2005) 187; H Duffy The
‘War on Terror’ and the Framework of International Law 2 ed (2015); T Becker Terrorism and the
State (2006); B Saul Defining Terrorism in International Law (2008); B Saul Terrorism (2010); G Nesi
(ed) International Co-operation in Counter Terrorism (2005).
[142] The text of the Convention appears in MO Hudson International Legislation vol 7, no 499 (1941).
[143] J Dugard ‘International terrorism and the just war’ (1977) 12 Stanford Journal of International
Studies 21.
[144] International Convention against the Taking of Hostages (1979) 18 ILM 1456.
[145] Convention on the Prevention and Punishment of Crimes against Internationally Protected
Persons, including Diplomatic Agents (1974) 13 ILM 43.
[146] Annex to Resolution 49/60 of 9 December 1994.
[147] (1998) 37 ILM 249.
[148] Article 2.
[149] Article 6.
[150] Article 6(4).
[151] Article 5.
[152] Article 11.
[153] (2000) 39 ILM 268. Over 170 states are party to this Convention.
[154] For the texts of such instruments, see United Nations International Instruments related to the
Prevention and Suppression of International Terrorism (2008); Inter-American Convention against
Terrorism (2003) 42 ILM 19; Council framework decision of the European Union, 13 June 2002, Official
Journal L 164 vol 45 (22 June 2002).
[155] International Instruments (n 154) 222 article 3. See on the adoption and implementation of
terrorism conventions in Africa, A Thomashausen ‘The “war on terror” in Africa in international law and
state practice’ (2007) 32 SAYIL 85.
[156] The Rome Statute of the International Criminal Court, 1998, which created the ICC, would only
be applicable to acts of terrorism when committed as a crime against humanity (that is on a widespread
or systematic basis).
[157] A Sambei, A du Plessis and M Polaine Counter-terrorism Law and Practice: An International
Handbook (2009) 2.44 at 35–6. See also the discussion of the Constitutional Court’s decision in S v
Okah below.
[158] Madan Singh v State of Bihar (2004) INSC 225 (2 April 2004).
[159] Enrique Lautaro Arancibia Clavel 259 (Arg Supreme Court) (2004) 51–2.
[160] Case No STL-11-01/1 para 85. See, too, para 111. For criticisms of this decision, see the
Symposium in (2011) 24 Leiden Journal of International Law 651–700.
[161] Act 33 of 2004. See A Cachalia’s critical comments on this Act in ‘Counter-terrorism and
international cooperation against terrorism—An elusive goal: A South African perspective’ (2010)
26 SAJHR 510.
[162] The preamble makes this clear in setting out all the treaties which South Africa is obliged to
enforce.
[163] S v Okah 2018 (1) SACR 492 (CC) paras 35–7.
[164] Convention for the Suppression of Unlawful Seizure of Aircraft (n 127).
[165] Convention for the Suppression of Unlawful Acts against the Safety of Civil Aviation (n 131).
[166] Act 13 of 2009.
[167] Section 56(1)(h) of Act 46 of 1999, read with ss 34A and 56A inserted by the Schedule to the
Protection of Constitutional Democracy against Terrorist and Related Activities Act 33 of 2004 (s 27).
[168] Section 15(1)(b) and (c) of Act 33 of 2004.
[169] Section 15(5) to (8) of Act 33 of 2004.
[170] S v Okah (n 163).
[171] The Supreme Court of Appeal’s judgment is reported as S v Okah 2017 (1) SACR 1 (SCA).
[172] S v Okah (n 163) para 26.
[173] Ibid para 43.
[174] Sections 25 and 26.
 [175] Article 5.
[176] Article 6.
[177] Questions of Interpretation and Application of the 1971 Montreal Convention arising from the
Aerial Incident at Lockerbie (Libyan Arab Jamahiriya v United States of America), Provisional
Measures (order of 14 April 1992) 1992 ICJ Reports 114 para 42.
[178] Article 3(1) of which provides that ‘the struggle waged by peoples in accordance with the
principles of international law for their liberation or self-determination, including armed struggle against
colonialism, occupation, aggression and domination by foreign forces shall not be considered as terrorist
acts’.
[179] Act 67 of 1962.
[180] Section 27.

Page 245

Chapter 10
International Criminal Courts, the
International Criminal Court, and South
Africa’s Implementation of the Rome Statute

Max du Plessis and Eshed Cohen

International Criminal Courts

The idea of a permanent international criminal court was on the international agenda
for much of the last century. [1] After World War I, unsuccessful attempts were made to
bring the German Emperor to trial before an international tribunal [2] and, later, to try
Turks responsible for the genocide of Armenians before a tribunal designated by the
Allied powers. [3] In 1937, following the assassination in 1934 of King Alexander of
Yugoslavia by Croatian nationalists in Marseilles, treaties were drafted to outlaw
international terrorism [4] and to provide for the trial of terrorists before an international
tribunal, [5] but states lost interest in this venture as war approached: no state ratified
the treaty for an international criminal court and only one ratified the treaty outlawing
international terrorism. The aggressive war conducted by Germany, and the atrocities
committed by its officials
Page 246
and soldiers during World War II, provided the requisite impetus for the creation by
Allied powers of an ad hoc international military tribunal at Nuremberg in 1945. [6] (A
similar tribunal was constituted in Tokyo [7] in respect of crimes committed by Japan’s
leaders.) The establishment of the Nuremberg and Tokyo international military
tribunals, which tried the principal leaders of the Nazi and Japanese regimes after World
War II for crimes against the peace, war crimes and crimes against humanity, was a
natural culmination of the pre-war debate over an international criminal court.
Inevitably, however, there was criticism of the fact that these tribunals were
established by the victors to try the vanquished. [8] The United Nations was nonetheless
energised by the work of these tribunals to adopt, on 9 December 1948, a resolution
mandating the International Law Commission to begin work on the draft statute of an
international criminal court. [9] The enthusiasm generated by Nuremberg and Tokyo for
a permanent court in the immediate post-war period was, however, abandoned during
the Cold War. Even the consensus between East and West over apartheid failed to
produce the court that had been proposed to try apartheid’s criminals in the late
1970s. [10]
By the 1980s, a wide range of factors combined to strengthen the case for the
establishment of an international criminal court. These included: the increase in the
number of international crimes provided by treaties outlawing hijacking, hostage-
taking, torture, seizure of ships on the high seas and attacks on diplomats; the
emergence of powerful drug cartels capable of subverting the judicial systems of weak
states; and above all, the conviction that international law had progressed sufficiently
to enable it to condemn individuals before an international criminal court for violating
international norms. The final contributing factor was the end of the Cold War—it was
thereafter possible for a more unified United Nations to renew its interest in a
permanent international criminal court.
Page 247
The idea of a permanent criminal court for the world was placed back on the
international agenda through a proposal by Latin American states who envisaged such a
court as their last resort to prosecute international drug-traffickers. [11]
Thereafter, the International Law Commission was directed by the UN General
Assembly to consider the drafting of a statute for an international criminal court. The
Commission prepared a draft statute for such a court in the early 1990s, and by 1994,
a formal Draft Statute for an International Criminal Tribunal had been adopted by the
ILC and forwarded to the General Assembly for consideration. [12] During the time that
the Commission was preparing the Draft Statute, events compelled the creation of a
court on an ad hoc basis to respond to the atrocities that were being committed in the
former Yugoslavia. That tribunal, the International Criminal Tribunal for the Former
Yugoslavia (ICTY), was established by the Security Council in 1993 and mandated to
prosecute persons responsible for serious violations of international humanitarian law
committed in the territory of the former Yugoslavia since 1991. [13] Then, in November
1994, and acting on a request from Rwanda, the Security Council voted to create a
second ad hoc tribunal (ICTR), charged with the prosecution of genocide and other
serious violations of international humanitarian law committed in Rwanda and in
neighbouring countries during 1994. [14] Since they share statutes that are virtually
identical, the two tribunals are close relatives. [15]
The Rwanda and Yugoslav tribunals furthered the widespread belief that a
permanent international criminal court was desirable and practical. When delegates
convened in Rome in 1998 to draft a
Page 248
statute for a permanent international criminal court, the tribunals could provide a
reassuring model of how such a court might function. In addition to the example which
the tribunals provided of a working criminal justice system, the innovative international
criminal law jurisprudence that they had produced—such as the progressive view that
crimes against humanity could be committed in peacetime, [16] and the finding that war
crimes could be committed during an internal armed conflict [17]—fed into the debates
at Rome, and eventually came to be reflected in the Rome Statute. [18]
 The Statute of the International Criminal Court was adopted on 17 July 1998 by an
overwhelming majority of the states attending the Rome Conference. The conference
was specifically organised to secure agreement on a treaty for the establishment of a
permanent international criminal tribunal. After five weeks of intense negotiations, 120
countries voted to adopt the treaty. Only seven countries voted against it (including
China, Israel, Iraq and the United States) and 21 abstained. By the 31 December 2000
deadline, 139 states had signed the treaty. The treaty came into force upon 60
ratifications. Sixty-six countries—six more than the threshold needed to establish the
court—had ratified the treaty by 11 April 2002. This was much sooner than was
generally expected. To date, the Rome Statute has been ratified by 124 states [19] and a
significant proportion—33—are African. [20] South Africa remains a party to the Statute,
although of late its continued membership of the ICC is in doubt (see discussion
Page 249
further below). One significant absentee amongst the ratifications is that of the United
States. [21]
Along with the ad hoc international criminal tribunals for Rwanda and Yugoslavia,
together with, inter alia, the Special Court for Sierra Leone, [22] the International
Criminal Court stands as a working model of international criminal justice—one in which
an international criminal forum applies rules of international law, is staffed by
independent prosecutors and judges, and holds persons individually responsible for
crimes against humanity and war crimes, after allowing them a fair trial.

The International Criminal Court

The International Criminal Court (ICC) is situated in The Hague, in the Netherlands. The
ICC is an independent international organisation, and is not part of the United Nations
system. [23] The judges for
Page 250
the Court were sworn in on 11 March 2003 at the Court’s inaugural session. [24] In
2018, of the 20 judges, four are from Africa, including the judge president. The current
prosecutor is Fatou Bensouda from The Gambia.
The ICC is divided into an Appeals Division, a Trial Division and a Pre-Trial Chamber
Division. [25] The Office of the Prosecutor is responsible for receiving and examining
referrals and substantiated information on alleged crimes, and conducting
investigations and prosecutions before the Court. [26] The Office of the Prosecutor is
headed by the prosecutor, who has full authority over the management and
administration of the office. [27] In the interests of efficiency and consistency, the
prosecutor relies extensively on the registry for administrative services. The registry is
responsible for the non-judicial aspects of the administration and servicing of the Court,
without prejudice to the functions and powers of the prosecutor. The registry is headed
by the registrar, who is elected by the judges and who exercises her functions under
the authority of the president of the Court. [28] The work of the Court is overseen by an
Assembly of States Parties, which provides management oversight, considers and
decides the budget for the Court, conducts elections, and performs other functions. The
Assembly meets at least once a year. [29]

ICC Crimes
The Court can take up only the most serious crimes of concern to the international
community as a whole—genocide, crimes against humanity and war crimes—all of
which are defined in the Statute. [30] Aggression also falls within the competence of the
ICC but an accepted definition of this crime has only recently been added to the
Page 251
Statute. [31] Treaty crimes (such as terrorism or drug trafficking) do not fall within the
ICC’s jurisdiction but, like the crime of aggression, may be added later after
consideration by a review conference. [32] For the purposes of interpreting and applying
the definitions of crimes found in the Rome Statute, reference must also be made to the
Elements of Crimes, a 50-page document adopted in June 2000 by the Preparatory
Commission for the International Criminal Court. [33]

1 Genocide

Genocide is understood as involving the intentional mass destruction of entire groups,

or members of a group as members of that group. [34] The first criminal prosecution of
‘genocide’ took place at the International Military Tribunal at Nuremberg in 1947, when
the Nazi leaders were tried for ‘crimes against humanity’ under the Nuremberg
Charter. [35] It took almost half a century, with the establishment of the ICTY and ICTR
in 1993/4, before ‘genocide’ came before an international tribunal once again to be
prosecuted at an international level. [36]
Page 252
Article 6 of the Rome Statute, following article IV of the Genocide Convention,
defines the underlying conduct that, along with genocidal intent, constitutes the crime
of genocide:
(a)
killing members of a national or ethnic, racial, or religious group (meaning their
‘murder’, ie intentional, voluntary killing); [37]
(b)
causing serious bodily or mental harm to members of the group (these terms ‘do
not necessarily mean that the harm is permanent or irremediable’); [38]
(c)
deliberately inflicting on the group conditions of life calculated to bring about its
physical destruction (including, inter alia, ‘subjecting a group of people to a
subsistence diet, systematic expulsion from homes and the reduction of essential
medical services below minimum requirement[s]’); [39]
(d)
imposing measures intended to prevent birth within the group (such measures
consist of ‘sexual mutilation, the practice of sterilization, forced birth control [and
the] separation of the sexes and prohibition of marriage’); [40] or
(e)
forcibly transferring children of the group to another group.
The victim of the crime of genocide is the group itself and not the individual. [41] The
definition of genocide in the Rome Statute provides that genocide is any one of the
enumerated acts committed with intent to destroy, in whole or in part, a national,
ethnical, racial or religious group, as such. This chapeau is known as genocidal intent.
The common criterion in the four types of groups protected under the Genocide
Convention (national, ethnic, racial or religious) is that ‘membership in such groups
would seem to be normally not challengeable by its members, who belong to it
automatically, by birth, in a continuous and often irremediable manner’. [42] The ‘victim’
group therefore does not extend to what might be called ‘political’ and ‘social’ groups.
In respect of the Rome Statute, the drafters have evinced a clear intention to limit the
groups to the four identified by the Genocide Convention. The idea of including a
‘cultural group’ in the ICC Statute was rejected at the Rome Conference. The drafters
Page 253
were quick to point out that the Genocide Convention was aimed at preventing physical
destruction of a group, not cultural destruction. [43]
Genocide is known as the crime of all crimes, [44] as evidenced in the high threshold
set for the mental element required for proof of genocide. In the Jelisic Case, the ICTY
explained that ‘it is in fact the mens rea which gives genocide its speciality and
distinguishes it from an ordinary crime and other crimes against international
humanitarian law’. [45] Both customary and conventional definitions of genocide require
a prosecutor to establish a form of aggravated criminal intention, or specific intent
(dolus specialis) in addition to the criminal intent accompanying the underlying offence.
The accused must commit the underlying offence with the intent to produce the result
charged; that is, the intent to destroy, in whole or in part, a national, ethnic, racial or
religious group, as such. Genocide is therefore a crime perpetrated against a
‘depersonalised’ victim, and carried out for no other reason than that he or she is a
member of a specific national, ethnic, racial or religious group. [46]
The intention must be to destroy a group ‘in whole or in part’. Genocide can thus be
committed through the destruction of a large number of the group (a quantitative
attempt at destruction) or the destruction of a limited number of the group who are
targeted because of the potential impact of their destruction on the survival of
Page 254
the group as a whole (a qualitative attempt at destruction). An example of the latter
would be the act of destroying young, fertile women in a group who are of childbearing
age. The element of specific intent in the context of genocide ‘may, in the absence of
direct explicit evidence, be inferred from a number of facts and circumstances, such as
the general context, the perpetration of other culpable acts systematically directed
against the same group, the scale of the atrocities committed, the systematic targeting
of victims on account of their membership of a particular group, or the repetition of
destructive and discriminatory acts’. [47] In Akayesu, for example, the ICTR Trial
Chamber found that the accused had the requisite mens rea to commit genocide, and
had exhibited that aggravated criminal intention through, inter alia, the systematic rape
of Tutsi women. According to the ICTR, the systematic rape of Tutsi women was part of
the campaign to mobilise the Hutus against the Tutsi, and the sexual violence was
aimed at destroying the spirit, will to live, or will to procreate, of the Tutsi group. [48]

2 Crimes against humanity

The term ‘crime against humanity’ was first used in its contemporary sense to condemn
the atrocities committed by the Turkish forces against their own Greek and Armenian
subjects during World War I in 1915. Although no prosecutions ultimately took place,
the immediate response of the Allied powers to the massacres was for France, the UK
and Russia to proclaim enthusiastically that all members of the Turkish government
would be held responsible together with its agents for the ‘crimes against humanity and
civilization’. [49] At Nuremberg, the idea of a crime against humanity arose again. The
crime was the brainchild of Hersch Lauterpacht, a professor of international law in
England. [50] Through his influence, the Nuremberg and Tokyo tribunals utilised the
technical term ‘crime against humanity’ to secure, for the first time, the prosecution of
individuals for crimes that, by their nature, offended ‘humaneness’ and thereby became
the concern of the international community. [51]
Page 255
At Nuremberg, the notion of crimes against humanity was limited to those acts that
occurred only during an international armed conflict. [52] Today, in international criminal
law, the nexus between crimes against humanity and war has disappeared, and
customary international law prohibits crimes against humanity whether they are
committed in times of war or peace. [53]
Crimes against humanity are prohibited under article 7 of the Rome Statute. The
term ‘crimes against humanity’, under the Statute, covers actions that have a common
set of features: [54]
(a)
The offences are particularly egregious in that they constitute a serious attack on
human dignity or a grave degradation or humiliation of one or more human
beings.
(b)
They are not isolated or sporadic events but are acts that form part of
governmental policy, or of a widespread or systematic practice of atrocities
tolerated, condoned or acquiesced in by a government or de facto authority.
(c)
Their prohibition extends regardless of whether they are perpetrated in times of
war or peace.
(d)
Under the Rome Statute (and the Statutes of the ICTY and the ICTR) the victims
of the crimes are civilians or, in the case of crimes committed during armed
conflict, persons who do not take part (or no longer take part) in armed
hostilities.
Page 256
The specific acts or classes of offences that make up crimes against humanity under the
Rome Statute are those commonly associated with egregious abuses of human rights
and include: murder; [55] extermination [56] (involving mass or large-scale killing, [57] or
intentional infliction of conditions of life; inter alia, the deprivation of food and medicine
calculated to bring about the destruction of part of a
population); [58] enslavement; [59] deportation or forcible transfer of
population [60] (being the ‘forced displacement of the persons concerned by expulsion or
other coercive acts from the area in which they are lawfully present, without grounds
permitted under international law’); [61] imprisonment or other severe deprivation of
physical liberty in violation of fundamental rules of international
law; [62] torture [63] (being ‘the intentional infliction of severe pain or suffering, whether
physical or mental, upon a person in the custody or under the control of the
accused’); [64] sexual violence (which includes ‘rape, sexual slavery, enforced
prostitution, forced pregnancy, enforced sterilization, or any other form of sexual
violence of comparable gravity’); [65] persecution [66] (being ‘the intentional and severe
deprivation of fundamental rights contrary to international law by reason of the identity
of the group or collectivity’); [67] enforced disappearance [68] (being ‘the arrest,
detention or abduction of persons by, or with the authorisation, support or
acquiescence of, a State or a political organisation, followed by a refusal to
acknowledge that deprivation of freedom or to give information on the fate or
whereabouts of those persons, with the intention of removing them from the protection
of the law for a prolonged period of time’); [69] the crime of apartheid [70] (which
includes ‘inhumane acts of a character similar to [other crimes
Page 257
against humanity], committed in the context of an institutionalized regime of
systematic oppression and domination by one racial group over any other racial group
or groups and committed with the intention of maintaining the regime’; [71] and other
inhumane acts [72] that are acts ‘of a similar character to [other crimes against
humanity] intentionally causing great suffering, or serious injury to body or to mental
or physical health’). [73]
The Rome Statute then adds a chapeau to these underlying acts to elevate them to
the level of crimes against humanity. This chapeau is that any conduct listed in article
7(1) of the Rome Statute must be committed as part of a widespread or systematic
attack directed against any civilian population, with knowledge of the attack. An
‘“[a]ttack directed against any civilian population” means a course of conduct involving
the multiple commission of acts referred to in [article 7(1)] against any civilian
population, pursuant to or in furtherance of a State or organisational policy to commit
such attack’. [74]
The first requirement of the chapeau is that the act complained of must be part of a
widespread or systematic attack. Article 7(2) provides elucidation when it says that an
attack is ‘a course of conduct involving the multiple commission of acts referred to in
[article 7(1)] against any civilian population, pursuant to or in furtherance of a state or
organizational policy to commit such attack’. The Elements of Crimes say that ‘“policy
to commit such attack” requires that the State or organization actively promote or
encourage such an attack against a civilian population’. [75] There is an ongoing debate
in international law on whether the requirement of a state or organisational policy to
commit an attack in the Rome Statute is at odds with customary international
law. [76] The second requirement is that the attack must be directed against a civilian
population. This distinguishes it from many war crimes which may be targeted at both
civilians and combatants, and the requirement also distinguishes the Rome Statute
from customary international law, which accepts that a crime against humanity may be
committed against civilians and military personnel. Lastly, a crime against humanity
cannot be committed unless a specific form of intention is present. Article 7(1)
Page 258
provides that a ‘crime against humanity’ means any of the enumerated acts when
committed as part of a widespread or systematic attack directed against any civilian
population, ‘with knowledge of the attack’. This requirement amounts to a form of
specific intent, which sets another threshold that must be crossed before a particular
offence can be regarded as a crime against humanity. [77]

3 War crimes

War crimes have an ancient lineage and historically belligerent states took it upon
themselves to determine those acts committed in time of war for which they would try
the combatants or civilians belonging to the enemy. Of the core crimes in the Rome
Statute, ‘war crimes’ were the first to have been prosecuted at international law.
German soldiers were convicted of ‘acts in violation of the laws and customs of war’ at
Leipzig in the early 1920s, pursuant to articles 228 and 230 of the Treaty of
Versailles. [78]
Generally speaking, war crimes are crimes committed in violation of international
humanitarian law applicable during armed conflicts. The sources of international
humanitarian law are vast, and are broadly divided into two categories of substantive
rules—’the law of The Hague’ [79] and ‘the law of Geneva’ [80]—and which constitute the
Page 259
rules concerning behaviour that is prohibited in the case of an armed conflict.
Drawing extensively from these existing sources of humanitarian law, the drafters of
the Rome Statute in article 8 have set out an elaborate ‘codification’ of the rules
concerning behaviour that is prohibited in situations of armed conflict. Various
preconditions for a war crimes prosecution are built into the Statute. First, in order to
constitute a violation of article 8 of the Rome Statute, there must be a nexus between
the criminal conduct and the armed conflict. War crimes may be committed during
either international or internal armed conflicts, [81] with states parties to the Rome
Statute having accepted that responsibility for war crimes can be founded during times
of civil war. Secondly, the Rome Statute directs the Court’s attention ‘in particular’ to
those war crimes that are ‘committed as part of a plan or policy or as part of a large-
scale commission of such crimes’. [82] This so-called ‘non-threshold threshold’ built into
article 8 ensures that two jurisdictional triggers—(1) that the war crime is committed as
part of a plan or policy, and (2) that the war crime is committed alongside other war
crimes on a large scale—should ordinarily be met before the ICC will be seized with the
case. Thirdly, article 30 of the Statute provides that to found criminal responsibility for
a war crime requires intent and knowledge: intent in relation to the conduct, namely,
that the person means to engage in the conduct;
Page 260
and knowledge in relation to the consequence, namely, that the person means to cause
that consequence or is aware that it will occur in the ordinary course of events.
Drawing on the Geneva Conventions and international humanitarian law, the Statute
adopts a four-part division to its elucidation of ‘war crimes’—the first two divisions
cover war crimes committed during an international armed conflict; the last two
divisions cover war crimes committed during an internal armed conflict.

(a) War crimes in times of international armed conflict

Grave breaches of the Geneva Conventions (article 8(2)(a))

Article 8(2)(a) of the Rome Statute provides that any of the following acts (’grave
breaches’) committed during an international armed conflict against persons or
property protected under the provisions of the relevant Geneva Convention will amount
to a war crime:
•
wilful killing;
•
torture or inhuman treatment, including biological experiments;
•
wilfully causing great suffering, or serious injury to body or health;
•
 extensive destruction and appropriation of property, not justified by military
necessity and carried out unlawfully and wantonly;
•
compelling a prisoner of war or other protected person to serve in the forces of a
hostile power;
•
wilfully depriving a prisoner of war or other protected person of the rights of fair
and regular trials;
•
unlawful deportation or transfer or unlawful confinement;
•
taking of hostages.
The persons protected by article 8(2)(a) are combatants who are considered hors de
combat because of injury, shipwreck or illness, or because they have been taken as
prisoners of war, and civilians. The notion of ‘protected property’ is not defined in any
of the Geneva Conventions but is generally regarded as property found in territories
occupied by foreign forces. Such property, which would include medical units and
establishments, medical transports, and hospital ships, amongst others, may not be
destroyed except in cases of military necessity. [83]

Other serious violations of the laws and customs applicable in international armed
conflict, within the established framework of international law (article 8 (2)(b))
Article 8(2)(b) sets out the second category of war crimes and which are limited to
international armed conflict. The ‘serious violations of
Page 261
the laws and customs applicable in international armed conflict’ are generally drawn
from the law of The Hague. Unlike the focus of the grave breaches crimes under article
8(2)(a), which aim to protect the innocent victims of war or those who are hors de
combat, the focus of the crimes under article 8(2)(b) is on the combatants themselves.
These crimes are a continuation of ancient rules of chivalry reflecting a code of conduct
amongst warriors. [84] As a general overview, article 8(2)(b) of the Rome Statute
includes prohibitions on attacks against the civilian population, [85] attacks against
civilian objects, [86] as well as attacks that violate the principle of proportionality [87] and
attacks against undefended places. [88] Civilians are also protected against ‘misuse’, for
instance, the use of civilians or protected persons as a means to render certain points
or areas immune from military operations. [89] The starvation of civilians as a method of
warfare is prohibited, as is any attack against objects indispensable to the survival of
the civilian population. [90] The ‘destruction of property’ is outlawed in that the
destruction or seizing of the enemy’s property is considered a war crime unless such
destruction or seizure is imperatively demanded by the necessities of war. [91] The
improper use of signs and perfidy is rendered a war crime, [92] and there is a prohibition
on killing or wounding persons who are hors de combat. [93] Lastly, there is a prohibition
placed on declaring that no quarter will
Page 262
be given; that is, ordering that there shall be no survivors, threatening an adversary
therewith, or conducting hostilities on this basis. [94]
Several of the provisions of article 8(2)(b) deal with prohibited weapons—for
example, poison or poisoned weapons, [95] poisonous gases and all analogous liquids,
materials or devices, [96] and dum-dum bullets [97]—and render their use a war crime.
 In addition to the provisions reflecting The Hague Rules, there are some ‘new crimes’
under para (b) and which have now been codified by the drafters at Rome. They cover,
for instance, the protection of humanitarian and peacekeeping missions [98] and prohibit
environmental damage. [99] Another new war crime under the Statute is the conscription
or enlistment of children under the age of 15 into the national armed forces or to use
them to participate actively in hostilities. [100] Another development relates to ‘sexual
crimes’. In terms of article 8(2)(b)(xxii) it is a war crime to commit rape, sexual
slavery, enforced prostitution, forced pregnancy, [101] enforced sterilisation or any other
form of sexual violence also constituting a grave breach of the Geneva
Conventions. [102] While the terms rape and enforced prostitution already appear in the
Fourth Geneva Convention and Protocol I of 1977, the outlawing of ‘sexual slavery’,
‘forced pregnancy’ and ‘enforced sterilization’ are essentially new crimes.

(b) War crimes in times of non-international armed conflict

Violations of Common Article 3 of the Geneva Conventions (Rome Statute article

8(2)(c) and (d))
The criminal acts proscribed by this section are the Common Article 3 [103] crimes of
violence to life and person, in particular murder of all
Page 263
kinds, mutilation, cruel treatment and torture; committing outrages upon personal
dignity, in particular humiliating and degrading treatment; taking of hostages; and the
passing of sentences and the carrying out of executions without previous judgment
pronounced by a regularly constituted court affording all judicial guarantees that are
generally recognised as indispensable. The prohibited acts listed under article 8(2)(c) of
the Rome Statute, like the ‘grave breaches’ in article 8(2)(a), are acts that are
committed against ‘protected persons’. Such ‘protected persons’ are described in article
8(2)(c) as ‘persons taking no active part in the hostilities [civilians], including members
of armed forces who have laid down their arms and those placed hors de combat by
sickness, wounds, detention or any other cause’.
These standards represent a ‘common denominator of core human rights’ [104] that
must be respected by those engaged in hostilities, whether they are engaged in an
international or non-international armed conflict. While the prohibitions apply to armed
conflicts ‘not of an international character’, the ICC Statute provides that these
protections do not extend ‘to situations of internal disturbances and tensions, such as
riots, isolated and sporadic acts of violence or other acts of a similar
nature’. [105] Internal disturbances and acts of terrorism that do not amount to an
armed conflict are therefore not subject to the laws of armed conflict at all, although
the state (but not the rebels) will be subject to the provisions of any human rights
treaties to which the state is a party.

Other serious violations of the laws and customs applicable in armed conflicts not of an
international nature (Rome Statute article 8(2)(e))
Protocol II of 1977 to the Geneva Conventions largely serves as the inspiration for the
prohibitions contained in article 8(2)(e) of the Rome Statute. [106] The article prohibits
attacks against the civilian population, [107] ‘killing or wounding treacherously a
combatant adversary’, [108] declaring that no quarter will be given, [109] or destroying or
seizing the property of an adversary unless such destruction or
Page 264
seizure is imperatively demanded by the necessities of the conflict, [110] and ‘pillaging a
town or place, even when taken by assault’. [111]
The following special protections are included under article 8(2)(e): intentional
attacks directed against buildings, materials, medical units and transport, and
personnel using the distinctive emblems of the Geneva Conventions in conformity with
international law, are prohibited; [112] intentional attacks directed against personnel,
installations, material, units or vehicles involved in a humanitarian assistance or
peacekeeping mission in accordance with the Charter of the United Nations is prohibited
so long as they are entitled to the protection given to civilians or civilian objects under
the international law of armed conflict; [113] and intentional attacks against buildings
dedicated to religion, education, art, science or charitable purposes, historic
monuments, hospitals and places where the sick and wounded are collected are
prohibited, provided they are not military objectives.
In addition, the following provisions serve to protect against violations of human
rights more generally. Article 8(2)(e)(xi) makes it a war crime to subject persons who
are in the power of another party to the conflict to physical mutilation or to medical or
scientific experiments of any kind which are neither justified by the medical, dental or
hospital treatment of the person concerned nor carried out in his or her interest, and
which cause death to or seriously endanger the health of such person or persons.
Article 8(2)(e)(vi) prohibits rape, sexual slavery, enforced prostitution, forced
pregnancy, enforced sterilisation, and any other form of sexual violence also
constituting a serious violation of Common Article 3 to the Geneva Conventions, while
article 8(2)(e)(vii) outlaws conscripting or enlisting children under the age of 15 years
into armed forces or groups or using them to participate actively in hostilities. Article
8(2)(e)(viii) makes it a war crime to order the displacement of the civilian population
for reasons related to the conflict, unless the security of the civilians involved or
imperative military reasons so demand.
The prohibitions contained in article 8(2)(e) of the ICC Statute apply to armed
conflicts not of an international character but not to ‘situations of internal disturbances
and tensions, such as riots, isolated and sporadic acts of violence or other acts of a
similar nature’. [114]
Page 265

4 Aggression

Although aggression was dubbed the ‘supreme international crime’ by the Nuremberg
Tribunal as far back as 1946, [115] there has only recently been agreement on the
definition of the crime of aggression for the purposes of the Rome Statute.
Article 5 of the Rome Statute lists the crime of aggression as one of the core crimes
under the Court’s jurisdiction. However, in contrast to the other three crimes (genocide,
crimes against humanity and war crimes) the Court remained unable to exercise
jurisdiction over the crime of aggression as the Statute did not define the crime or set
out jurisdictional conditions. [116]
After two weeks of intense debate and years of preparatory work, on 11 June 2010,
the Review Conference of the Rome Statute (held in Kampala, Uganda) adopted by
consensus amendments to the Rome Statute which include a definition of the crime of
aggression. [117] Far more controversially, the Conference also established the
conditions under which the Court is empowered to exercise jurisdiction in respect of the
crime. [118]
On 14 December 2017, the ICC’s jurisdiction over the crime of aggression was
activated as of 17 July 2018 [119] for ICC member states which have ratified or accepted
the amendment to the Rome Statute. It also stipulates that the ICC will not have
jurisdiction over ICC member states, or their nationals, that have not ratified or
accepted these amendments in the case of a state referral or proprio motu (initiated by
the ICC prosecutor) investigation.
The Kampala amendment is a complex legal provision which is further complicated
by numerous ‘understandings’ that accompanied its adoption. [120] The ‘understandings’
of Kampala are purportedly designed to enable the Court to ‘dismiss frivolous, or
politically
Page 266
motivated, allegations of aggression and to protect military missions based on self-
defence, humanitarian intervention, or other legitimate purposes consistent with the UN
Charter’. [121]
It is generally accepted that prosecuting the crime of aggression consists of two
parts: the definition of the crime, and the manner in which the Court may exercise
jurisdiction over it.

(a) The definition of the crime

Broadly speaking, the conduct prohibited by the crime of aggression is the unlawful and
unjustified use of armed force against the territory of another state. [122] According to
article 8bis(1), the crime of aggression contains an important inbuilt threshold, since it
means the planning, preparation, initiation or execution of an act of aggression that, ‘by
its character, gravity and scale, constitutes a manifest violation of the Charter of the
United Nations’. Furthermore, in order to qualify as the crime of aggression, the act of
aggression must be performed by a person ‘in a position effectively to exercise control
over or to direct the political or military action of a State’. [123] This second qualification
thus retains the notion, held at Nuremberg, that aggression is a ‘leadership crime’ that
cannot be committed ‘by minions and footsoldiers’. [124] An ‘act of aggression’ means
‘the use of armed force by a State against the sovereignty, territorial integrity or
political independence of another State, or in any manner inconsistent with the Charter
of the United Nations’. [125] Article 8bis(2) then lists the various acts, following United
Nations General Assembly Resolution 3314 (XXIX) of 14 December 1974, which qualify
as acts of aggression:
Page 267
(a)
the invasion or attack by the armed forces of a state of the territory of another
state, or any military occupation, however temporary, resulting from such
invasion or attack, or any annexation by the use of force of the territory of
another state or part thereof;
(b)
bombardment by the armed forces of a state against the territory of another
state or the use of any weapons by a state against the territory of another state;
(c)
the blockade of the ports or coasts of a state by the armed forces of another
state;
(d)
an attack by the armed forces of a state on the land, sea or air forces, or marine
and air fleets of another state;
(e)
the use of armed forces of one state that are within the territory of another state
with the agreement of the receiving state, in contravention of the conditions
provided for in the agreement, or any extension of their presence in such
territory beyond the termination of the agreement;
(f)
the action of a state in allowing its territory, which it has placed at the disposal of
another state, to be used by that other state for perpetrating an act of
aggression against a third state;
(g)
the sending by or on behalf of a state of armed bands, groups, irregulars or
mercenaries that carry out acts of armed force against another state of such
gravity as to amount to the acts listed above, or its substantial involvement
therein.
Importantly, the listed acts constitute acts of aggression ‘regardless of a declaration of
war’. While Resolution 3314 (XXIX) is open-ended, the Court will have to interpret
whether the wording of article 8bis(2) leaves room for the extension of aggression to
other acts of aggression. [126]
Even if the particular conduct is shown to be an ‘act of aggression’ that accords with
one of the listed items, it must meet the further requirement of constituting a ‘manifest
violation of the Charter of the United Nations’ when examining the character, gravity
and scale of the act of aggression. Criticism has been levelled at the lack of clear
guidance on how criteria such as ‘its character, gravity and scale’ should be
interpreted. [127]
Calls to extend the definition of an act of aggression to include acts by non-state
entities (for example, terrorist armed groups, liberation movements and organised
insurgents) against a state were not
Page 268
accommodated by the definition. [128] Accordingly, conduct by non-state entities cannot
constitute aggression.

(b) Jurisdiction for the crime of aggression

Perhaps the most controversial issue surrounding aggression has been the jurisdictional
triggers of the crime, in particular, the Security Council’s role in the prosecution of the
crime. [129] Indeed, there had been so much debate around the issue that the Special
Working Group on the Crime of Aggression’s working document, leading into the Review
Conference in Kampala, proposed six alternative formulations of how jurisdiction would
operate. [130] The version adopted in the Kampala Amendment set up two distinct
schemes for the prosecution of aggression. [131] The first allows states and the ICC
prosecutor to trigger a prosecution (article 15bis) and the second provides for the
Security Council to do so (article 15ter).

State referral and proprio motu prosecutions

The Court may exercise jurisdiction over a crime of aggression ‘arising from an act of
aggression committed by a state party, unless that state party has previously declared
that it does not accept such jurisdiction by lodging a declaration with the
Registrar’. [132] It has been argued that this exclusion provision creates an ‘asymmetry’
amongst states parties since it allows a state party to displace the Court’s jurisdiction
over their acts committed against other states parties by making a declaration opting
out of jurisdiction. However, it still affords them protection against acts of aggression
committed against them by states parties which have not made such a
declaration. [133] Ferencz argues that this asymmetry may potentially be at odds with
the conventional treaty law notion of a ‘contract among consenting states’. [134]
Accordingly, the ICC can prosecute crimes of aggression only if the offending state
has consented to the Court’s jurisdiction—both by signing the Rome Statute and by
subsequently not entering a
Page 269
declaration vitiating the Court’s jurisdiction over aggression prior to the act in
question. [135] The Court also has no jurisdiction over the crime of aggression when
committed by nationals of or on the territory of a state that is not a party to the Rome
Statute. [136]
Earlier drafts of the amendment proposed that there must have been a ‘pre-
determination made, of the commission of a state act of aggression, by the United
Nations Security Council’. [137] The adopted amendment does not require a Security
Council determination for the Court’s jurisdiction to be triggered. However, before
proceeding with any investigation the prosecutor must first ascertain whether the
Security Council has ‘made a determination of an act of aggression committed by the
State concerned’. [138] If no such determination has been made within six months of the
prosecutor informing the Secretary-General of the UN of his intention to launch an
investigation, the prosecutor may proceed with such an investigation ‘provided that the
Pre-Trial Division of the Court (with the approval of six judges) has authorised the
commencement of the investigation . . . and the Security Council has not decided
otherwise in accordance with article 16’. [139] Some commentators have referred to this
mechanism as an ‘internal filter’ against ‘politicized’ allegations. [140]

Referral by the Security Council

Article 15ter of the Statute provides that the Court ‘may exercise jurisdiction over the
crime of aggression in accordance with article 13, paragraph (b), subject to the
provisions of this article’. However, the only real proviso is the general clause in relation
to aggression providing that the Court may only exercise jurisdiction after the
amendment has been ratified by 30 states and not before 1 January 2017. This proviso
has been satisfied. Accordingly, the Security Council’s authority to refer a matter to the
Court in respect of aggression will be no different to that which it has in respect of the
Page 270
other crimes. [141] The Security Council’s power to refer an act of aggression to the
Court is furthermore not ‘conditioned by the requirement of any state’s consent to the
Kampala amendments’. [142]
Few had thought that consensus in Kampala ‘on a comprehensive package on the
crime of aggression’ was likely or even feasible. [143] Thus, notwithstanding the
jurisdictional shortcomings and complexities canvassed above, the attitude of most
commentators towards the amendment has been one of ‘cautious optimism’. [144]

Jurisdiction
The Rome Statute strictly defines the jurisdiction of the Court. Aside from having
jurisdiction only over the most serious crimes of concern to the international
community, the temporal jurisdiction of the Court is limited to crimes occurring after
the entry into force of the Statute on 1 July 2002. [145] For those states that become
party to the Statute after 1 July 2001, the ICC has jurisdiction only over crimes
committed after the entry into force of the Statute with respect to that state. [146] Thus,
the Court is not a remedy for crimes of the past, which must be addressed by national
or other international or hybrid initiatives.
The jurisdictional triggers for the Court to exercise its competence are set out in
article 12 of the Statute. This article provides that the Court may exercise jurisdiction
if: (a) the state where the alleged crime was committed is a party to the Statute
(territoriality); or (b) the state of which the accused is a national is a party to the
Statute (nationality). In terms of article 14 of the Statute, any state party may refer to
the Court a ‘situation’ in which one or more crimes within the jurisdiction of the Court
appear to have been committed, so long as the preconditions to the Court’s exercise of
jurisdiction have been met, namely, that the alleged perpetrators of the crimes are
nationals of a state party or the crimes are committed on the territory of a state
Page 271
party. [147] There has been considerable debate on whether this article allowed a state
party to refer alleged crimes committed in its own territory or whether the mechanism
was simply intended as an inter-state referral mechanism. [148] However, the decision of
the ICC Appeals Chamber in the Katanga Case has now clarified that self-referrals are
permissible under article 14 of the Statute and that such referrals are indeed consistent
with the Rome Statute’s ‘object and purpose of eradicating impunity for international
crimes’. [149] There have been self-referrals of situations to the ICC by Uganda, the
Central African Republic, the Democratic Republic of the Congo and, most recently,
Mali. [150]
The ICC prosecutor is also authorised by article 15 of the Rome Statute to initiate
independent investigations on the basis of information received from any reliable
source. The granting to the prosecutor of a proprio motu power to initiate investigations
was one of the most debated issues during the negotiations of the Rome Statute. In the
end, the drafters of the Statute determined that in order for the prosecutor to exercise
this power, the alleged crimes must have been committed by nationals of a state party
or have taken place in the territory of a state party—the preconditions set out in terms
of article 12. [151]
Proposals that the principle of universal jurisdiction should apply in respect of state
referrals were rejected at the Rome Conference. That being said, under the Statute, the
UN Security Council is empowered to refer to the Court ‘situations’ in which crimes
within the jurisdiction of the Court appear to have been committed. [152] The referral
power is a mechanism by which the Court is accorded jurisdiction over an offender,
regardless of where the offence took place and by whom it was committed, and
regardless of whether the state concerned has ratified the Statute or accepted the
Court’s
Page 272
jurisdiction. [153] The Statute provides that the Council may only make such a referral by
acting under Chapter VII of the United Nations Charter, which means that it must
regard the events in a particular country as a threat to the peace, a breach of the
peace, or an act of aggression. In determining whether a ‘threat to the peace’ exists,
the Council will be guided by the gravity of the crimes committed, the impunity enjoyed
by the crimes’ perpetrators and the effectiveness or otherwise of the national
jurisdiction in the prosecution of such crimes. [154] Having had regard to these factors, in
March 2005 the Security Council referred the atrocities committed in the Darfur region
of Sudan to the ICC for investigation. [155] Then, on 26 February 2011, the Security
Council unanimously referred the situation in Libya to the ICC. [156] So far, these are the
only two Security Council referrals to the Court, both involving African non-members of
the ICC. The referrals have helped to heighten critical sentiment of the Court,
particularly from African states [157] (as we discuss further below).

Admissibility

The International Criminal Court is not expected to supersede national prosecutions of

persons guilty of international crimes. Investigations and prosecutions under the Rome
Statute are premised on the principle of ‘complementarity’, whereby the Court is
required to rule a case inadmissible when it is being appropriately dealt with by a
national justice system. [158] States parties to the Court therefore retain their right and
responsibility to investigate offences committed on their territory, or where their
nationals stand accused
Page 273
of committing ICC crimes anywhere else in the world. The ICC will be able to step in
only where a national judicial system is unwilling or unable to genuinely
investigate. [159] The principle of complementarity ensures that the ICC operates as a
system of international criminal justice which buttresses the national justice systems of
states parties. It is ‘an attempt to balance the principle of state sovereignty and the
need to establish an international regime that effectively intervenes when states fail’ to
do so. [160]
What about amnesties that are accorded by a national state in lieu of prosecution?
For centuries, successor regimes have sought to secure peace through the pardoning of
their enemies, and modern history is replete with examples where a regime has
granted amnesty to officials of the previous regime who were guilty of torture and
crimes against humanity, rather than prosecuted them (eg Uruguay, Argentina and El
Salvador). So, too, there are examples of outgoing regimes which used their last days
of political power to ensure that their members were granted an official ‘pardon’ from
prosecution before the new regime took office (eg Chile). With the advent of truth
commissions, it has become possible to channel the grant of amnesty through the
commission. So far, only the South African Truth and Reconciliation Commission (TRC)
and the recent truth commission in East Timor have been accorded the power to grant
amnesty. [161] As the South African experience demonstrates, the prospect of amnesty
in exchange for truth is a good incentive to the guilty to provide
Page 274
detailed accounts of the acts they have committed. [162] In any event, the political
reality for many transitional governments is that giving a truth commission the power
of amnesty rather than criminally prosecuting past offenders is the only realistic and
peaceful way in which an existing regime will be persuaded to relinquish power.
Whatever the form of amnesty (whether it be granted by a truth commission or by the
outgoing or ingoing government as a political act of reprieve) the question to be
confronted is how such trumps to prosecution are to be dealt with by the ICC.
 The Rome Statute is silent on amnesty, and commentators argue that this is because
the Rome Statute was never drafted with the intention of allowing amnesty to trump
the Court’s jurisdiction. [163] Assuming therefore that the relevant jurisdictional
requirements for an ICC prosecution are met, national amnesties granted by a truth
commission or by governmental sleight of hand would not per se prevent action by the
ICC. [164]
While amnesties do not in principle bar the ICC from exercising criminal jurisdiction
over an individual who has been granted amnesty, the political reality is that in some
instances it might be expedient or a requirement of justice not to push ahead with the
prosecution of such a person. Article 53(2)(c) of the Rome Statute therefore provides
the prosecutor with a discretion to refuse prosecution at the instance of a state or the
Security Council where, after investigation, he concludes that ‘a prosecution is not in
the interests of justice, taking into account all circumstances’. Of course, the type of
amnesty at issue will play an important role in the prosecutor’s decision. No clear rules
can be enunciated to distinguish between permissible and impermissible amnesties
under international law, but it has been suggested that ‘international recognition might
be accorded where amnesty has been granted as part of a truth and reconciliation
inquiry and each person granted amnesty has been obliged to make full disclosure of
his or her criminal acts as a precondition of amnesty and the acts were politically
motivated’. [165] The blanket amnesty in Chile passed by the Pinochet regime would
Page 275
thus not meet the required standard (in the Pinochet case, [166] before the House of
Lords, it was not even argued by Pinochet’s lawyers that his amnesty in Chile could
constitute a bar to his extradition from Britain to Spain) [167] while the South African
amnesties, granted by a quasi-judicial amnesty committee functioning as part of a TRC
process established by a democratically elected government, may well do so. [168] It is
also important to note that the nature of certain offences precludes the grant of
amnesty to their perpetrators. [169] It is still open to states to grant amnesty for
international crimes without violating a rule of international law, but international
lawyers are largely in agreement that states are not permitted to grant amnesty for the
crimes of genocide, torture and ‘grave breaches’ under the Geneva Convention. [170] The
preamble of the Statute of the ICC, while binding only in respect of parties to it,
confirms this trend when it declares that it is the duty of every state to exercise
criminal jurisdiction over those responsible for international crimes. [171]
As a result, whatever form of amnesty the prosecutor is forced to consider, it is clear
that she will be more disposed towards those amnesties that have been limited in terms
of the nature of the offence
Page 276
(at the very least it appears that amnesty afforded for the international crimes of
torture and genocide will be disregarded) and which have been granted as part of a
truth and reconciliation inquiry, in which amnesty recipients have been obliged to make
full disclosure of their criminal acts as a precondition of amnesty and to prove that their
acts were politically motivated. [172]

Situations before the ICC

Africa is high on the Court’s agenda. All of the ICC’s convictions have been for crimes
committed by Africans in Africa. Ten of the eleven situations before the ICC involve
African states and leaders. Of these ten, there have been two important
convictions, [173] two acquittals [174] and numerous arrest warrants.
However, the ICC is currently investigating situations around the globe. The Office of
the Prosecutor is investigating situations in, for example, Afghanistan, Colombia,
Honduras, Iraq/the UK, Mexico, Palestine [175] and Korea. [176] Whether prosecutions will
result from these investigations remains to be seen.
A situation that deserves particular mention is the one in Sudan. On 31 March 2005,
the UN Security Council passed Resolution 1593, referring the prosecution of those
allegedly responsible for the numerous atrocities committed in the Darfur region in
Sudan to the ICC. [177] On 4 March 2009, Pre-Trial Chamber I issued a warrant of arrest
against Sudanese President Omar al-Bashir for his alleged responsibility under article
25(3)(a) of the Statute for the crimes
Page 277
against humanity and war crimes alleged by the prosecution. [178] A second warrant of
arrest was then issued on 12 July 2010, which included numerous counts of
genocide. [179] Al-Bashir is still at large. Various countries have failed to arrest him,
including South Africa. The ICC has consistently held that failure to arrest al-Bashir is in
contravention of international law. [180]
The Sudan referral has been the subject of much debate and has highlighted
numerous contentious issues in international law. Chiefly, it is the first time a sitting
president has been investigated for international crimes before the ICC. [181] This has
raised the issue of immunity, both before the ICC and domestic tribunals. The issue of
immunities is discussed later in this chapter under its own heading.

1 An international criminal court for the ‘usual

suspects’? [182]

International criminal law has sometimes been criticised for ‘providing victors in a
conflict with an opportunity to demonise their opponents, sanitise their crimes and
perpetuate injustice’. [183] Similarly, since the ICC was established there have been
concerns that the
Page 278
Court has only concentrated on the ‘usual suspects’, with some arguing that it has
illustrated a bias towards prosecuting situations in Africa while neglecting similar
violations of the Rome Statute on other continents. [184] They are captured in
statements to the effect that the ICC is a ‘hegemonic tool of western powers which is
targeting or discriminating against Africans’. [185] On the other hand, there are concerns
that this ‘rhetoric of condemnation’ (that the ICC is ‘anti-African, and merely an agent
of neocolonialism or neo-imperialism’) may damage the institution to such an extent
that it is simply abandoned. [186]
These concerns are borne out by the reported requests by Kenya to withdraw from
the Rome Statute, by Burundi’s withdrawal from the ICC, and by South Africa’s recent
attitude towards its ICC obligations. [187] They are also evidenced in the position
adopted within the African Union (AU) in response to the ICC’s investigation of
President al-Bashir of Sudan. While the ICC warrant of arrest for al-Bashir was
welcomed by human rights organisations, [188] the AU called on the Security Council to
defer the ICC’s investigation into al-Bashir by invoking article 16 of the Rome Statute,
which allows for a suspension of prosecution or investigation for a period of up to 12
months. [189] On 3 July 2009, at an AU meeting in Sirte, Libya, the AU
Page 279
took a resolution (the Sirte Resolution) calling on its members to defy the international
arrest warrant issued by the ICC for al-Bashir. [190]
This AU decision placed African states party to the Rome Statute in the ‘unenviable
position of having to choose between their obligations as member states of the AU on
the one hand, and their obligations as states party to the Rome Statute, on the
other’. [191] To date, even though al-Bashir is the subject of an arrest warrant by the
ICC, several states have failed to enforce the warrant after inviting al-Bashir to visit
their territory. [192]
The international law questions arising (eg of immunities, commitment to the ICC by
African states, Security Council and ICC reform, and the politics of international criminal
justice) have generated a plethora of scholarly writing. [193] The issue continues to
simmer, with the AU taking a decision in January 2018, inter alia, to ‘reiterate its
previous decisions on the deferral or termination of proceedings against President Omar
Al-Bashir of the Republic of the Sudan in accordance with Article 16 of the Rome
Statute’ and ‘to immediately place on the agenda of the United Nations General
Assembly a request to seek an advisory opinion from the International Court of Justice
on the question of immunities of a Head of State and Government and other Senior
Officials as it relates to the relationship between Articles 27 and 98 and the obligations
of States Parties under International Law’. [194]
The friction is of continuing and serious concern, and raises
Page 280
debates about the effectiveness of the international criminal justice project more
generally, reform of the ICC and the Security Council’s relationship with the Court, and
a sober reflection of how to resolve the tensions between the AU and the ICC. [195]

South Africa’s implementation of the Rome Statute

In order to give effect to its complementarity obligations under the Rome Statute,
South Africa incorporated the Rome Statute into its domestic law by means of the
Implementation of the Rome Statute of the International Criminal Court Act 27 of 2002
(the ‘ICC Act’). [196] Prior to the ICC Act, South Africa had no municipal legislation on
the subject of war crimes or crimes against humanity, [197] and no domestic
prosecutions of international crimes had taken place in South Africa.
Under the ICC Act, a structure is created for national prosecution of crimes in the
Rome Statute. The Act takes seriously the ‘complementary’ obligation on South African
courts to domestically investigate and prosecute the ICC offences of crimes against
humanity, war crimes and genocide. The preamble, for instance, speaks of South
Africa’s commitment to bring ‘persons who commit such atrocities to justice . . . in a
court of law of the Republic in terms of its domestic law where possible’. And s 3 of the
Act defines as one of its objects to enable, ‘as far as possible and in accordance with
the principle of complementarity . . . the national prosecuting authority of the Republic
to adjudicate in cases brought against any person accused of having committed a crime
in the Republic and beyond the borders of the Republic in certain circumstances’. Like
the Rome Statute, the
Page 281
ICC Act does not reach back into the past. The Act provides expressly that ‘[n]o
prosecution may be instituted against a person accused of having committed a crime if
the crime in question is alleged to have been committed before the commencement of
the Statute’. [198]

1 Incorporation of ICC crimes

The advantage of the Rome Statute of the International Criminal Court is that it brings
together in one place a codified statement of the elements which make up the crimes of
genocide, war crimes and crimes against humanity. The drafters of the ICC Act, aware
of this benefit of codification, incorporated the ICC Statute’s definitions of the core
crimes directly into South African law through a schedule appended to the Act. In this
regard, Part 1 of Schedule 1 to the ICC Act follows the wording of article 6 of the ICC
Statute in relation to genocide; Part 2 of the Schedule mirrors article 7 of the Statute in
respect of crimes against humanity; and Part 3 does the same for war crimes, as set
out in article 8 of the ICC Statute. It is clear that these crimes now form part of South
African law through the Act. One of the objects of the Act is ‘to provide for the crime of
genocide, crimes against humanity and war crimes’, [199] and s 4(1) of the Act provides
that ‘[d]espite anything to the contrary in any other law in the Republic, any person
who commits a crime [defined as genocide, crimes against humanity and war crimes],
is guilty of an offence’. The ICC Act, at this stage, requires amendment in order to
provide for the definition and jurisdictional regime for the crime of aggression following
the Kampala amendment discussed earlier.
While the Act usefully incorporates the definitions of these crimes into South African
domestic law, neither the ICC Act nor Schedule 1 refers specifically to article 9 of the
Rome Statute on Elements of Crimes. [200] There is nothing, however, that prevents a
South African court from having regard to the Elements of Crimes, were it to be
involved in the domestic prosecution of an ICC offence. However, in the interests of
clarity and completeness, it is suggested that South
Page 282
Africa follow the example of other states parties [201] and incorporate, by regulation, the
Elements of Crimes. [202]

2 Grounds of jurisdiction

Section 4(1) of the ICC Act creates jurisdiction for a South African court over ICC
crimes by providing that ‘[d]espite anything to the contrary in any other law of the
Republic, any person who commits [an ICC] crime, is guilty of an offence and liable on
conviction to a fine or imprisonment’. Section 4(3) of the Act goes further and provides
for extraterritorial prescriptive jurisdiction. [203] In terms of that section, the jurisdiction
of a South African court will be triggered when a person commits an ICC crime outside
the territory of the Republic and that person:
(a)
is a South African citizen; or
(b)
is not a South African citizen but is ordinarily resident in the Republic; or
(c)
after the commission of the crime, is present in the territory of the Republic; or
(d)
 has committed the said crime against a South African citizen or against a person
who is ordinarily resident in the Republic.
When a person commits a core crime outside the territory of the Republic in one of
these four circumstances, s 4(3) deems that crime to have been committed in the
territory of the Republic.
The jurisdictional ‘triggers’ in the ICC Act are largely uncontroversial. Section 4(1)
appears to assert the traditional principle of territoriality; namely, that a state has
competency in respect of all acts which occur in its territory. Section 4(3), which
provides for extraterritoriality, begins in trigger (a) with the recognised nationality basis
for prescriptive jurisdiction. That is, international law has long accepted that states
have the competency to exercise prescriptive jurisdiction over their nationals for crimes
committed anywhere in the world. Trigger (b) extends, in similar fashion, prescriptive
jurisdiction over South African residents on the basis that they have a close and
substantial connection with South Africa at the time of the offence. Trigger (c) of the
ICC Act extends prescriptive jurisdiction to a person who, ‘after the commission of the
crime, is present in the
Page 283
territory of the Republic’. There is no mention here of the person’s nationality or
residency, and one must assume, given that triggers (a) and (b) already provide
jurisdiction in respect of crimes committed abroad by South African nationals and
residents, that trigger (c) is referring to individuals who commit a core crime and who
do not have a close and substantial connection with South Africa at the time of
offence. [204] The jurisdiction in trigger (c) is thus grounded on the idea of universal
jurisdiction; that is, prescriptive jurisdiction which exists for all states in respect of
certain crimes that attract universal prescriptive jurisdiction by their egregious nature,
and consequently over the perpetrators of such crimes on the basis that they are
common enemies of humankind. This form of prescriptive jurisdiction is to be welcomed
because genocide, crimes against humanity and war crimes are among the crimes of
most serious concern to the international community as a whole, and as such, are often
regarded as giving rise to ‘universal jurisdiction’. [205] Trigger (d) is founded on the
passive personality principle in international law. In terms of that principle, a state has
the competency to exercise prescriptive jurisdiction over an individual who causes harm
to one of its nationals abroad.
The ICC Act provides that a South African court, charged with the prosecution of a
person allegedly responsible for a core crime, shall apply ‘the Constitution and the
law’. [206] The South African Bill of Rights in s 35 sets out a range of rights for arrested,
detained and accused persons. These protections will obviously need to be afforded to
any person who is being tried under the ICC Act. In addition, the Rome Statute sets out
a comprehensive framework of general principles of liability and defences in Part 3 of
the Statute. While the drafters of the ICC Act have not chosen to expressly adopt Part
3, s 2 of the Act says that applicable law for any South African court hearing any matter
arising under the Act includes ‘conventional international law, and in particular the
[Rome] Statute’. [207] Accordingly, the general principles of international criminal law
applicable to the prosecution of genocide, war crimes and crimes
Page 284
against humanity (including the available defences contained in the Rome Statute, such
as superior orders) ought to find application before a South African court.
3 Complementarity

The ICC Act gives effect to the complementarity scheme by creating the structure
necessary for national prosecutions under the ICC Statute. The procedure for the
institution of prosecutions in South African courts is set out in s 5 of the Act. This
procedure involves different governmental departments and officials. First, the ICC Act
requires that the consent of the National Director of Public Prosecutions must be
obtained before any prosecution may be instituted against a person accused of having
committed a crime. [208] The National Director must, when reaching a decision about a
prosecution, recognise South Africa’s obligation, in the first instance, under the principle
of complementarity in the Rome Statute, to exercise jurisdiction over and to prosecute
persons accused of having committed an ICC crime. [209] Given the importance of any
such prosecution, it is clear that a specialised court would need to be designated. The
Act provides that, after the National Director has consented to a prosecution, an
appropriate High Court must be designated for that purpose. Such designation must be
provided in writing by the ‘Cabinet member responsible for the administration of justice
. . . in consultation with the Chief Justice of South Africa and after consultation with the
National Director’. [210] The ICC Act does not provide any specific trial procedure or
punishment regime for domestic courts. All that the ICC Act provides is for the
designation of ‘an appropriate High Court in which to conduct a prosecution against any
person accused of having committed [an ICC] crime’. [211] Presumably the usual trial
procedure for a criminal trial in the High Court will be followed and the High Court will
be empowered to issue any of the sentences which it would ordinarily be entitled to
impose in terms of its domestic criminal sentencing jurisdiction. Such punishments
would include life imprisonment, imprisonment, a fine and correctional supervision.
The expectation under the Act, flowing from South Africa’s obligations under the
complementarity scheme, is that a prosecution will take place within the Republic.
Accordingly, if the National Director declines to prosecute a person under the Act, the
Director-General
Page 285
of Justice must be provided with the full reasons for that decision. [212] The Director-
General is then obliged to forward the decision, together with reasons, to the Registrar
of the International Criminal Court in The Hague. [213]

4 Co-operation with the International Criminal Court

(a) Arrest and surrender

The ICC Act is premised on the understanding that the International Criminal Court will
in most circumstances have to rely on the intercession of national jurisdictions to gain
custody of suspects. As a result, the ICC Act envisages two types of arrest: one in
terms of an existing warrant issued by the ICC, and another in terms of a warrant
issued by South Africa’s National Director of Public Prosecutions. In both scenarios the
warrant (whether endorsed or issued) must be in the form and executed in a manner as
near as possible to that which exists in respect of warrants of arrest under existing
South African law. [214]
Dealing with the first scenario (an arrest in terms of an existing warrant issued by
the ICC), in terms of s 8 of the ICC Act, when South Africa receives a request from the
ICC for the arrest and surrender of a person for whom the ICC has issued a warrant of
arrest, it must refer the request to the Director-General of Justice with the necessary
documentation to satisfy a local court that there are sufficient grounds for the surrender
of the person to The Hague. [215] The Director-General must then forward the request
(along with the necessary documentation) to a magistrate, who must endorse the ICC’s
warrant of arrest for execution in any part of the Republic. [216]
Section 9 details the second scenario (an arrest in terms of a warrant issued by the
National Director). In this situation, the Director-General of Justice is mandated to
receive a request from the ICC for the provisional arrest of a person who is suspected
or accused of having committed a core crime, or has been convicted by the ICC. The
Director-General is then obliged under the ICC Act to immediately forward the request
to the National Director of Public Prosecutions, who must then apply for the warrant
before a magistrate. [217]
Page 286
After being arrested pursuant to a warrant (whether that warrant was issued by the
ICC or by the National Director) the arrestee is to be brought ‘before a magistrate in
whose area of jurisdiction he or she has been arrested or detained’, ‘within 48 hours
after that person’s arrest or on the date specified in the warrant for his or her further
detention’. [218]
Having laid their hands on the arrestee, the South African authorities then become
engaged in what is known as the ‘surrender’ of an arrestee to the International Criminal
Court—his or her ‘delivery’ to The Hague. To make a committal order, with a view to
the surrender of an arrestee to the International Criminal Court, the magistrate has to
be satisfied of three things only. First, the magistrate must be satisfied that the person
before court is the individual named in the warrant. [219] Secondly, the person must
have been arrested in accordance with the procedures set down by domestic
law. [220] Thirdly, the arrestee’s rights, as contemplated in the Bill of Rights, must have
been respected, if, and to the extent to which, they are or may be applicable. [221] The
nature of these three requirements makes it clear that surrender to the ICC is different
to extradition in international law. There is no mention of the double criminality rule
which has become so central to extradition proceedings; and unlike many extradition
proceedings, there is no requirement in the ICC Act that a prima facie case be shown
against the suspect. [222] Section 10(5) of the ICC Act provides as the primary test that,
if, after considering the evidence adduced at the inquiry, the magistrate is satisfied that
the three requirements outlined above are met, then the magistrate ‘must issue an
order committing that person to prison pending his or her surrender to the Court’. Of
course, the magistrate also has to be satisfied that the International Criminal Court has
a genuine interest in the surrender of the arrestee, and to this end, s 10(5) stipulates
that, in addition to the three requirements being met, the magistrate must be content
that the person concerned may be surrendered to the Court: (a) for prosecution for the
alleged crime; (b) for the imposition of a sentence by the Court for the crime in respect
of which the person has been convicted, or (c) to serve a sentence already imposed by
the Court. [223] There is little indication in the Act what level of proof must be proffered
by the prosecution in respect of these additional requirements, such as whether the
court must inquire if there is
Page 287
evidence to justify the person’s trial for the offence he or she is alleged to have
committed. [224] Presumably any of these three factual conditions will have been proved
by the terms of the International Criminal Court’s request, either for the endorsement
of its own warrant of arrest within South Africa (in terms of s 8 of the ICC Act) or for
South Africa to issue a provisional warrant of arrest pursuant to the Court’s request (in
terms of s 9 of the ICC Act), such that these additional requirements may be regarded
as being satisfied on the strength of the ‘material supporting the request’ for surrender
provided by the International Criminal Court. [225]

(b) Forms of assistance offered to the Court in fulfilment of article 93 of the Rome
Statute
Article 93 of the Rome Statute requires states parties to assist the ICC by co-operating
in relation to investigations and prosecutions. Part 2 of the ICC Act sets out a variety of
circumstances in which the relevant competent authorities in the Republic must
‘cooperate with, and render assistance to, the Court in relation to investigations and
prosecutions’. There are many areas of co-operation (detailed in s 14 of the Act) such
as the questioning of suspects, the identification and whereabouts of persons or items,
the taking of evidence (including expert opinions), inspections in loco (including the
exhumation and examination of grave sites) and the execution of searches and
seizures, to name but a few. [226] The areas of co-operation must be undertaken in
terms of the relevant law applicable to investigations in South Africa, as well as the
applicable rules in the Rome Statute, [227] and with the ultimate aim of assisting the
ICC.
Page 288
Certain acts of co-operation are subject to comprehensive regulation in the ICC Act
and others are not. For example, in the context of questioning suspects, the ICC Act
stipulates in s 14(c) no more than that the competent South African authorities must
assist with ‘the questioning of any person being investigated or prosecuted’. South
African authorities will therefore have to turn to the Rome Statute and South African
law for assistance. In this respect, the Bill of Rights in s 35 and the Rome Statute in
article 55 equally guarantee certain rights to a person under investigation, such as the
right against self-incrimination, the right to remain silent and the right to legal
assistance.
Those means of co-operation that are subject to detailed regulation under the ICC
Act include the examination of witnesses, [228] the transfer of a prisoner to the ICC for
the purposes of giving evidence or to assist in an investigation, [229] the service of
process and documents, [230] acts of entry, search and seizure, [231] and the making of
forfeiture or confiscation orders. [232]

(c) Specialised units

In order that South Africa’s obligations under the ICC Act may be fulfilled, a Priority
Crimes Litigation Unit (PCLU) has been established within the National Prosecuting
Authority and is headed by a Special Director of Public Prosecutions appointed in terms
of s 13(1)(c) of the National Prosecuting Authority Act 32 of 1998. Section
13(1)(c) provides that the President may, by proclamation in the Government Gazette,
appoint one or more Directors of Public Prosecutions (referred to as Special Directors)
to perform functions assigned to him by the President.
Page 289
The Special Director’s appointment was confirmed in terms of Government
Gazette 24876 of 23 May 2003. [233] The Special Director was given two powers: first, to
head the PCLU and, secondly, to ‘manage and direct the prosecution of crimes
contemplated in the Implementation of the Rome Statute of the International Criminal
Court Act’. The Unit is thus specifically tasked with dealing with the ICC crimes set out
in the ICC Act.
The PCLU has expressed the view that when it comes to the investigation of the
alleged perpetrator, it is dependent on the co-operation of the South African Police
Services. A Directorate for Priority Crimes Investigation (DPCI) has recently been
established within the police, and the crimes under the ICC Act fall within its purview
for investigation. In practice, this means that requests by individuals or civil society
groups for investigation and prosecution under the ICC Act should be directed jointly to
the PCLU and DPCI.
On the assumption that the PCLU, with the assistance of DPCI, takes up the
investigation and issues a warrant of arrest (in camera or otherwise) and the suspect or
suspects are arrested, the matter will then move to the prosecution stage. The ICC Act
stipulates that no prosecution may be instituted against a person accused of having
committed a core crime without the consent of the National Director of Public
Prosecutions. Assuming that such consent is provided, the matter will proceed to court
and the PCLU will adopt responsibility for the prosecution of the matter.

(d) Discretionary measures of assistance

In terms of the ICC Act, the President may, at the request of the ICC and by
proclamation in the Government Gazette, declare any place in the Republic to be the
seat of the ICC. [234] Should such a declaration be made, then the ICC Act sets out a
variety of privileges and immunities for the Court. First, the Court is accorded such
rights and privileges of a South African court of law in the Republic as may be
necessary to enable it to perform its functions. [235] Furthermore, the judges, the
prosecutor, the deputy prosecutors and the registrar of the Court, while performing
their functions in the Republic, enjoy the same immunities and privileges that are
accorded to a representative of another state or government in terms of s 4(2) of the
South African Diplomatic Immunities and Privileges Act 37 of 2001. [236] Those
immunities include immunity from the criminal and civil jurisdiction of the courts of the
Republic, and the privileges enjoyed are those
Page 290
which (a) a special envoy or representative enjoys in accordance with the rules of
customary international law, or (b) are provided for in any agreement entered into with
a state, government or organisation whereby immunities and privileges are conferred
upon such special envoy or representative.
The deputy registrar, the staff of the Office of the Prosecutor and the staff of the
registry of the Court enjoy the privileges and facilities necessary for the performance of
their functions in the Republic as may be published by proclamation in the Government
Gazette, as provided for in s 7(2) of the Diplomatic Immunities and Privileges Act. [237]
The Minister of Foreign Affairs may, after consultation with the Minister of Justice,
confer immunities and privileges on any other member of the staff of the Court or any
person performing functions for purposes of the ICC Act. Such immunities and
privileges are conferred by the Minister of Foreign Affairs publishing a notice in
the Government Gazette, on such conditions as he or she deems necessary. [238] Any
person who is accorded immunities or privileges in terms of the ICC Act must have his
or her name entered into a register as contemplated in s 9(1) of the Diplomatic
Immunities and Privileges Act. [239]
(e) Enforcement of sentences
The Rome Statute stresses that ‘States Parties should share the responsibility for
enforcing sentences of imprisonment, in accordance with principles of equitable
distribution’. [240] The International Criminal Court has no prison, and states are
therefore expected to volunteer their services, indicating their willingness to allow
convicted prisoners to serve the sentence within their domestic penal institutions. [241]
After sentencing an offender, the ICC will, in terms of article 103(1)(a) of the Rome
Statute, designate the state where the term is to be served. In so doing, the Court
must take into account the views of the sentenced prisoner, his or her nationality, and
‘widely accepted international treaty standards governing the treatment of
prisoners’. [242] In addition, conditions of detention must be neither more nor
Page 291
less favourable than those available to prisoners convicted of similar offences in the
state where the sentence is to be enforced. [243]
In order to give effect to this enforcement scheme, the ICC Act provides that the
Minister of Correctional Services must consult with the Cabinet and seek the approval of
Parliament with the aim of informing the ICC whether South Africa can be placed on the
list of states willing to accept sentenced persons. [244] If the Republic is placed on the
list of states and is designated as a state in which an offender is to serve a prison
sentence, then such person must be committed to prison in South Africa. [245] The
provisions of the Correctional Services Act [246] and South African domestic law then
apply to that individual. However, the sentence of imprisonment may only be modified
at the request of the ICC, after an appeal by the prisoner to, or review by, the Court in
terms of the Rome Statute. [247]
It is commendable that the ICC Act requires the government to indicate its
availability to assist in enforcing the ICC sentences. It is not clear, however, that South
Africa will be placed on the list of states available for enforcement duty. The Rome
Statute stresses that there can be ‘no question of sending a prisoner to a State with
prison conditions that do not meet international standards’. [248] This is a particular
problem for South Africa, given the poor state of its prisons. [249]
The Rome Statute also enables the ICC to impose a fine, but only ‘[i]n addition to
imprisonment’. [250] On top of this, the ICC is empowered to address the issue of
reparations to victims, and may ‘make an order directly against any convicted person’
specifying reparation. [251] Such an order will no doubt often take the form of monetary
compensation. The ICC Act makes provision for the execution of such fines and
compensation orders within the Republic. [252] Such orders must be registered with a
court in the Republic
Page 292
which has jurisdiction. [253] Once the order has been registered, that sentence or order
‘has the effect of a civil judgment of the court at which it has been registered’ and the
Director-General of Justice must pay over to the ICC any amount realised in the
execution of the sentence or the order, minus any expenses incurred by the Republic in
the execution thereof. [254]

Immunities
While the ICC Act provides South African courts with potential jurisdiction over persons
who may have committed ICC crimes, the issue of immunity from jurisdiction for high-
ranking officials remains contentious. The most heated debate has been around the
extent to which serving heads of state and other senior government officials can
justifiably claim immunity, on the basis of their official status, from proceedings brought
against them for allegedly committing international crimes.
This question dramatically rears its head in the al-Bashir case before the ICC. The
ICC has issued two warrants of arrest for al-Bashir because of his alleged crimes in
Darfur. These were forwarded to South Africa with requests to assist in arresting
him. [255] South Africa’s obligations under these warrants were also made clear in
various communications and consultations with the ICC leading up to al-Bashir’s
visit. [256] When al-Bashir was present in South Africa in 2015, South Africa did not
arrest and surrender him to the ICC. The government’s main reason for not doing so
was that al-Bashir, as an incumbent head of state, is immune from being arrested and
surrendered to—then tried before—the ICC.
The High Court in Pretoria [257] and the Supreme Court of Appeal [258] found that
South Africa had violated its obligations under international and domestic law by not
arresting and surrendering al-Bashir. The ICC, in its own judgment, reached the same
conclusion. [259] This is
Page 293
not the first or last time the ICC has held that a state breached its obligations under the
Rome Statute by relying on immunity to refuse a request to arrest al-Bashir. [260]
Immunity from jurisdiction under international law is discussed elsewhere in this
book. [261] The issue of immunity under the ICC Act and the Rome Statute can be
divided into four parts that warrant discussion here. First, does South African law, on a
domestic level, grant immunity to incumbent heads of state (or other senior state
officials) from arrest and surrender to the ICC? Secondly, does South African law grant
immunity to incumbent heads of state from prosecution before a South African court?
Thirdly, does international law grant immunity to incumbent heads of state from
prosecution before the ICC? Fourthly, what are South Africa’s international legal
obligations regarding incumbent senior officials who are implicated in international
crimes, like al-Bashir?

1 Immunity from arrest and surrender to the ICC

South African law does not grant personal or functional immunity to those who are
sought for surrender to the ICC. [262] This is for four reasons. First, s 10(9) of the ICC
Act provides that the arrested person being a head of state ‘does not constitute a
ground for refusing to issue an order’ for committal and surrender by a
magistrate. [263] Tladi explains as follows: [264]
As du Plessis points out, this provision [s 10(9)] is unambiguous in its effect, ie the mere fact
that a person is entitled to inviolability is in itself not a justification for not ordering surrender.
This means that even if a South African court itself cannot exercise jurisdiction over a head of
state like al-Bashir, this does not apply to the arrest and surrender processes described above.
It is noteworthy that while Article 98 of the Rome Statute provides an exception to the duty to
cooperate on the basis of immunity as described above, a similar provision does not exist in
the Implementation Act. Indeed section 10(9) of the Implementation Act, stating that the
status of a person is
Page 294
 not a ground for refusing surrender, suggests that the legislator intended to explicitly exclude
the effects of Article 98.
Secondly, s 10(1), which lists the factors a magistrate must consider when deciding
whether to issue an order of surrender to the ICC, does not include the question
whether the arrested person enjoys any immunity. [265]
Thirdly, s 8 provides for the procedure for receiving and executing a warrant of
arrest from the ICC. It is silent about immunity. It must be read consistently with s
10(9). To avoid s 10(9) being rendered nugatory, one must read s 8 as not allowing
immunity to prevent arrests of persons who stand to be surrendered to the ICC under s
10. An interpretation which renders s 10(9) nugatory is to be avoided. [266]
Finally, not providing immunity against arrest and surrender accords with South
Africa’s international law obligations. [267] It thus promotes the object of the Act: lawful,
effective co-operation between the ICC and South Africa. It also ensures that legislation
is reasonably interpreted to be consistent with international law. [268]
Therefore, South African law does not grant immunity from arrest and surrender to
the ICC to incumbent heads of state. This election by the South African Parliament to
preclude claims to immunity is a progressive step, which is arguably ahead of the
international law curve. It accords with the ‘prudent approach’ advocated by Cassese et
al in their leading text on the Rome Statute of the International Criminal Court:
To avoid these difficulties [regarding immunities for officials], a prudent approach [for the
state’s legislature in drafting its ICC implementation legislation] would be to provide that any
issue of immunities will not bar arrest or surrender to the ICC. In essence, this approach
leaves the issue to be decided by the ICC and not by national courts. In this manner, an
implementing State can ensure that it will not find itself stuck with a legislative provision—or a
judicial interpretation—on international immunities that hinders compliance with an ICC
request. [269]
Page 295

2 Immunity from domestic prosecution

In the groundbreaking Pinochet cases, the House of Lords accepted that serving
international functionaries (such as current heads of state) retain absolute
immunities rationae personae (ie personal immunity on account of their status) before
domestic tribunals, irrespective of the nature of the crime alleged, unless waived by the
sending state. The House of Lords denied immunity to Pinochet in his capacity as
a former head of state. However, it made it clear that if he had still been an acting head
of state, this immunity in international law would have continued to subsist. [270]
The International Court of Justice has affirmed this immunity in its decision in
the Arrest Warrant case. [271] With regard to the provisions precluding immunity found
in the constitutive instruments of a myriad of international criminal tribunals (the most
recent being the Rome Statute of the ICC) the Court expressly held that this exception
to customary international law was not applicable to national courts. [272] This case law
is therefore read as authority for the proposition that there is a dichotomy between
immunity for international crimes before international courts and national courts: the
personal immunity of an accused person prevents national courts from dealing with
allegations of international crimes unless that immunity has been waived, or the senior
official has left office.
But what is the position under South African law? In Minister of Justice v SALC, the
Supreme Court of Appeal said that the ICC Act diverges from the position under
customary international law. Wallis JA noted how s 4(2)(a) of the Act provides that the
fact that the accused is or was a head of state does not constitute a defence for a crime
before a South African court. Because of this, he held, it cannot
Page 296
then be that personal immunity is a procedural bar to domestic prosecution in South
Africa. Otherwise s 4(2) is rendered nugatory. The ICC Act would include a defence that
no person could rely on, unless South African courts have jurisdiction over heads of
state in the first place. Therefore, s 4(2) does not give personal immunity to incumbent
heads of state from domestic prosecution in South Africa. [273]
Some commentators have echoed this. Dugard and Abraham argue that s 4(2)(a) of
the Act represents a choice by the legislature not to follow the ‘unfortunate Arrest
Warrant decision of which it must have been aware’. [274] In effect, the ICC Act simply
trumps personal immunity from domestic prosecution. [275]
But there are arguments the other way. For example, some contend that the
ordinary meaning of the section does not lift immunity from domestic prosecution, since
the provision refers to a ‘defence’, which is substantive criminal law, while immunity is
a procedural matter. [276] Ultimately, though, the Supreme Court of Appeal has put the
issue to rest. In Minister of Justice v SALC, the Court found as follows at para 103:
I conclude therefore that when South Africa decided to implement its obligations under the
Rome Statute by passing the Implementation Act it did so on the basis that all forms of
immunity, including head of state immunity, would not constitute a bar to the prosecution of
international crimes in this country.
As a safety net, any decision by the South African authorities not to prosecute entitles
the ICC to do so in South Africa’s place. Section 5(6) of the ICC Act confirms this in
stating that a decision by the National Director of Public Prosecutions ‘not to prosecute
a person under this section does not preclude the prosecution of that person in the
[International Criminal] Court’. In addition, article 98(1) of the Rome Statute entails
that states parties have a duty of co-operation with the Court, requiring—as the SCA
has now confirmed in Minister of Justice v SALC—South Africa to arrest and surrender
to the Court persons charged with an ICC crime where those persons are on South
African territory and the subject of an arrest warrant from the ICC.
Page 297

3 Immunity from international prosecution before the ICC

Immunity, including personal immunity from prosecution for international crimes, is

waived if states sign up to a tribunal whose charter provides that immunity is not a bar
to prosecution. [277] Article 27(2) of the Rome Statute provides that neither functional
nor personal immunity can prevent prosecution before the ICC. So, if a state signs up
to the Rome Statute, they waive whatever immunity their officials or heads of state
would otherwise enjoy as a bar to their prosecution at the ICC.
The issue is that some countries, including Sudan, have not signed the Statute. In
multiple decisions, though its reasoning was not always entirely consistent, the ICC has
held that al-Bashir’s immunity from prosecution before the ICC has been waived by
Sudan by means of Security Council Resolution 1593 (‘the Resolution’) [278] which
referred the situation in Sudan to the ICC. Sudan is bound by the Security Council’s
decision that Sudan is to co-operate fully with the ICC. [279] Moreover, full co-operation
necessitates the waiving of al-Bashir’s immunity—otherwise the ICC is hampered in
prosecuting al-Bashir. Therefore, al-Bashir’s immunity has been waived. [280]
Some object that if the Resolution intended to remove al-Bashir’s immunity, it would
have done so explicitly. [281] However, the Security Council cannot be expected to
predict all permutations of ‘co-operation’. To do so would have undermined the ICC’s
judicial independence. The Security Council has subsequently not objected to the
issuing of al-Bashir’s arrest warrants. [282] And ‘cooperate fully’ was the term that was
used in the resolutions creating the ICTY and ICTR. These resolutions were considered
to remove immunities. [283]
Akande has argued, and the ICC has held, that the Resolution entails an actual
waiver by Sudan and that it binds Sudan to accept the ICC’s jurisdiction in terms of the
Statute. The ICC’s jurisdiction, as per the Statute, includes article 27(2), which
becomes binding on
Page 298
Sudan. This amounts to waiving immunity from prosecution before the ICC. [284]

Gaeta responds to Akande by arguing that the Resolution simply triggers the ICC’s
jurisdiction. It cannot go further and bind Sudan (like a member state) to article
27(2). [285]
Despite this debate, the ICC’s reasoning in DRC left open the question whether
Sudan is, via the SC Resolution, bound to article 27. Akande’s argument could explain
why the ICC believed that Sudan had actually waived al-Bashir’s
immunity. [286] However, in Jordan and South Africa, the ICC made it clear that Sudan is
bound to article 27 via the Resolution. Therefore, as things currently stand, al-Bashir
does not enjoy immunity in international law from prosecution before the ICC. The
Appeals Chamber, in the Jordan matter, [287] is currently seized with giving a definitive
ruling from the ICC on the issue—and the seriousness of the question moved the Court
to make the unprecedented ‘Order inviting expressions of interest as amici curiae in
judicial proceedings (pursuant to rule 103 of the Rules of Procedure and Evidence)’. [288]

4 South Africa’s obligations regarding senior officials

Article 98(1) of the Rome Statute provides that member states cannot, when co-
operating with the ICC, disregard their obligations to respect the immunities enjoyed by
heads of non-member states. [289] So, ordinarily, member states cannot arrest or
surrender al-Bashir (or other incumbent heads of state) because of his personal
immunity. This is unless the Court can obtain the co-operation of that third state for the
waiver of the immunity.
As already indicated, in DRC, the ICC found that the Resolution removes al-Bashir’s
immunity as a bar to member states co-operating with the ICC by arresting and
surrendering al-Bashir. The Resolution, which required Sudan to ‘cooperate fully’ and
‘provide any necessary assistance to the Court’, constitutes the ‘cooperation of that
third State [Sudan] for the waiver of the immunity’, as required under
Page 299
article 98(1) of the Statute. [290] Therefore, because of the Resolution, South Africa
would not have breached its obligation to respect al-Bashir’s immunity in international
law had South Africa arrested him.
 The purpose of article 98(1) is to ensure that immunity is consensually waived by a
non-member state. [291] According to the ICC, then, Sudan has given this consent by
agreeing to abide by the Resolution. But the ICC’s focus on ‘co-operation’, while it
makes al-Bashir’s position clear, leaves questions open about the relationship between
articles 27 and 98.
Chapter 12 examines the correct position regarding this relationship between articles
27 and 98. [292] An answer to this question from the ICC is expected in
the Jordan appeal, which at the time of writing is still to be decided.
As things presently stand, the ICC has resolved this uncertainty in South
Africa and Jordan by finding that because Sudan is bound (via the Resolution) to article
27, article 98(1) is inapplicable. There is no immunity to be waived per article 98(1)
because it has already been waived through article 27. [293] Therefore, the obligation
under article 98(1) does not bar South Africa (or any state party) from arresting al-
Bashir.
So, given that South Africa was under an obligation to arrest al-Bashir pursuant to
the request from the ICC, South Africa breached its obligation by failing to do so.

South Africa and the ICC

South Africa was instrumental in the adoption and drafting of the Rome Statute.
Nothing less would be expected from a country that has experienced its own crimes
against humanity. But the growing resentment within the AU of the ICC and the recent
al-Bashir saga have challenged South Africa’s commitment to the Rome Statute. At the
time of writing, South Africa is still a party to the Rome Statute and the ICC Act is still
in force. But South Africa’s withdrawal from the ICC, and a repeal of the ICC Act, are
mooted by the government.
On 19 October 2016, about a year after al-Bashir left South Africa, the Zuma
government took the decision to withdraw from the Rome Statute and delivered its
withdrawal instruments to the UN Secretary-General.
Page 300
An opposition party and various civil society organisations urgently applied to the High
Court to review the decision to withdraw from the Statute. A full bench of the High
Court granted the order sought and declared the decision invalid. [294] The primary
grounds for doing so was that Parliament needed to approve the withdrawal, and that
withdrawal could not occur before the ICC Act is repealed. [295]
The government was then forced to reverse itself: on 7 March 2017 it deposited with
the UN Secretary-General in New York a document headed ‘South Africa: Withdrawal of
notification of withdrawal’.
Thereafter, the Minister of Justice, Michael Masutha, announced on 6 December 2017
at the Assembly of States Parties meeting of the ICC in New York that South Africa was
attempting for a second time to withdraw from the Court.
He indicated there that he would submit to Parliament South Africa’s intention to
withdraw from the ICC. To this end, the executive has tabled the International Crimes
Bill. [296] Among others, the purpose of the Bill is to repeal the ICC Act. The Bill takes
the unfortunate view that the Rome Statute and the ICC Act are inconsistent with
international law on immunity. In the first place, the Bill gives effect to South Africa’s
view that it must withdraw from the ICC. But the arguments in support of its
withdrawal have been subjected to withering criticism by many, including a number of
retired Justices of South Africa’s Constitutional Court. [297] Furthermore, as explained
above, while the Supreme Court of Appeal in Minister of Justice v SALC expressly
rejected the notion that customary international law today recognises an all-
encompassing exception to immunities in the case of international crimes, it held that
South Africa’s ICC Act meant that President al-Bashir was not entitled to immunity
while he was in South Africa. As the Court stressed: in adopting this approach ‘South
Africa was taking a step that many other nations have not yet taken. If that puts this
country in the
Page 301
vanguard of attempts to prevent international crimes . . . that seems to me to be a
matter for national pride rather than concern. It is wholly consistent with our
commitment to human rights, both at a national and at an international
level.’ [298] Moreover, if South Africa successfully withdraws from the Rome Statute, this
will constitute a wasted opportunity to show leadership. The ICC is by no means a
flawless institution above reproach. On the contrary, it is in need of constructive
engagement from its member states, and South Africa could be playing a leading role in
helping improve the ICC and its relationship with many African states who have
increasingly become dissatisfied with the Court’s case selection and efficiency. [299]
Finally, the ICC has found that al-Bashir lacked immunity because Sudan, via the
Security Council resolution, had waived his immunity. For that reason, the Bill changes
very little for al-Bashir when it says that the Bill does not apply to anyone who enjoys
immunity under customary international law. [300] In any event, repealing the Act and
withdrawing from the Rome Statute does not revoke obligations incurred by South
Africa while party to the Statute. [301] This means that South Africa will continue to carry
its ICC obligations to arrest al-Bashir if he were to arrive in South Africa. [302]
The Bill is in many ways similar to the ICC Act. It domesticates international
crimes [303] and sets up the domestic processes for the surrender of a person to an
‘entity’ that has jurisdiction over the international crimes of which the person is
accused. [304] An entity is defined as ‘the International Criminal Court, any international
organisation, international tribunal, international court, or similar international body
which has jurisdiction in respect of an international crime’. [305] Section 12 of the Bill
envisages entities making requests for the surrender of sought persons. The obvious
difference
Page 302
is that once South Africa withdraws from the Statute, the ICC will not have jurisdiction
over future crimes or criminals in South Africa.
At the time of writing, the Bill is still at its early stages. It may well be amended as it
progresses through the parliamentary process—and it could face possible legal
challenge. There is also the hope that South Africa yet sees its position differently.
While there is no question that immunities for heads of state is a thorny issue, the
details of those debates, and their importance for the future of the ICC, are precisely
where South Africa’s strength lies. Instead of its strategy of rejection and withdrawal,
South Africa could help work with other African states to improve the ICC from within
and to help set the agenda of that institution—the world’s first and only permanent
international criminal court that is able to ensure accountability for the world’s worst
crimes. [306]
[1] For an account of this history, see B Ferencz An International Criminal Court: A Step towards World
Peace—A Documentary History and Analysis (1980).
[2] Article 227 of the Treaty of Versailles (UK Treaty Series No 1 (1919)) provided for the trial of the
Emperor for ‘a supreme offence against international morality and the sanctity of treaties’ before a special
tribunal composed of five judges appointed by the United Kingdom, the United States, France, Italy and
Japan. The attempt to bring the Emperor to trial was thwarted when he was granted asylum by the
Netherlands.
[3] The ‘unratified’ Treaty of Sevres of 1920 (UK Treaty Series No 11 (1920)) provided for the
surrender by Turkey of persons ‘responsible for the massacres committed during the continuance of the
state of war on territory which formed part of the Turkish Empire’ (article 230) but in 1923, the Treaty of
Lausanne (UK Treaty Series No 16 (1923), Part VIII) granted amnesty to these persons. See VN Adrian
‘Genocide as a problem of national and international law: The World War I Armenian Case and its
contemporary legal ramifications’ (1989) 14 Yale Journal of International Law 221.
[4] See Chapter 9.
[5] Convention for the Creation of an International Criminal Court, in MO Hudson International
Legislation vol 7 no 500 (1941).
[6] See T Taylor The Anatomy of the Nuremberg Trials (1992). The judgment of the Nuremberg
Tribunal is published at (1947) 41 AJIL 172.
[7] AC Brackman The Other Nuremberg: The Untold Story of the Tokyo War Crimes Trials (1989).
[8] For an insightful overview of the criticisms of the Nuremberg trials, see R Overy ‘The Nuremberg
trials: International law in the making’ in P Sands (ed) From Nuremberg to The Hague—The Future of
International Criminal Justice (2003) 1.
[9] See WA Schabas An Introduction to the International Criminal Court (2004) 8.
[10] In 1979 the United Nations Human Rights Commission instructed Professor M Cherif Bassiouni to
draft a statute for an international court to try offenders under the 1973 International Convention on the
Suppression and Punishment of the Crime of Apartheid. A statute was drafted but no action was taken on
the project: see M Cherif Bassiouni A Draft International Criminal Code and Draft Statute for an
International Criminal Tribunal (1987) 10–11. See further on the crime of apartheid, Chapter 9.
[11] See K Kittichaisaree International Criminal Law (2001) 27.
[12] See J Crawford ‘The ILC’s draft statute for an International Criminal Tribunal’ (1994) 88 AJIL 140
and ‘The ILC adopts a statute for an International Criminal Court’ (1995) 89 AJIL 404.
[13] SC Resolution 808 of 22 February 1993 and SC Resolution 827 of 25 May 1993. For detailed
accounts of the creation of the ICTY, see M Cherif Bassiouni and P Manikas The Law of the International
Criminal Tribunal for the Former Yugoslavia (1996) Chapters I–III. See, too, V Morris and M Scharf An
Insider’s Guide to the International Criminal Tribunal for the Former Yugoslavia: A Documentary History
and Analysis (1995).
[14] SC Resolution 955 of 8 November 1994. For details, see C Scheltema and W van der Wolf
(eds) The International Tribunal for Rwanda: Facts, Cases, Documents (1999).
[15] The ICTR closed on 31 December 2015, and the ICTY closed on 31 December 2017. The essential
functions of both tribunals have been preserved through a small and temporary structure (the so-called
‘Residual Mechanism’) which has the power, inter alia, to prosecute the most senior persons indicted by
the Tribunal after the completion of their mandates, and persons who interfere with the administration of
justice. See Security Council Resolution 1966 (2010) available at http://www.icty.org/en/press/security-
council-adopts-resolution-international-residual-mechanism-criminal-tribunals-irmct.
[16] At Nuremberg, ‘crimes against humanity’ were prosecuted as crimes associated with one of the
other crimes within the Nuremberg Tribunal’s jurisdiction, namely, war crimes and crimes against peace.
Since Nuremberg, several variants of crimes against humanity have developed, not all with a nexus to
armed conflict. (The most prominent example is genocide—the most egregious form of crime against
humanity—which the Genocide Convention of 1948 defines as an offence which can be committed in
times of peace and war.) The requirement of a nexus with armed conflict was firmly done away with by
the Yugoslavia Tribunal in its celebrated decision in Prosecutor v Tadic Case No IT-94-1-AR72 see n 81
below (2 October 1995), (1996) 35 ILM 32. Article 7 of the Rome Statute codifies this evolution of crimes
against humanity as being crimes committed either in times of peace or war.
[17] See the Tadic case (n 16). Interesting developments have also come out of the Rwanda Tribunal’s
decisions. For instance, in Prosecutor v Akayesu (Judgment) ICTR-96-4-Y T Ch 1 (2 September 1998) the
Rwanda Tribunal came to the enlightened conclusion that rape could constitute an act of genocide.
[18] Schabas (n 9) 12.
[19] For the latest ratification status, see www.iccnow.org.
[20] For status of African ratification, see https://asp.icc-
cpi.int/en_menus/asp/states%20parties/african%20states/Pages/african%20states.aspx. For recent
African discontent with the ICC, see discussion further below.
[21] There is vast literature critiquing the failure by the United States to join the Court. For selected
reading, see MP Scharf ‘The United States and the International Criminal Court: The ICC’s jurisdiction
over nationals of non-party states—A critique of the US position’ (2001) Law and Contemporary
Problems 64; M du Plessis ‘Seeking an International International Criminal Court—Some reflections on the
United States opposition to the ICC’ (2002) 15 SACJ 301; WA Schabas ‘United States hostility to the
International Criminal Court: It’s all about the Security Council’ (2004) 15 EJIL 710.
[22] On 2 November 2002, the Special Court for Sierra Leone was established pursuant to Security
Council Resolution 1315. This tribunal is the result of an agreement between the UN and Sierra Leone to
try ‘those who bear the greatest responsibility’ for crimes against humanity and disrupting the peace
process. The Court is a hybrid, staffed by local and international personnel, and has an international
prosecutor. Its temporal jurisdiction to prosecute international crimes under its Statute stretches back to
crimes committed since 30 November 1996. The case that has attracted the most attention is the trial of
former Liberian President Charles Taylor. Taylor was indicted on 7 March 2003 for crimes against
humanity, war crimes, and other serious violations of international humanitarian law. These crimes
include terrorising the civilian population and collective punishments, unlawful killings, sexual violence,
physical violence, use of child soldiers, abductions, forced labour and looting. Due to concerns about
regional security if the trial was held in Sierra Leone, the Special Court arranged for the trial to be held at
The Hague in the Netherlands. On 26 April 2012 the Trial Chamber found Taylor guilty on all 11 counts,
on the modes of liability of planning of crimes and for aiding and abetting of crimes committed by rebel
forces in Sierra Leone. On 30 May 2012 the former Liberian president was given a single sentence of 50
years in prison. The conviction and sentence were upheld by the Appeals Chamber. See Prosecutor v
Charles Ghankay Taylor SCSL-03-01-A (Appeal Chamber Decision) available at
http://www.rscsl.org/Taylor.html.
[23] There is a wealth of literature on the ICC. See, for example, A Cassese et al (eds) The Rome
Statute of the International Criminal Court: A Commentary 3 vols (2002); O Triffterer (ed) Commentary
on the Rome Statute of the International Criminal Court 3 ed (2016); WA Schabas The International
Criminal Court: A Commentary on the Rome Statute 2 ed (2016). A number of other books deal with the
ICC in the context of general international criminal law: see, for example, A Cassese (ed) The Oxford
Companion to International Criminal Justice (2010); C Stahn and L van den Herik (eds) Future
Perspectives on International Criminal Justice (2010); G Boas et al The International Criminal Law
Practitioner 3 vols (2008–2011); R Cryer et al An Introduction to International Criminal Law and
Procedure 3 ed (2014).
[24] Judges are elected for terms of office of nine years by the Assembly of States Parties to the Rome
Statute, and are not eligible for re-election.
[25] Articles 34 and 39 of the Rome Statute.
[26] Article 42(1). A summary of the submissions received by the Office of the Prosecutor is publicly
available. See the ICC’s website.
[27] Article 42(2).
[28] Article 43.
[29] Article 112.
[30] Articles 5–8.
[31] Article 5(2). See further, WA Schabas ‘The unfinished work of defining aggression: How many
times must the cannonballs fly before they are forever banned?’ in D McGoldrick, P Rowe and E
Donnelly The Permanent International Criminal Court—Legal and Policy Issues (2004) 123.
[32] Article 123(1). Consequently, the ICC may only prosecute terrorist acts if the particular acts meet
the definitional requirements of the current crimes, the most likely example being crimes against
humanity.
[33] See the Finalized Draft Text of the Elements of Crimes (PCNICC/2000/INF/3/Add.2).
[34] The crime of genocide has been committed throughout history, the pre-eminent example being the
mass killing of Jews by the Nazis during World War II, and more recently, the slaughter of Tutsis by
Hutus in Rwanda. The term ‘genocide’ is a combination of the Latin words genus (kind, type, race)
and cide (to kill) and was first coined by Raphael Lemkin writing in response to the events of the Second
World War. See R Lemkin Axis Rule in Occupied Europe (1944) at 79–95; R Lemkin ‘Genocide as a crime
under international law’ (1947) 41 AJIL 145. For an historical account of the origins of genocide, see P
Sands East West Street: On the Origins of Genocide and Crimes against Humanity (2016).
[35] There was no reference to the crime of genocide in the Charter or the judgment of the tribunal,
even though it did appear in the indictment and was referred to by the prosecution from time to time.
[36] See Kittichaisaree (n 11) 67. There was at least one national prosecution of genocide prior to the
ICTY and ICTR’s existence; namely, the prosecution of Eichmann before the District Court of Jerusalem
(1968). Eichmann was tried for crimes against the Jewish people, an offence under Israeli law which
incorporated all the elements of the definition of genocide (see A Cassese International Criminal
Law (2003) 97).
[37] Prosecutor v Akayesu (Judgment) ICTR-96-4-T TCh 1 (2 September 1998) paras 500–1.
[38] Akayesu paras 502–4.
[39] Akayesu paras 505–6.
[40] Akayesu para 507.
[41] Kittichaisaree (n 11) at 69.
[42] Akayesu (n 37) para 511.
[43] The same view has been expressed by the ICTY Trial Chamber in its ruling in the Krstic Case
(Judgment) IT-98-33-T (2 August 2001), (2001) 40 ILM 1346. There it confirmed that ‘customary
international law limits the definition of genocide to those acts seeking the physical or biological
destruction of all or part of a group’, with the result that an ‘enterprise attacking only the cultural or
sociological characteristics of a human group in order to annihilate these elements which give to that
group its own identity distinct from the rest of the community would not fall under the definition of
genocide’ (para 580). Genocide appears therefore to be limited to material destruction of a group, rather
than the destruction of the national, linguistic, religious, cultural or other identity of that group. It is for
this reason that the Australian courts have held that degradation of aboriginal people through confiscation
of traditional lands cannot amount to genocide by the responsible Ministers, since the confiscation was
not aimed at material destruction of the group as such (see G Robertson Crimes against Humanity (2000)
at 230).
[44] The Prosecutor v Jean Kambanda (Judgement and Sentence) ICTR 97-23-S, International Criminal
Tribunal for Rwanda (ICTR) 4 September 1998 para 16.
[45] Jelisic (Appeal) ICTY 5 July 2001, IT-95-10-A, (2001) 40 ILM 1295 para 66; H Vest ‘A structure-
based concept of genocidal intent’ (2007) 5 JICJ 781 at 783.
[46] The specific intention of destroying all or part of the group must have been formed by the accused
prior to the commission of the genocidal act. Put differently, the underlying genocidal act (killing, causing
serious bodily or mental harm etc) should be done to further the genocidal goal of ensuring the group’s
destruction (see Kayishema and Ruzindana (Judgment), ICTR-96-1-T, T Ch (21 May 1999) para 91).
[47] Jelisic (Appeal) (n 45) para 47. See also the decision of the ICTR Appeals Chamber
in Kayishema and Ruzindana Case ICTR Appeals Chamber, Case No ICTR 95-1-A (1 June 2001).
[48] Akayesu (n 37) para 732.
[49] See n 3.
[50] See Sands (n 34).
[51] This use of the idea of crimes against humanity—to initiate prosecutions against individuals for
atrocities committed within their own territories—led to a measure of discomfort for the Allied powers,
who were concerned about the ramifications for their treatment of minorities within their own countries
and colonies. As a result, the Nuremberg notion of ‘crime against humanity’ had an important rider
attached to it: a crime against humanity was committed if it was associated or linked with one of the
other crimes under the Tribunal’s jurisdiction, being war crimes and crimes against the peace
(aggression). What this meant is that there had to be a link between crimes against humanity
and international armed conflict. In part, that is why the Nuremberg trials are spoken of as ‘war crimes
trials’—since the crimes against humanity there could only be tried if they were attendant on either a
crime against peace or war crimes (see Schabas (n 9) 42). See also M Koskenniemi ‘Between impunity
and show trials’ (2002) 6 Max Planck UNYB 1.
[52] However, within weeks of the Nuremberg judgment the United Nations expressed its
dissatisfaction with this limited scope of crimes against humanity when the General Assembly asserted in
the Genocide Convention of 1948 that genocide (the most egregious form of crimes against humanity)
could be committed during times of war and peace. On that and other developments that gradually led to
the link between crimes against humanity and war being dropped, see A Cassese ‘Crimes against
humanity’ in A Cassese et al (eds) The International Criminal Court: A Commentary vol I (2002) 73.
[53] Cassese (n 52). The most recent developments relate to the ICTR and ICTY. The establishment of
the ICTR—to punish those guilty of crimes committed in an internal conflict—in itself reiterates the point
that crimes against humanity do not have to be attendant on an international armed conflict. See, too,
the ICTY Appeals Chamber decision in Prosecutor v Tadic (1997) 105 ILR 453 para 141.
[54] Cassese (n 52) 64.
[55] Article 7(1)(a).
[56] Article 7(1)(b).
[57] In Vasiljevic (ICTY Trial Chamber, 29 November 2002) the Tribunal held that for criminal
responsibility to attach for extermination, the accused must have been responsible for a ‘large number of
deaths’. See also D Mundis ‘Current developments at the ad hoc International Criminal Tribunals’ (2003)
1 JICJ 520 at 521.
[58] Article 7(2)(b).
[59] Article 7(1)(c).
[60] Article 7(1)(d).
[61] Article 7(2)(d)
[62] Article 7(1)(e).
[63] Article 7(1)(f).
[64] Article 7(2)(e).
[65] Article 7(1)(g).
[66] Article 7(1)(h).
[67] Article 7(2)(g).
[68] Article 7(1)(i).
[69] Article 7(2)(i).
[70] Article 7(1)(j).
[71] Article 7(2)(h).
[72] Article 7(1)(k).
[73] Ibid.
[74] Article 7(2)(a).
[75] This can be via a ‘deliberate failure to take action’. See fn 6 of the Elements of Crimes.
[76] See, for example, G Werle Principles of International Criminal Law 3 ed (2014) 299–300.
[77] Schabas (n 9) 45; Judgment pursuant to Article 74 of the Statute ICC-01/05-01/08-3343 (21
March 2016): The Prosecutor v Jean-Pierre Bemba Gombo paras 166–9; Prosecutor v Dragoljub
Kunarac ICTY Appeals Chamber, Case No IT-96-23 & 23/1 (12 August 2002) para 102.
Clearly, each of the underlying acts committed (in terms of the greater event—the attack) require their own
form of intent. However, overall, these acts must be committed with a specific intention that is associated with
the main event—the attack that gives the individual acts their ‘crime against humanity’ character.
[78] In articles 228–30 of the Treaty of Versailles, Germany recognised the jurisdiction of the Allied
powers to try persons accused of violating the laws and customs of war as well as the obligation to hand
over such accused persons to the Allies for that purpose. None of these provisions was implemented due
to later German pressure. Instead, Germany proposed to try its own nationals accused of war crimes
before the Supreme Court of Leipzig, a proposal which produced mock trials which resulted in only 13
convictions out of 901 cases, and with insignificant sentences which ultimately were not executed. See G
Abi-Saab ‘The concept of “war crimes”’ in S Yee and W Tieya (eds) International Law in the Post-Cold War
Area (2001) 99–118.
[79] The ‘law of The Hague’ is made up of the Hague Conventions of 1868, 1899 and 1907, which,
generally speaking, set out rules regarding the various categories of lawful combatants, and which
regulate the means and methods of warfare in respect of those combatants. The Hague Rules also deal
with the treatment of persons who do not take part in armed hostilities or who no longer take part in
them, but in this respect The Hague Rules have been supplanted by the Geneva Rules, which cover this
aspect of humanitarian law in more detail. See further Chapter 25 below on Humanitarian Law.
[80] The ‘law of Geneva’, so called because it comprises the four Geneva Conventions of 1949 plus the
two Additional Protocols thereto of 1977, regulates the treatment of persons who do not take part in the
armed hostilities (such as civilians, the wounded and the sick) and those who used to take part but no
longer do (such as prisoners of war). An exception here is the Third Geneva Convention, which, in
addition to the focus on treatment of persons no longer involved in the conflict, also regulates the various
classes of lawful combatants and thereby updates The Hague Rules. The Hague Rules have been further
updated by the First Additional Protocol to the Geneva Convention of 1977, which deals with the means
and methods of combat, with a particular emphasis on sparing civilians as far as is possible in an armed
conflict. See further Chapter 25 below on humanitarian law.
[81] Tadic (Appeal) (Decision on the defence motion for Interlocutory Appeal on Jurisdiction) ICTY
Appeals Chamber, case No IT-94-1-AR 72 (2 October 1995) cf n 16 paras 96–136. Until this decision, the
scope of international responsibility for war crimes was the subject of much confusion. The two major
sources of humanitarian law—war crimes codified in the Geneva Conventions and their Protocols, which
addressed the protection of the victims of armed conflict from ‘grave breaches’, and war crimes as
understood under The Hague Convention, which focused on the methods and materials of warfare—did
not appear to extend international criminal responsibility to those who committed the prohibited acts
during times of internal armed conflicts. In Tadic, the ICTY Appeals Chamber stated that international
criminal responsibility for war crimes included acts committed during internal armed conflict; that is,
during times of civil war.
[82] Article 8(1).
[83] Kittichaisaree (n 11) 41; Cryer et al (n 23) cited 2014 above 302.
[84] Schabas (n 9) 60; Cryer (n 23) 303–8.
[85] Article 8(2)(b)(i).
[86] Article 8(2)(b)(ii).
[87] Article 8(2)(b)(iii), which prohibits an attack that is intentionally launched in the knowledge that it
will cause incidental loss of life or injury to civilians, or damage to civilian objects, or widespread, long-
term and severe damage to the natural environment which would be clearly excessive in relation to the
concrete and direct overall military advantage anticipated.
[88] Article 8(2)(b)(v), such undefended places being defined as towns, villages, dwellings or buildings
that are undefended and that are not military objectives.
[89] Article 8(2)(b)(xxiii), which prohibits utilising the presence of a civilian or other protected person
to render certain points, areas or military forces immune from military operations.
[90] Article 8(2)(b)(xxv): intentionally starving civilians ‘as a method of warfare by depriving them of
objects indispensable to their survival, including wilfully impeding relief supplies as provided for under the
Geneva Conventions’.
[91] Article 8(2)(b)(xiii).
[92] Article 8(2)(b)(vii): ‘Making improper use of a flag of truce, of the flag or of the military insignia
and uniform of the enemy or of the United Nations, as well as of the distinctive emblems of the Geneva
Conventions, resulting in death or serious personal injury’.
[93] Article 8(2)(b)(vi).
[94] Article 8(2)(b)(xii), as read with article 40 of Additional Protocol I of 1977.
[95] Article 8(2)(b)(xvii).
[96] Article 8(2)(b)(xviii).
[97] Article 8(2)(b)(xix).
[98] See article 8(2)(b)(iii), which prohibits intentionally directing attacks against personnel,
installations, material, units or vehicles involved in a humanitarian assistance or peacekeeping mission in
accordance with the Charter of the United Nations, as long as they are entitled to the protection given to
civilians or civilian objects under the international law of armed conflict.
[99] Article 8(2)(b)(iv).
[100] Article 8(2)(b)(xxvi).
[101] Which is defined in article 7, para 2(f) as ‘the unlawful confinement of a woman forcibly made
pregnant, with the intent of affecting the ethnic composition of any population or carrying out other grave
violations of international law’.
[102] Article 8(2)(b)(xxii).
[103] Common Article 3 to the Geneva Conventions of 1949 proscribes the following acts, even when
committed during non-international armed conflicts: ‘(a) violence to life and person, in particular murder
of all kinds, mutilation, cruel treatment and torture; (b) taking of hostages; (c) outrages upon personal
dignity, in particular humiliating and degrading treatment; (d) the passing of sentences and the carrying
out of executions without previous judgment pronounced by a regularly constituted court, affording all
the judicial guarantees which are recognized as indispensable by civilized peoples.’
[104] Schabas (n 9) 65; Cryer (n 23) 276; Cassese et al International Criminal Law: Cases and
Commentary (2011) 12.
[105] Article 8(2)(d).
[106] See further, on Protocol II, Chapter 25 below on Humanitarian Law.
[107] Article 8(2)(e)(i).
[108] Article 8(2)(e)(ix).
[109] Article 8(2)(e)(x).
[110] Article 8(2)(e)(xii).
[111] Article 8(2)(e)(v).
[112] Article 8(2)(e)(ii).
[113] Article 8(2)(e)(iii).
[114] See the limitation contained in article 8(2)(f).
[115] Judgment of the International Military Tribunal for the Trial of German Major War Criminals,
Nuremberg, 30 September and 1 October 1946.
[116] Article 5(2) of the Rome Statute prior to amendment by RC/Res 6, 11 June 2010.
[117] For a very insightful account on the negotiation process and the different proposals put forward
at Kampala, see the article by the President of the Assembly of States Parties to the Rome Statute, C
Wenaweser ‘Reaching the Kampala compromise on aggression: The Chair’s perspective’ (2010) 23 Leiden
Journal of International Law 883–7. See also N Blokker and C Kress ‘A consensus agreement on the crime
of aggression: Impressions’ (2010) 23 Leiden Journal of International Law 890.
[118] RL Manson ‘Identifying the rough edges of the Kampala compromise’ (2010) 21(3) Criminal Law
Forum at 417.
[119] Resolution ICC-ASP/16/Res.5, adopted at the 13th plenary meeting on 14 December 2017 by
consensus of the Assembly of State Parties.
[120] A du Plessis and C Gevers ‘Africa and the codification of aggression: A pyrrhic victory?’ (2010)
2 African Legal Aid Quarterly 6.
[121] C Stahn ‘The “end”, the “beginning of the end” or the “end of the beginning”? Introducing
debates and voices on the definition of “aggression”’ (2010) 23 Leiden Journal of International Law 875 at
879. However, Ferencz argues that determining when military action falls under the exception
‘humanitarian intervention’ will be difficult since ‘what looks to some as humanitarian intervention may
appear to others as self-interested adventurism’; see DM Ferencz ‘The crime of aggression: Some
personal reflections on Kampala’ (2010) 23 Leiden Journal of International Law 905 at 907.
[122] See also C Kress and L von Holtzendorff ‘The Kampala compromise on the crime of aggression’
(2010) 8 JICJ 1179 at 1190.
[123] Ibid.
[124] See the House of Lords opinion by Lord Bingham in R v Jones [2006] UKHL 16 para 16.
[125] Article 8bis(2).
[126] Stahn (n 121) 876.
[127] Ibid 879.
[128] Stahn (n 121) 876.
[129] In particular, whether the Security Council would have the exclusive power to determine that an
act of aggression had taken place for the purposes of an ensuing prosecution. For instance, article 23(2)
of the ILC’s 1994 Draft Statute ‘suggested making ICC proceedings for the crime of aggression dependant
upon a prior determination of the Security Council of an act of aggression’; see Kress and Von
Holtzendorff (n 122) at 1194.
[130] Du Plessis and Gevers (n 120) 6.
[131] Blokker and Kress (n 117) 893.
[132] Article 15bis(4).
[133] KJ Heller ‘The sadly neutered crime of aggression’ available at opiniojuris.org.
[134] Ferencz (n 121) 906.
[135] Du Plessis and Gevers (n 120) 8. These limits of jurisdiction apply regardless of whether the
prosecution of the crime concerned is occasioned by a state party referral under article 14 or the exercise
of the Prosecutor’s proprio motu powers under article 15.
[136] Article 15bis(5).
[137] Manson (n 118) 419.
[138] Article 15bis(6).
[139] Scheffer notes that the Statute is oddly silent on what the Prosecutor may do if the Security
Council determines there has not been aggression: nothing in article 15bis ‘explicitly prohibits the ICC
from forging ahead even if the Security Council renders a negative determination’ (D Scheffer ‘The
complex crime of aggression under the Rome Statute’ (2010) 23 Leiden Journal of International Law 897
at 901).
[140] Manson (n 118) 419.
[141] It may, acting under Chapter VII of the UN Charter, refer a situation in which one or more of
such crimes appears to have been committed (including aggression) to the Prosecutor for investigation
under article 13(b). It may also, by way of a Chapter VII resolution, defer any investigation or
prosecution (including for aggression) for a period of 12 months—renewable under the same conditions—
under article 16 of the Statute. For more discussion, refer to Du Plessis and Gevers (n 120).
[142] Kress and Von Holtzendorff (n 122) 1211.
[143] Wenaweser (n 117) 887.
[144] Stahn (n 121) 882.
[145] Article 11.
[146] Article 11(2).
[147] See P Kirsch and D Robinson ‘Trigger mechanisms’ in A Cassese et al (eds) The Rome Statute of
the International Criminal Court: A Commentary vol 1 (2002) 623–5.
[148] P Akhavan ‘Self-referrals before the International Criminal Court: Are states the villains or the
victims of atrocities?’ (2010) 21 Criminal Law Forum 103 at 103; also see WA Schabas ‘First prosecutions
at the International Criminal Court’ (2006) 27 HRLJ 27; MH Arsanjani and WM Reisman ‘Law-in-action of
the International Criminal Court’ (2005) 99 AJIL 385.
[149] See Akhavan (n 148) and the ICC Appeals Chamber Prosecutor v Germain Katanga and Mathieu
Ngudjolo Chui (’Judgment on the appeal of Mr Germain Katanga against the oral decision of Trial
Chamber II of 12 June 2009 on the admissibility of the case’) ICC-01/04-01/07 OA8 (25 September
2009) para 78.
[150] See the discussion below of situations before the ICC for detail.
[151] See Kirsch and Robinson (n 147) 661–3.
[152] Article 13(b).
[153] See Kirsch and Robinson (n 147) 634.
[154] See, in general, Kirsch and Robinson (n 147) 630–1.
[155] SC Resolution 1593, 31 March 2005.
[156] SC Resolution 1970, 26 February 2011. The UN Security Council referred this situation to the
ICC, ‘condemning the violence and use of force against civilians, deploring the gross and systematic
violation of human rights, including the repression of peaceful demonstrators, expressing deep concern at
the deaths of civilians, and rejecting unequivocally the incitement to hostility and violence against the
civilian population made from the highest level of the Libyan government’, then under Muammar Gaddafi.
[157] See further M du Plessis, T Maluwa and A O’Reilly ‘Africa and the International Criminal Court’ (1
July 2013) Chatham House Report, available at
https://www.chathamhouse.org/sites/default/files/public/Research/International%20Law/0713pp_iccafric
a.pdf.
[158] Preamble, para 10 article 17. The principle of complementarity was acknowledged by the
Constitutional Court in S v Basson 2007 (3) SA 582 (CC) when it stated that the establishment of the ICC
in no way deprives national courts of responsibility for trying international crimes (para 172).
[159] See article 17 of the Rome Statute. A country may be determined to be ‘unwilling’ if it is clearly
shielding someone from responsibility for ICC crimes, and may be deemed ‘unable’ if its legal system has
collapsed.
[160] Schabas (n 9) 8; Cassese et al (n 104) 523; R Cryer (n 23) 153. See C Gevers ‘Back to the
future? Civil society, the “turn to complementarity” in Africa and some critical concerns’ 2016 Acta
Juridica 95, who argues that there is an emerging trend to prosecute international crimes domestically,
and that this could be beneficial to the international criminal legal project. For the role civil society has
and can play in the domestic prosecution of such crimes, see H Woolaver ‘Partners in complementarity:
The role of civil society in the investigation and prosecution of international crimes in South Africa’
2016 Acta Juridica 129. For discussion on complementarity and the important lessons from African states,
see M du Plessis, A Louw and O Maunganidze ‘African efforts to close the impunity gap: Lessons for
complementarity from national and regional actions’ Institute for Security Studies Paper, No 241
(November 2012).
[161] On the South African TRC’s amnesty process, see A McDonald ‘A right to truth, justice and a
remedy for African victims of serious violations of international humanitarian law’ (1999) 2 Law,
Democracy and Development 164–70; P Hayner Unspeakable Truths (2002) 98 et seq. On the amnesty
process in East Timor, see C Stahn ‘Accommodating individual criminal responsibility and national
reconciliation: The UN Truth Commission for East Timor’ (2001) 95 AJIL 962–5.
[162] On the importance of the TRC as truth-finder, see Azapo v President of the Republic of South
Africa 1996 (4) SA 671 (CC) 681–5.
[163] J Dugard ‘Conflicts of jurisdiction with truth commissions’ in A Cassese et al (eds) The Rome
Statute of the International Criminal Court: A Commentary (2002) 700–1.
[164] Where a criminal prosecution is instituted by a state under its domestic incorporating legislation,
amnesty does not have an extraterritorial effect and the prosecuting state is not required to recognise the
amnesty granted to human rights offenders by another state. See Dugard (n 163) 699.
[165] Dugard (n 163) 700.
[166] R v Bow Street Metropolitan Stipendiary Magistrate, Ex p Pinochet Ugarte (No 3) (1999) 2 All ER
97 (HL).
[167] Dugard (n 163) 699.
[168] Ibid. See, too, R Mani Beyond Retribution: Seeking Justice in the Shadows of War (2002) 112–
13.
[169] There is a vast body of literature on whether there is an international legal obligation (whether
founded in customary or conventional law) obliging states to punish past crimes. See, for example, D
Orentlicher ‘Settling accounts: The duty to prosecute human rights violations of a prior regime’ (1991)
100 Yale Law Journal 2537; N Roht-Arriaza ‘State responsibility to investigate and prosecute grave
human rights violations in international law’ (1990) 78 California Law Review 449. See also Dugard (n
163) 697.
[170] Dugard (n 163) 699.
[171] It is noteworthy that this trend has been reflected in the mandate of East Timor’s truth
commission. While the mandate is clearly supportive of individualised amnesty in exchange for truth, the
commission may grant ‘no immunity’ to persons who have committed a ‘serious criminal offence’, which
includes the international crimes of genocide, crimes against humanity, war crimes and torture, as well as
the domestic crimes of murder and sexual offences, as defined by the Indonesian Criminal Code. In 1999,
pro-Indonesian militia, supported by Indonesian security forces, used violence, threats and intimidation in
an attempt to coerce the East Timorese population to support continued integration in Indonesia in the
UN-organised 1999 referendum on independence for the island. In apparent revenge for the
overwhelming vote in favour of independence, an estimated 1000 supporters of independence were killed
and hundreds of thousands fled their homes or were forcibly expelled to Indonesia. After these events,
the United Nations took control of East Timor and, through its United Nations Transitional Administration
in East Timor, established the Commission for Reception, Truth and Reconciliation in East Timor. See
Stahn (n 161) 952–3.
[172] See further M du Plessis and J Ford ‘Transitional justice: A future truth commission for Zimbabwe’
(2009) 58 ICLQ 73 at 111–15; M du Plessis ‘Truth and reconciliation processes: Lessons for Zimbabwe?’
South African Institute of International Affairs, Report 44 (2002).
[173] Prosecutor v Thomas Lubanga Dyilo ‘Public redacted judgment on the appeal of Mr Thomas
Lubanga Dyilo against his conviction’ Appeals Chamber, ICC-01/04-01/06-3121-Red (1 December 2014);
and Prosecutor v Germain Katanga ‘Judgment rendu en application de l’article 74 du Statut’ Trial
Chamber II, ICC-01/04-01/07-3436 (20 March 2014).
[174] Prosecutor v Mathieu Ngudjolo Chui ‘Judgment on the prosecutor’s appeal against the decision of
Trial Chamber II entitled “Judgment pursuant to article 74 of the Statute”’ Appeals Chamber, ICC-01/04-
02/12-271-Corr (7 April 2015); Prosecutor v Jean-Pierre Bemba Gombo ‘Judgment on the appeal of Mr
Jean-Pierre Bemba Gombo against Trial Chamber III’s “Judgment pursuant to article 74 of the Statute”’
Appeals Chamber, ICC-01/05-01/08-3636-Red (8 June 2018).
[175] In this regard, see J Dugard ‘Palestine and the ICC: Institutional failure or bias?’ (2013)
11 JICJ 563.
[176] For a full account of the current preliminary investigations, see the 2017 report of the Office of
the Prosecutor on preliminary investigations, available at https://www.icc-
cpi.int/Pages/item.aspx?name=171204-rep-otp-PE.
[177] SC Resolution 1593, 31 March 2005.
[178] See Warrant of Arrest for Omar Hassan Ahmad al-Bashir, ICC-02/05-01/09 (4 March 2009)
available at www.icc-cpi.int/iccdocs/doc/doc639078.pdf.
[179] Prosecutor v Omar Hassan Ahmad al-Bashir ‘Second decision on the prosecution’s application for
a warrant of arrest’ ICC-02/05-01/09-94 (12 July 2010).
[180] ‘Decision under article 87(7) of the Rome Statute on the non-compliance by Jordan with the
request by the Court for the arrest and surrender or Omar al-Bashir’ Pre-Trial Chamber II, ICC-02/05-
01/09-309 (11 December 2017) para 38 (‘Jordan’); ‘Decision under article 87(7) of the Rome Statute on
the non-compliance by South Africa with the request by the Court for the arrest and surrender of Omar
al-Bashir’ Pre-Trial Chamber II, ICC-02/05-01/09-302 (6 July 2017) para 87 (‘South Africa’); ‘Decision on
the cooperation of the DRC’ ICC-02/05-01/09 (9 April 2014) para 26 (‘DRC’); ‘Decision on the non-
compliance by the Republic of Uganda with the request to arrest and surrender Omar al-Bashir to the
Court and referring the matter to the United Nations Security Council and the Assembly of State Parties
to the Rome Statute’ Pre-Trial Chamber II, ICC-02/05-01/09-267 (11 July 2016) para 12 (‘Uganda’);
‘Decision on the non-compliance by the Republic of Djibouti with the request to arrest and surrender
Omar al-Bashir to the Court and referring the matter to the United Nations Security Council and the
Assembly of the State Parties to the Rome Statute’ Trial Chamber II, ICC-02/05-01/09-266 11 July 2016
para 12 (‘Djibouti’); Prosecutor v Omar Hassan Ahmad al-Bashir ‘Decision Pursuant to Article 87(7) of the
Rome Statute on the Failure by the Republic of Malawi to Comply with the Cooperation Requests Issued
by the Court with Respect to the Arrest and Surrender of Omar Hassan al-Bashir’ ICC-02/05-01/09 (12
December 2011) para 40 (‘Malawi’).
[181] S Williams and L Sherif ‘The arrest warrant for President al-Bashir: Immunities of incumbent
heads of state and the International Criminal Court’ (2009) 14 Journal of Conflict & Security Law 71 at
71.
[182] M Mandel How America Gets away with Murder: Illegal Wars, Collateral Damage and Crimes
against Humanity (2004) 207.
[183] Schabas (n 9) 1; see also CS Igwe ‘The ICC’s favourite customer: Africa and international
criminal law’ (2008) 40 CILSA 294 at 295.
[184] See Igwe (n 183) 294. However, other situations were at the time under preliminary examination
by the Office of Prosecutor in Afghanistan, Colombia, Chad, Georgia and Guinea; in this regard, see M
Ssenyonjo ‘The International Criminal Court arrest warrant decision for President al-Bashir of Sudan’
(2010) 59 International & Comparative Law Quarterly 205.
[185] M du Plessis ‘Recent cases and developments: South Africa and the International Criminal Court’
(2009) 3 SACJ 441 at 442. See also EO Asaala ‘Rule of law or realpolitik? The role of the United Nations
Security Council in the International Criminal Court processes in Africa’ (2017) 17 AHRLJ 265; E Keppler
‘Challenges for international criminal justice in Africa and the role of civil society’ 2016 Acta Juridica 66.
[186] N Fritz ‘Black–white debate does no justice to a nuanced case’ Business Day 13 August 2008.
[187] See further discussion below.
[188] See Human Rights Watch ‘ICC: Bashir warrant is warning to abusive leaders’ 4 March 2009,
available at http://www.hrw.org/en/news/2009/03/04/icc-bashir-warrant-warning-abusive-leaders—
Human Rights Watch stated that the ICC warrant indicates that ‘[n]ot even presidents are guaranteed a
free pass for horrific crimes’; see also Amnesty International ‘ICC Issues Arrest Warrant for Sudanese
President al-Bashir’ 4 March 2009, available at http://www.amnesty.org/es/node/9632.
[189] Du Plessis (n 185) 443. Article 16 empowers the Security Council to defer an investigation or
prosecution for one year if it is necessary for the maintenance of international peace and security under
Chapter VII of the UN Charter. The Security Council would need to make a determination that the
continued involvement of the ICC is a greater threat to international peace and security than suspending
the ICC’s work.
[190] Du Plessis (n 185) 443. See also African Union Assembly ‘Decision on the meeting of African
States Parties to the Rome Statute of the International Criminal Tribunal (ICC)
(Doc.Assembly/AU/13(XIII))’ 3 July 2009, Assembly/AU/Dec.245(XIII) Rev.1; African Union Assembly
‘Decision on the progress report of the Commission on the Implementation of Decision
Assembly/AU/Dec.270(XIV) on the second ministerial meeting on the Rome Statute of the International
Criminal Court (ICC) Doc.Assembly/AU/10(XV)’ 27 July 2010, Assembly/AU/Dec.296(XV); African Union
Assembly ‘Decision on the implementation of the decisions on the International Criminal Court (ICC) Doc.
EX.CL/639(XVIII)’ 30–31 January 2011, Assembly/AU/Dec.334(XVI); African Union Assembly ‘Decision on
the implementation of the assembly decisions on the International Criminal Court—Doc.EX.CL/670 (XIX)’
30 June–1 July 2011, Assembly/AU/Dec.366 (XVII).
[191] D Tladi ‘The African Union and the International Criminal Court: The battle for the soul of
international law’ (2009) 34 SAYIL 57 at 57.
[192] See n 180.
[193] See, for example, E Ankumah (ed) The International Criminal Court and Africa: One Decade
On (2016); C Jalloh and I Bantekas (eds) The International Criminal Court and Africa (2017); M du
Plessis ‘The International Criminal Court that Africa wants’ Institute for Security Studies monograph
(August 2010).
[194] See the AU decisions, declarations and resolutions at the 30th Ordinary Session of the Assembly,
28–29 January 2018, Addis Ababa, Ethiopia, available at
https://au.int/sites/default/files/decisions/33908-assembly_decisions_665_-_689_e.pdf.
[195] See further M du Plessis ‘Prosecutor v al-Bashir: Decision under article 87(7) of the Rome Statute
on the non-compliance by South Africa with the request by the Court for the arrest and surrender of
Omar al-Bashir (Int’l Crim Ct)’ (2017) 56 ILM 1061.
[196] The ICC Act came into force on 16 August 2002. For further information on the Act, see M du
Plessis ‘Bringing the International Criminal Court home: The implementation of the Rome Statute of the
International Criminal Court Act’ (2003) 16 SACJ 1.
[197] Although customary international law forms part of South African law, a South African court
confronted with the prosecution of a person accused of an international crime would have been hard
pressed to convict, since the principle of nullum crimen sine lege would probably have constituted a bar
to any such prosecution. The same principle would most likely have also put paid to prosecutions under
the Geneva Conventions of 1949. South Africa has not incorporated the Geneva Conventions into
municipal law nor, prior to the ICC Act, enacted legislation to punish grave breaches. It would therefore
have been unlikely for a South African court to try a person for a grave breach of the Conventions in the
absence of domestic legislation penalising such conduct. This proposition was challenged before the South
African Constitutional Court in S v Basson 2007 (3) SA 582 (CC). However, the Court found it
unnecessary ‘to consider whether customary international law could be used . . . as the basis in itself for
a prosecution under the common law’ (para 172, fn 147). See Chapter 9 above.
[198] Section 5(2) of the ICC Act.
[199] See s 3(c) of the ICC Act.
[200] As mentioned previously in this chapter, for the purposes of interpreting and applying the
definitions of crimes found in articles 6, 7 and 8 of the Rome Statute, reference must also be made to the
‘Elements of Crimes’, a 50-page document adopted in June 2000 by the Preparatory Commission for the
International Criminal Court. See the ‘Finalized draft text of the elements of crimes’
(PCNICC/2000/INF/3/Add.2).
[201] For example, the secretary of state in the United Kingdom has, by regulation, made the Elements
of Crimes applicable to proceedings in a service court within the United Kingdom. See The International
Criminal Court Act 2001 (Elements of Crimes) Regulations 2001, available at
http://www.hmso.gov.uk/si/si2001/20012505.htm.
[202] In terms of s 38 of the ICC Act, the Minister of Justice may make regulations regarding the ICC
Act. In terms of s 1 of the Act, such regulations would be included as part of the Act.
[203] See Chapter 9 for a discussion on the different aspects to jurisdiction.
[204] The UK’s implementing legislation, for example, provides more clearly that, aside from the
traditional bases of jurisdiction (territoriality and nationality), the UK courts will have jurisdiction over a
person who ‘commits acts outside the United Kingdom at a time when he is not a United Kingdom
national, a United Kingdom resident or a person subject to UK service jurisdiction and who subsequently
becomes resident in the United Kingdom’ (see s 68(1) of the United Kingdom’s International Criminal
Court Act 2001).
[205] See, for example, Cassese et al The Rome Statute of the International Criminal Court (n 23) vol
2 at 1862.
[206] Section 2, ICC Act.
[207] See s 2(a).
[208] Section 5(1).
[209] Section 5(3).
[210] Section 5(4).
[211] Section 5(5).
[212] Section 5(5).
[213] Ibid.
[214] Section 9(3).
[215] Section 8(1).
[216] Section 8(2).
[217] Section 9(1).
[218] Section 10(1).
[219] Section 10(1)(a).
[220] Section 10(1)(b).
[221] Section 10(1)(c).
[222] See Chapter 11 for a full discussion of extradition.
[223] One must assume that the listing of these conditions is in the disjunctive.
[224] By contrast, the United Kingdom’s ICC Act, for example, makes it clear that a court, when
making an order for surrender, ‘is not concerned to enquire’ whether the warrant was duly issued by the
ICC or, where the person to be surrendered is ‘alleged to have committed an ICC crime, whether there is
evidence to justify his trial for the offence he is alleged to have committed’ (see s 5(5) of the
International Criminal Court Act 2001; see, too, the commentary on the Act by R Cryer ‘Implementation
of the International Criminal Court statute in England and Wales’ (2002) 51 ICLQ 733 at 736).
[225] Article 89 of the Rome Statute, which deals with surrender of persons to the Court, provides that
the Court ‘may transmit a request for the arrest and surrender of a person, together with the material
supporting the request’ to a state party, so this material would be before the magistrate. Prior to this, to
obtain a warrant of arrest from the ICC, the prosecutor would have had to convince a pre-trial chamber of
the Court (consisting of three judges) that there were ‘reasonable grounds to believe’ the suspect had
committed an ICC offence.
[226] The full list of areas of co-operation is set out in s 14(a)–(l). The list is modelled on article 93 of
the Rome Statute.
[227] Section 14 reads that the ‘relevant competent authorities in the Republic must, subject to the
domestic law of the Republic and the Statute, cooperate with, and render assistance to, the Court’
(emphasis added). The Constitution, where applicable, will no doubt provide the background standards
against which the relevant ‘co-operation’ is undertaken. So, for example, when it comes to searches and
seizures in terms of s 14(h), read with s 30 of the ICC Act, the relevant provisions of the Act will need to
be read in conjunction with ss 10, 12(1)(a)–(d), 12(2)(b), 14, 21, 35(5) and 36(1) of the Constitution.
[228] See ss 15, 16, 17, 18 and 19 of the ICC Act. The sections outline the procedure for the
examination of witnesses before a magistrate, the rights and privileges of the witness, the offences which
a witness might commit, and the procedure by which the attendance of a witness might be secured in
proceedings before the International Criminal Court.
[229] See s 20 of the ICC Act.
[230] Ibid s 21.
[231] Ibid s 30. This section is in many respects similar to those provisions of the Criminal Procedure
Act 51 of 1977 relating to search and seizure (ss 19–36) but with modifications to reflect the fact that the
request for co-operation has been made by the ICC for the purposes of its investigation and not to assist
South Africa in criminal investigations unrelated to the ICC.
[232] Sections 14(k), 22(1) and 27(1). For fuller discussion, see Du Plessis (n 196) 10–12.
[233] Proclamation 43 of 2003.
[234] Section 6.
[235] Section 7(1).
[236] Section 7(2).
[237] Section 7(3).
[238] Section 7(4).
[239] Section 7(5).
[240] See article 103(3)(a) of the Rome Statute as well as rule 201 of the Rules of Procedure and
Evidence.
[241] If no state offers its prison services, the host state of the ICC—the Netherlands—will perform the
task (see article 103(4) of the Rome Statute).
[242] Article 103(3).
[243] Article 106(2).
[244] Section 31.
[245] Section 32.
[246] 111 of 1998.
[247] Section 32(4)(b). This provision is a reflection of the prescription in article 110(2) of the Rome
Statute whereby the ICC ‘alone shall have the right to decide any reduction of sentence’.
[248] Schabas (n 9) 75.
[249] South African jails are notoriously overcrowded and prisoners are often kept in inhuman
conditions. See, for example, Sonke Gender Justice v Government of South Africa (unreported, ZAWCHC
Case No 24087/15, 22 December 2016 and 23 February 2017) available at
http://genderjustice.org.za/publication/pollsmoor-court-order/.
[250] Article 77(2)(a).
[251] Article 75(2).
[252] Sections 25 and 26.
[253] Section 25(2) and (3).
[254] Section 26.
[255] ‘Decision following the prosecutor’s request for an order clarifying that the Republic of South
Africa is under the obligation to arrest and surrender Omar al-Bashir’ ICC-02/05-01/09 (13 June 2015)
para 2.
[256] ‘Prosecution’s submissions for article 87(7) hearing’ ICC-02/05-01/09-289 (17 March 2017)
paras 90 et seq.
[257] Southern Africa Litigation Centre v Minister of Justice and Constitutional Development 2015 (5)
SA 1 (GP).
[258] Minister of Justice and Constitutional Development v Southern Africa Litigation Centre 2016 (3)
SA 317 (SCA) (‘Minister of Justice v SALC’).
[259] ‘Decision under article 87(7) of the Rome Statute on the non-compliance by South Africa with the
request by the Court for the arrest and surrender of Omar al-Bashir’ Pre-Trial Chamber II, ICC-02/05-
01/09-302 (6 July 2017).
[260] After South Africa, the ICC made the same finding in respect of Jordan (although this decision is
subject to appeal). Before South Africa, Malawi, Chad, the Democratic Republic of Congo, Djibouti and
Uganda were held to have violated their obligations. See n 180.
[261] Chapter 12.
[262] For a discussion of the various sorts of immunity, see Chapter 12 generally and in particular see
the discussion there on the important distinction between immunity as a defence or bar to jurisdiction
and immunity in respect of a request for judicial assistance.
[263] Read with s 4(2)(a) of the Act.
[264] D Tladi ‘The duty on South Africa to arrest and surrender President al-Bashir under South African
and international law’ (2015) 13 JICJ 1027 at 1039 (referencing M du Plessis ‘South Africa’s
implementation of the ICC Statute: An African example’ (2007) 5 JICJ 460).
[265] Minister of Justice v SALC (n 258) para 99.
[266] Ibid para 101.
[267] Discussed below.
[268] Section 233 of the Constitution.
[269] Cassese et al The Rome Statute of the International Criminal Court (n 23) vol 2 at 1857
(emphasis added). And as Wallis JA held in Minister of Justice v SALC (n 258) para 103: ‘I conclude
therefore that when South Africa decided to implement its obligations under the Rome Statute by passing
the Implementation Act it did so on the basis that all forms of immunity, including head of state
immunity, would not constitute a bar to the prosecution of international crimes in this country or to South
Africa cooperating with the ICC by way of the arrest and surrender of persons charged with such crimes
before the ICC, where an arrest warrant had been issued and a request for cooperation made. I accept, in
the light of the earlier discussion of head of state immunity, that in doing so South Africa was taking a
step that many other nations have not yet taken. If that puts this country in the vanguard of attempts to
prevent international crimes and, when they occur, cause the perpetrators to be prosecuted, that seems
to me a matter for national pride rather than concern. It is wholly consistent with our commitment to
human rights both at a national and an international level. And it does not undermine customary
international law, which as a country we are entitled to depart from by statute as stated in s 232 of the
Constitution. What is commendable is that it is a departure in a progressive direction.’
[270] For instance, Lord Nicholls in the first Pinochet case held that ‘there can be no doubt that if
Senator Pinochet had still been the head of the Chilean state, he would have been entitled to immunity’
(see R v Bow St Magistrate, Ex p Pinochet Ugarte [1998] 4 All ER (Pinochet 1) 938). Lord Millett in the
third Pinochet case said that ‘Senator Pinochet is not a serving head of state. If he were, he could not be
extradited. It would be an intolerable affront to the Republic of Chile to arrest him or detain him’ (see R v
Bow St Magistrate, Ex p Pinochet (No 3) [1999] 2 WLR 824 at 905H).
[271] Arrest Warrant of 11 April 2000 (Democratic Republic of the Congo v Belgium) 2002 ICJ Reports
3. For criticisms of the decision, see the discussion on immunity in criminal proceedings in Chapter 12.
[272] Paragraph 58.
[273] Minister of Justice v SALC (n 258) para 95. See further MJ Ventura ‘Escape from Johannesburg?:
Sudanese President al-Bashir visits South Africa, and the implicit removal of head of state immunity by
the UN Security Council in light of Al-Jedda’ (2015) 13 JICJ 995.
[274] J Dugard and C Abraham ‘Public international law’ 2002 Annual Survey 140 at 166.
[275] Du Plessis (n 264) 474.
[276] Tladi (n 264) 1038; Woolaver (n 160) fn 78.
[277] See Arrest Warrant (n 271) para 61. A different reasoning was offered by the Special Court for
Sierra Leone in Prosecutor v Charles Ghankay Taylor Case No SCSL-2003-1-AR72(E) (31 May 2004). The
Court found that customary international law and the international nature of the Court, not state consent,
automatically excluded immunity as a procedural bar. For a criticism of this reasoning, see Cryer (n 83)
552. For its support, see P Gaeta ‘Does President al-Bashir enjoy immunity from arrest?’ (2009)
7 JICJ 315 at 317.
[278] See n 180.
[279] Article 25 of the UN Charter.
[280] Ventura (n 273) 1013–14.
[281] Ibid 1018. Jordan made this argument in Jordan (n 180) para 16 and South Africa made this
argument in South Africa (n 180) para 35.
[282] Ventura (n 273) 1021.
[283] Cryer (n 23) 557; Malawi (n 180) para 40.
[284] D Akande ‘The legal nature of Security Council referrals to the ICC and its impact on al-Bashir’s
immunities’ (2009) 7 JICJ 333 at 341–2. Jordan (n 180) para 38; South Africa (n 180) para 91.
[285] Gaeta (n 277) 324.
[286] Cf the Malawi Decision Case No ICC-02/05-01/09 (12 December 2011).
[287] See the Hashemite Kingdom of Jordan’s appeal against the ‘Decision under article 87(7) of the
Rome Statute on the non-compliance by Jordan with the request by the Court for the arrest and
surrender o[f] Omar al-Bashir’ ICC-02/05-01/09-326 (12 March 2018).
[288] See the order of the Appeals Chamber, ICC-02/05-01/09 OA2 (29 March 2018).
[289] DRC supra (n 180) para 27.
[290] DRC supra (n 180) para 29.
[291] Treaty provisions, moreover, are to be interpreted in light of their purpose. See article 31(1) of
the Vienna Convention on the Law of Treaties.
[292] See further on this subject, G Mettraux, J Dugard and M du Plessis ‘Heads of state immunities,
international crimes and President Bashir’s visit to South Africa’ (2018) 18 International Criminal Law
Review 1.
[293] South Africa (n 180) para 93; Jordan (n 180) para 39.
[294] Democratic Alliance v Minister of International Relations and Cooperation (Council for the
Advancement of the South African Constitution Intervening) 2017 (3) SA 212 (GP). The government
chose not to appeal the decision.
[295] Ibid para 53. See, for a critical discussion, M du Plessis and G Mettraux ‘South Africa’s failed
withdrawal from the Rome Statute: Politics, law, and judicial accountability’ (2017) 15 JICJ 361.
 [296] [B 37–2017].
[297] See ‘South Africa should not withdraw from the International Criminal Court: Briefing submitted
to the Portfolio Committee on Justice and Correctional Services’ International Commission of Jurists (in
collaboration with a number of leading jurists), available at
http://www.lhr.org.za/sites/lhr.org.za/files/south_africa_icc_briefer_formatted2.pdf?ct=t%28LHR-
ICJ+Press+Conference%29&mc_cid=b24c9ee660&mc_eid=%5BUNIQID%5D.
[298] Minister of Justice v SALC (n 258) para 103.
[299] See M du Plessis ‘Ramaphosa can renew South African leadership by recommitting to the ICC’ (6
February 2018) Chatham House Expert Comment, available at
https://www.chathamhouse.org/expert/comment/ramaphosa-can-renew-south-african-leadership-
recommitting-icc.
[300] Section 3(1).
[301] Article 127(2) of the Statute.
[302] See in this regard N Pillay and A Mudukuti ‘South Africa and the ICC: Dismantling the
international criminal justice system to protect one individual?’ Daily Maverick 19 June 2018, available at
https://www.dailymaverick.co.za/article/2018-06-19-south-africa-and-the-icc-dismantling-the-
international-criminal-justice-system-to-protect-one-individual/#.WzdjYy2B2gw.
[303] Chapter 2.
[304] Chapter 4. It also provides for the ‘extradition’ of persons to foreign states for international
crimes.
[305] Section 1, definition of ‘entity’.
[306] For details of how that might be done, see Du Plessis and Mettraux (n 295) 369; and M du Plessis
‘The Omar al-Bashir case: Exploring efforts to resolve the tension between the African Union and the
International Criminal Court’ in T Maluwa, M du Plessis and D Tladi (eds) The Pursuit of a Brave New
World in International Law: Essays in Honour of John Dugard (2017) 431.

Page 303

Chapter 11
Extradition

Anton Katz, Max du Plessis, Eshed Cohen and John Dugard

Extradition [1] may be defined as ‘the delivery of an accused or a convicted individual to
the state where he is accused of, or has been convicted of, a crime, by the state on
whose territory he happens for the time to be’. [2]
International law does not recognise any general duty on the part of states to
surrender criminals. [3] In practice, international obligations to return criminals are
sourced in extradition agreements between states. [4] These treaties can be bilateral or
multilateral. [5] State practice demonstrates a preference for bilateral treaties. The
United States, for example, has over 100 such treaties in force. [6] South Africa has
eight. [7] But South Africa is also party to the European Convention on Extradition, [8] a
multilateral extradition treaty between 50 (mostly European) states. [9]
Page 304
Some obligations pertaining to extradition are also sourced in multilateral treaties
that do not deal exclusively with extradition. For example, the Convention on the
Prevention and Punishment of the Crime of Genocide of 1948, in article 7, obliges
countries to grant extradition requests for those accused or convicted of
genocide. [10] Conversely, the Convention Against Torture prohibits extradition if the
requested state has substantial grounds for believing that the sought person will be
subject to torture. [11]
International law does not require extradition treaties to follow a particular form.
This means that there are general principles that have emerged from the practice of
states which are commonly incorporated into extradition agreements. In 1990 the
General Assembly of the United Nations approved a Model Treaty on Extradition
containing many of these principles, which aims to provide ‘a useful framework’ for
states in the negotiation and revision of bilateral agreements. [12]
Absent a treaty, a state could still grant an extradition request made by another
state. Some states, including South Africa, invoke reciprocity and comity as bases for
extradition in the absence of a treaty. Comity describes those actions between states
based solely on goodwill or courtesy. Reciprocity is when the request for surrender is
accompanied by assurances of reciprocal extradition in comparable
circumstances. [13] The primary difference between the two is that comity does not
presuppose similar counter-performances by the requesting state at a later stage. Both
reflect international principles of friendly co-operation among states. [14] They thus
enable a state to extradite a sought person on the request of another state in the
absence of a treaty. The conditions under which decisions on the basis of reciprocity or
comity are made vary from state to state—it all depends on the domestic empowering
provisions for the decision. [15]
Page 305
Before 1961, when South Africa became a Republic and left the Commonwealth,
extradition between South Africa and other states was governed by two international
arrangements. Extradition between South Africa and other Commonwealth countries
was regulated by the British Fugitive Offenders Act of 1881, [16] while extradition
between South Africa and non-Commonwealth states was governed by the British
Extradition Acts of 1870 to 1906, [17] which allowed the British government to extend
extradition treaties entered into between the British government and other states to
British colonies or dominions. After South Africa acquired full treaty-making power, the
South African government was able to enter into treaties with other states on its own
behalf, and in this case the treaties were made applicable to South Africa under the
Extradition Act of 1870 and the Royal Executive Functions and Seals Act of 1934. [18]
When South Africa left the Commonwealth in 1961, all its extradition arrangements
with Commonwealth countries, including the United Kingdom, came to an end. South
Africa’s departure from the Commonwealth and new Republican status had no effect,
however, on its extradition treaties with non-Commonwealth states, which remained in
force. [19] In 1962 the Extradition Act 67 of 1962 was enacted to govern the country’s
extradition relations. This statute, which remains in force, empowered the President to
enter into extradition agreements with foreign states. [20]
South Africa’s political isolation during the apartheid era made it almost impossible
for it to extend its network of extradition agreements. [21] Agreements entered into by
the British government on behalf of South Africa under the 1870 to 1906 Extradition
Acts were carefully preserved, [22] but new agreements were limited to states
Page 306
within the southern African region [23] and to fellow pariah states—Israel [24] and the
Republic of China (Taiwan). [25]
In these circumstances a special arrangement was made to allow South Africa to
return a fugitive to a state with which there was no extradition agreement. The State
President was empowered by s 3(2) of the 1962 Extradition Act to consent to the
surrender of such a person on an ad hoc basis.
South Africa’s return to international respectability has rescued it from isolation in
respect of extradition. In 1996, South Africa became covered by the Commonwealth
Scheme relating to the Rendition of Fugitive Offenders of 1990, [26] when the Extradition
Act was amended to allow extradition to a country designated [27] by the President
without the need for an extradition agreement. [28]
The Commonwealth Scheme is not a multilateral treaty but rather an agreed
guideline of principles on extradition which forms the basis of reciprocating legislation
enacted in Commonwealth states. The Extradition Amendment Act of 1996 [29] amends
the Extradition Act of 1962, inter alia, to bring South African law in line with the
guidelines of the Commonwealth Scheme. In 1998, Namibia, Zimbabwe and the United
Kingdom were ‘designated’ for the purposes of extradition. [30]
In the past decade, South Africa has added considerably to its arsenal of extradition
treaties. For instance, in 2001 it entered into a
Page 307
new agreement with the United States and, in 2003, it acceded to the European
Convention on Extradition of 1957. [31] It has also entered into treaties with Australia,
Canada, India and China. [32] In 2012, South Africa also signed and ratified the SADC
Protocol on Extradition, a multilateral treaty dealing with extradition among states in
the SADC region. [33]

The impact of the Constitution

1 Human rights

Extradition has not escaped the impact of human rights law. [34] Some human rights
principles have been adopted by extradition agreements; others have been used to
obstruct extradition despite their absence from the extradition agreement. In the latter
case, it is claimed that international human rights norms, whether based in treaty or
custom, ‘trump’ extradition treaty obligations on the ground that they enjoy a higher
status as part of the public order of the international community or of a particular
region.
In South Africa, the Constitutional Court has recognised that extradition involves an
invasion of fundamental human rights. [35] Because South Africa has a justiciable Bill of
Rights, extradition proceedings must protect the rights of those who are accused of
crimes by the requesting country. [36] Otherwise the extradition and the relevant law
could be invalid. [37]
The two principal human rights norms that have been adopted by extradition treaties
and legislation concern the death penalty and
Page 308
non-discrimination. Today it is common practice for extradition agreements to exclude
extradition where the crime in respect of which extradition is sought is punishable by
death in the state requesting extradition but not the requested state, unless the
requesting state provides a satisfactory assurance that the death penalty will not be
imposed, or, if imposed, will not be executed. [38] In Mohamed v President of the RSA
(Society for the Abolition of the Death Penalty in South Africa Intervening) [39] the
Constitutional Court held that it was unconstitutional to extradite any person (including
undocumented foreigners) to a country where he or she may face the death penalty if
put on trial. South Africa can only lawfully extradite a person if the requesting state
provides an assurance that the death penalty will not be sought, or, if imposed, will not
be carried out. As the Constitutional Court put it in Minister of Home Affairs and Others
v Tsebe and Others: [40]
We as a nation have chosen to walk the path of the advancement of human rights. By adopting
the Constitution we committed ourselves not to do certain things. One of those things is that
no matter who the person is and no matter what the crime is that he is alleged to have
committed, we shall not in any way be party to his killing as a punishment and we will not
hand such person over to another country where to do so will expose him to the real risk of the
imposition and execution of the death penalty upon him. This path that we, as a country, have
chosen for ourselves is not an easy one. Some of the consequences that may result from our
choice are part of the price that we must be prepared to pay as a nation for the advancement
of human rights and the creation of the kind of society and world that we may ultimately
achieve if we abide by the constitutional values that now underpin our new society since the
end of apartheid.
It is also increasingly the practice for treaties to exclude extradition where the
requested state ‘has substantial grounds for believing that a request for extradition for
an ordinary criminal offence has been made for the purpose of prosecuting or punishing
a person on account of his race, religion, nationality or political opinion, or that
Page 309
that person’s position may be prejudiced for any of these reasons’. [41] In 1996 the
Extradition Act was amended to include such a bar to extradition, extended to cover
gender discrimination. [42] The UN Model Treaty on Extradition [43] and the Convention
against Torture [44] prohibit extradition where there are substantial grounds for
believing that the extradited person will be subjected to torture or to cruel, inhuman or
degrading treatment or punishment in the requesting state. The UN Model Treaty on
Extradition also prohibits extradition if the extradited person is unlikely to receive a fair
trial. [45]
The manner in which an obligation contained in a human rights treaty may trump an
extradition treaty obligation is illustrated by the decision of the European Court of
Human Rights in the Soering Case. [46] Soering, a West German national, murdered his
girlfriend’s parents in Virginia and fled to the United Kingdom, from which his
extradition was requested by the United States. After the United Kingdom ordered his
extradition, he petitioned the European Commission of Human Rights, which referred
the case to the European Court of Human Rights. The Court held that the United
Kingdom was not required, by article 3 of the European Convention on Human Rights,
which prohibits torture and inhuman or degrading treatment or punishment, to
extradite Soering to the United States, where there was a real risk that he would be
subjected to inhuman or degrading treatment by being kept on death row for a
prolonged period in the state of Virginia. The Court found that the fact that the actual
human rights violation would take place outside the United Kingdom did not absolve the
United Kingdom from responsibility for any foreseeable consequence of extradition
suffered outside its jurisdiction. The same approach was followed by the United Nations
Human Rights
Page 310
Committee in Ng v Canada. [47] Here, the Committee held that Canada had violated its
obligations under article 7 of the International Covenant on Civil and Political Rights
prohibiting cruel, inhuman or degrading treatment or punishment by extraditing Ng to
the United States when it could reasonably have foreseen that, if sentenced to death in
California, he would be executed by gas asphyxiation, a form of punishment in violation
of that prohibition.
There are other human rights violations relating to the prohibition against cruel,
inhuman or degrading treatment or punishment that may, it has been suggested,
obstruct extradition. They are life imprisonment, harsh prison conditions, corporal
punishment and brutal pre-trial interrogation methods. [48]
This is a rapidly developing branch of extradition law. The traditional view that the
courts of a requested state will not examine the standards of justice applied in the
requesting state is no longer followed by many states. [49] Most courts today seriously
consider the human rights implications of extradition; and there is increasing practice in
favour of making extradition conditional upon the giving of satisfactory assurances by
the requesting state that the extradited person’s human rights will be respected in that
state. [50]
In Kaunda and Others v President of the Republic of South Africa and Others, [51] an
appeal was made to the above principles in highly peculiar circumstances. The
applicants, South African nationals, were arrested in Zimbabwe on suspicion of being
mercenaries en route to Equatorial Guinea to stage a coup against the President of that
state. Fearing that they might be extradited to Equatorial Guinea, where they
anticipated an unfair trial followed by the imposition of the death penalty, they brought
proceedings in South Africa to compel the South African government, inter alia, to seek
an assurance from the Zimbabwean government that they would not be extradited to
Equatorial Guinea, alternatively, if so extradited, to seek assurances that they would
receive a fair trial in Equatorial Guinea and not be sentenced to death. The
Constitutional Court dismissed the application largely on the ground that it was
premature because there was
Page 311
no clear evidence that the applicants were likely to be extradited to Equatorial
Guinea. [52] Moreover, the Court stressed that the arrest of the applicants in Zimbabwe
and ‘the possibility of their being extradited from Zimbabwe to Equatorial Guinea (were)
not the result of any unlawful conduct on the part of the government or of the breach of
any duty owed to them’. [53] The Court’s judgment does, however, give tentative
support to the need to respect human rights in the extradition process. The Court
stated:
[I]f the allegations by the applicants that they will not get a fair trial in Equatorial Guinea
prove to be correct, and they are convicted and sentenced to death, there will have been a
grave breach of international law harmful to our government’s foreign policy and its aspirations
for a democratic Africa . . . .
It cannot . . . be said that there is not a risk that the consequences that the applicants fear
will happen. Should that risk become a reality the Government would be obliged to respond
positively. Given its stated foreign policy, there is no reason to believe that this will not be
done. [54]
The Court furthermore stressed that ‘decisions made by the government in these
matters are subject to constitutional control’. [55]
The extradition agreement between South Africa and the People’s Republic of China
of 2005 contains a provision that would permit South Africa to deny extradition that
might result in the violation of human rights. Article 4(b) permits extradition to be
refused where ‘the probable penalty that may be imposed in the Requesting State is in
conflict with the fundamental principles of the laws of the Requested State’. [56]
 Mohamed did not explicitly deal with what would happen if a requesting state refused
to give an assurance that the death penalty would not be imposed on a sought person.
However, the Constitutional Court in Tsebe dealt with precisely this issue.
In Tsebe Botswana requested the extradition of Mr Tsebe and Mr Phale, who had fled
to and entered South Africa. Both citizens were accused of murder, and the extradition
was requested to try the two for murder. However, in Botswana, a person convicted of
murder could face the death penalty. The South African Minister of Justice, as
per Mohamed, thus requested an assurance from Botswana that the
Page 312
two sought persons would not be executed if convicted. Botswana refused to give this
assurance because its domestic legislation and the treaty with South Africa did not
provide for the giving of such an assurance. Despite all this, the Department of Home
Affairs decided to deport the two. The issue therefore was whether South Africa could
deport or extradite the two even though Botswana had not given the requested
assurance.
The Constitutional Court reiterated the principle in Mohamed that the government
has no power to extradite or deport or in any way remove from South Africa to a
retentionist state any person who, to its knowledge, if deported or extradited to such a
state, will face the real risk of the imposition and execution of the death penalty. To do
so would be to contravene the government’s obligation to fulfil and protect the right to
life. [57]
The Court went on to find that extradition or deportation cannot occur where a
requesting state refuses to give the assurance regarding the death sentence. That is,
the fact that a requesting state refuses to give the assurance does not change South
Africa’s obligations to protect and fulfil the right to life. And these obligations prevent
extradition or deportation without the requisite assurance. [58]
The Court also noted that article 6 of the extradition treaty between South Africa and
Botswana allowed either country to refuse extradition ‘if under the law of the requesting
Party the offence for which extradition is requested is punishable by death and if the
death penalty is not provided for such offence by the law of the requested
Party’. [59] South Africa, by refusing extradition, was therefore not breaching its duties
under international law to Botswana.
Moreover, article 5(c) of the SADC Protocol on Extradition, to which Botswana and
South Africa are party, allows for Botswana to make the requisite assurance, and allows
South Africa to refuse extradition in the absence of such an assurance. [60] However,
South Africa, if it refuses to extradite because there is no assurance, shall, if the other
state so requests, ‘submit the case to its competent authorities with a view to taking
appropriate action against the person for the offence for which extradition had been
requested’.
On the facts of Tsebe, therefore, Botswana was empowered under international law
to give an assurance that neither sought person would be executed if convicted. The
Court thus encouraged the South African government to continue to seek a political
solution to
Page 313
the extradition and acquire the assurance from Botswana. [61] Until then, however, the
Court held that neither could lawfully be deported or extradited to Botswana.
2 The relationship between international and municipal law

The Constitution’s treatment of international law is discussed in Chapter 4. Given how

the Constitution relates international law to the South African municipal system, there
are a number of constitutional anomalies arising from the present Extradition Act. In
the first place, it is difficult to reconcile the power of the President to consent to an ad
hoc extradition under s 3(2) [62] with the requirement in s 231 of the 1996 Constitution
that treaties be approved by Parliament to bind the Republic internationally, and that
they be incorporated by national legislation to have domestic effect. The President’s
consent to extradite under s 3(2) is preceded by an exchange of notes between South
Africa and the state requesting extradition, which appears to constitute an international
agreement, for which s 231 requires approval by Parliament as well as incorporation
into domestic law by national legislation. However, in Harksen v President of the
RSA, [63] the Constitutional Court held that the consent of the President under s 3(2)
‘was a domestic act never intended to create international legal rights and
obligations’, [64] and that the exchange of notes (in this case between South Africa and
Germany) did not constitute an agreement and therefore did ‘not provide support for
the conclusion that the President’s consent under s 3(2) was anything more than a
domestic act’. [65] The Court further dismissed the argument that the President’s
Page 314
consent under s 3(2) allowed the executive to bypass the requirement of parliamentary
approval for treaties laid down in s 231(2) of the Constitution. [66] In Geuking v
President of the Republic of South Africa, the Constitutional Court dismissed a challenge
to s 3(2) on the ground that it does not empower the President to consent to surrender
a person to a foreign state. [67] The Court seemed to hold that such a power was
necessarily implied in s 3(2). [68]
Secondly, another anomaly arising from the Extradition Act is that the procedure for
incorporating extradition agreements entered into after 1996, and ‘designation
arrangements’ made after this date, into domestic law, fails to follow the requirements
prescribed by s 231 of the 1996 Constitution. The Act provides, correctly, that such
agreements and designations must be approved by Parliament, [69] as required by s
231(2), but it does not provide for the incorporation of the agreement or ‘designation’
into domestic law by ‘national legislation’ as required by s 231(4). Instead, it merely
provides that the Minister of Justice shall give notice of Parliament’s approval of the
agreement or ‘designation’ in the Government Gazette, [70] which notice does not
qualify as ‘national legislation’. [71] This anomaly arises because the Extradition
Amendment Act 77 of 1996 was adopted while the Interim Constitution was still in force
and before the adoption of the Constitution of the Republic of South Africa,
1996. [72] The Interim Constitution did not require the legislative incorporation of
treaties. The failure of the Extradition Act to provide for incorporation as required by s
231(4) of the Constitution was challenged in President of the Republic of South Africa v
Quagliani, and Two Similar Cases. [73] The Constitutional Court avoided dealing directly
with the challenge and held instead—in a form of reasoning that is difficult to follow—
that ‘[e]ither the Agreement has “become law” in South Africa as a result of the prior
existence of the [Extradition] Act which constitutes the anticipatory enactment of the
Agreement for the purposes of section 231(4) of the Constitution. Or the Agreement
has not “become law” in the Republic as contemplated by section 231(4)
Page 315
but the provisions of the Act are all that is required to give domestic effect to the
international obligation that the Agreement creates.’ [74]
The decision in Quagliani has been criticised as not having adequately addressed the
constitutional difficulty. [75]

3 The rule of law

Most extraditions occur in terms of an extradition agreement. In such a case,

extradition proceedings are governed both by the particular extradition agreement in
terms of which the foreign state requests extradition and by the Extradition Act 67 of
1962. While the extradition agreement determines the offences in respect of which
extradition is possible and the circumstances in which extradition may be refused, the
Extradition Act prescribes the procedure to be followed in extradition proceedings and
some of the circumstances in which extradition may be refused.
Because the extradition process is regulated by an Act, the process entails the
exercise of public power by various actors at different stages. For the most part, there
is no issue that the powers given to the Minister of Justice and magistrates by the
Extradition Act are public in nature. Although an extradition enquiry before a magistrate
is judicial (and not administrative in nature) [76] the Supreme Court of Appeal held
in Zuma v National Director of Public Prosecutions: [77]
It is true, as counsel for the appellants reminded us, that the rule of law and the principle of
legality require State conduct (which includes the conduct of a judge) to be in accordance with
law.
This means that these powers must be exercised lawfully, in good faith and
rationally. [78] If not, such decisions can be reviewed and set aside. The review of
magistrates’ and the Minister’s decisions under the Extradition Act are dealt with below.
But it is important to note that since the advent of the Constitution, the bases and
grounds for reviewing these decisions have been tempered by the Constitution’s
commitment to the rule of law. In Fedsure Life Assurance Ltd v Greater Johannesburg
Transitional Metropolitan Council, the Constitutional
Page 316
Court explained that ‘[t]he rule of law—to the extent at least that it expresses this
principle of legality—is generally understood to be a fundamental principle of
constitutional law’. [79] The South African Constitution acknowledges this in s 1(c), which
mandates that South Africa is a Republic founded on the value of the rule of law.

Extradition distinguished from other forms of removal

1 Deportation (disguised extradition) from and to South

Africa

‘Disguised extradition’ may be achieved by deporting a fugitive to a state in which he is

accused of a crime, in accordance with deportation procedures. This practice is widely
condemned [80] as it deprives the deportee of the rights to which he would be entitled if
he were extradited. Inter alia, it deprives him of the right to raise the political offence
exception. [81]
 The threat posed to human rights by disguised extradition is starkly illustrated
by Mohamed v President of the RSA. [82] Mohamed, a Tanzanian national, fled to South
Africa after participating in the bombing of the US embassy in Dar es Salaam in 1998.
He entered South Africa on a false passport and under an assumed name on a visitor’s
visa. In 1999 Mohamed’s whereabouts were discovered by the US Federal Bureau of
Investigation (FBI) and it then entered into negotiations with the South African police
and immigration authorities for the expeditious transfer of Mohamed to New York,
where he had been indicted for murder for the bombing of the US embassy. Although
there was an extradition agreement in force between the United States and South
Africa, it was decided to declare him to be a prohibited person and deport him
immediately to the United States.
Page 317
As a result, he was not afforded the opportunity to require the South African
government to obtain an assurance from the government of the United States that, if
convicted, he would not be sentenced to death and executed. Thus, he was treated
differently from a co-accused whose extradition from Germany was made conditional
upon an assurance from the United States that he would not be sentenced to death. [83]
The Constitutional Court held that the deportation of Mohamed violated both the
Aliens Control Act, 1991 and the Constitution. The former statute permits deportation
only to a country of which the person is a national. [84] The Constitution had been
violated in that the South African immigration authorities had failed to obtain a prior
undertaking from the US government that, if convicted, the death penalty would not be
imposed on Mohamed. This failure infringed his rights to human dignity, to life and to
not to be punished in a cruel, inhuman or degrading manner. The Court made it clear
that a person ought not to be deported or extradited to another state where there was
a real risk that his basic human rights would be violated in that state. [85] It stated:
For the South African government to co-operate with a foreign government to secure the
removal of a fugitive from South Africa to a country of which the fugitive is not a national and
with which he had no connection other than that he is to be put on trial for his life there, is
contrary to the underlying values of our Constitution. It is inconsistent with the government’s
obligation to protect the right to life of everyone in South Africa, and it ignores the
commitment implicit in the Constitution that South Africa will not be a party to the imposition
of cruel, inhuman or degrading punishment. [86]
The Court ordered that its judgment be delivered to the Federal Court in New York
before which Mohamed stood trial. Mohamed was convicted but not sentenced to death.
The Constitutional Court’s decision in Mohamed has not always been scrupulously
followed by Home Affairs officials. That is best illustrated by the removal of Mr Rashid
from South Africa to Pakistan during 2005 on account of his alleged involvement in
terrorism, and under circumstances where Rashid appeared to be at risk of torture or
inhuman treatment if delivered to Pakistan. Such removals, to what have become
known as ‘black sites’ in the so-called
Page 318
war on terror, have been condemned judicially by South African courts and others. [87]

Where a person has been irregularly deported to South Africa in a ‘disguised

extradition’, a South African court should refuse to exercise jurisdiction. In S v
Rosslee [88] the Cape Provincial Division held otherwise, where an accused person was
deported to South Africa by Namibia (with which South Africa at that time had no
extradition agreement) after a ‘tip off’ from the South African police. This decision,
which cannot be reconciled with either the English decisions of R v Bow Street
Magistrates, Ex parte Mackeson [89] and Bennett v Horseferry Road Magistrates’
Court, [90] or the spirit of Mohamed, should not be followed. [91] This applies particularly
where the police of the receiving state are themselves complicit in the deportation.
Different considerations may, however, apply when the authorities of the receiving
state are in no way involved in the deportation. [92]

2 Abduction

One state may not exercise its police powers in the territory of another state.
Consequently, the abduction or kidnapping of a person from state ‘A’ by agents of state
‘B’ to stand trial in state B is a clear violation of the territorial sovereignty of state
A. [93] In such a case, the injured state is entitled to demand the return of the abducted
person and may—if there is an extradition agreement—request the extradition of the
abductors to stand trial on a charge of kidnapping. The abducted person may,
moreover, institute civil proceedings for
Page 319
damages against his abductors. [94] The most effective way to deter territorial violations
of this kind is probably for the courts of the abducting state to refuse to exercise
jurisdiction over the abductee as this would remove the incentive for the
abduction. [95] This, however, has not been the approach of the courts of most
countries. [96]
South African courts, relying on Anglo-American authority [97]—and Israel’s exercise
of criminal jurisdiction over Adolf Eichmann after his kidnapping from Argentina [98]—
repeatedly held that they had jurisdiction to try abducted political opponents of the
government during the apartheid era. [99] However, in 1991, in a matter that
commenced [100] before the abandonment of apartheid in February 1990, the Appellate
Division held in S v Ebrahim [101] that under Roman-Dutch law, a South African court
has no competence to try a person
Page 320
abducted from another state by agents of the prosecuting state. [102] Consequently, it
set aside the conviction and sentence of 20 years’ imprisonment imposed on Ebrahim,
an ANC operative kidnapped from Swaziland by agents of the South African state and
brought to trial before the Transvaal Provincial Division on a charge of treason. The
earlier South African decisions were distinguished on the facts [103] or rejected on the
ground that they were premised on English law and took no account of Roman-Dutch
law. Although the Court made it clear that its decision was based on Roman-Dutch law
and not international law, [104] it stated that the rule prohibiting the exercise of
jurisdiction over an abducted person was premised on considerations such as the
promotion of human rights, good inter-state relations and respect for territorial
sovereignty. [105]
Shortly afterwards, the Zimbabwe Supreme Court gave its approval to S v
Ebrahim in S v Beahan. [106] Here, however, Gubbay CJ placed greater emphasis on
international law in holding that:
There is an inherent objection to [exercising jurisdiction over an abductee] both on grounds of
public policy pertaining to international ethical norms and because it imperils and corrodes the
peaceful coexistence and mutual respect of sovereign nations. For abduction is illegal under
international law, provided the abductor was not acting on his own initiative and without the
authority or connivance of his government. A contrary view would amount to a declaration that
 the end justifies the means, thereby encouraging states to become law-breakers in order to
secure the conviction of a private individual. [107]
Considerations of this kind were ignored by the Supreme Court of the United States in
its much-criticised decision in United States v Alvarez-Machain, [108] in which it held that
the forcible abduction of a Mexican national from Mexico by US law enforcement agents
did not serve as a bar to his trial in the United States. In a dissenting opinion, Justice
Stevens invoked S v Ebrahim for his conclusion that ‘most courts throughout the
civilized world . . . will be deeply disturbed by the “monstrous” decision’ of the
majority. [109]
S v Ebrahim has been followed by the House of Lords in a case of
Page 321
South African origin containing elements of disguised extradition and abduction.
In Bennett v Horseferry Road Magistrates’ Court, [110] a New Zealand national wanted by
the English police on charges of fraud committed in the United Kingdom was arrested in
South Africa and forcibly returned to the United Kingdom by the South African police,
acting in collusion with the English police, under the pretext of deporting him to New
Zealand via the United Kingdom. This scheme was devised by the police forces of the
two countries as a result of the absence of an extradition agreement between South
Africa and the United Kingdom. After examining the judicial precedents on disguised
extradition and abduction, including S v Ebrahim, the House of Lords held that it would
decline to exercise jurisdiction over Bennett as the manner in which his presence had
been secured amounted to an abuse of the process of the Court. The Court stressed the
importance of refusing to exercise jurisdiction in such a case in order to discourage
practices of abduction and unlawful extradition. Lord Griffiths stated:
Extradition procedures are designed not only to ensure that criminals are returned from one
country to another but also to protect the rights of those who are accused of crimes by the
requesting country. Thus sufficient evidence has to be produced to show a prima facie case
against the accused and the rule of speciality protects the accused from being tried for any
crime other than that for which he was extradited. If a practice developed in which the police
or prosecuting authorities of this country ignored extradition procedures and secured the
return of an accused by a mere request to police colleagues in another country they would be
flouting the extradition procedures and depriving the accused of the safeguards built into the
extradition process for his benefit. It is to my mind unthinkable that in such circumstances the
court should declare itself to be powerless and stand idly by. [111]
In approving this approach, Lord Bridge declared:
There is, I think, no principle more basic to any proper system of law than the maintenance of
the rule of law itself. When it is shown that the law enforcement agency responsible for
bringing a prosecution has only been enabled to do so by participating in violations of
international law and of the laws of another state in order to secure the presence of the
accused within the territorial jurisdiction of the court, I think that respect for the rule of law
demands that the court take cognisance of that circumstance. To hold that the court may turn
a blind eye to executive lawlessness beyond the frontiers of its own jurisdiction is, to my mind,
an insular and unacceptable view. Having then taken cognisance of the lawlessness it would
again appear to me to be a wholly inadequate response for the court to hold that the only
remedy lies in
Page 322
civil proceedings at the suit of the defendant or in disciplinary or criminal proceedings against
the individual officers of the law enforcement agency who were concerned in the illegal action
taken. Since the prosecution could never have been brought if the defendant had not been
illegally abducted, the whole proceeding is tainted. [112]
The philosophy expounded in S v Ebrahim was applied to relations between South
Africa, Ciskei and Bophuthatswana, which remained nominally independent under South
African law until 1994. In Minister of Law and Order v Thandani, [113] S v
Wellem [114] and S v Buys, [115] the failure of the police forces of these territories to
respect each others’ borders, and to follow proper extradition procedures for the
rendition of fugitives from justice, was condemned and the exercise of jurisdiction
refused. In a number of other extraordinary decisions, however, both the lower
courts [116] and the Appellate Division [117] condoned cross-border arrests, police
deception and the refusal of the respective police forces of these territories to follow
extradition procedures.
The failure of courts, including the Appellate Division in S v Mahala [118] and S v
December, [119] to accept that the ‘collection’ or ‘apprehension’ (if not formal ‘arrest’) of
a fugitive by the South African police in the territory of Ciskei or Bophuthatswana had
been a violation of another state’s territorial sovereignty—and therefore contrary to the
fundamental jurisdictional rules proclaimed in the Lotus Case [120]—can only be
explained on the grounds that these courts did not actually believe that Ciskei and
Bophuthatswana were genuine sovereign independent states. While it is possible to
sympathise with this judicial cynicism about the ‘independence’ of the homeland states
or Bantustans, particularly as the political decision to re-incorporate them into South
Africa had already been made at this time, it is unfortunate that the judges did not
fashion a rule premised on the limited sovereignty of Ciskei and Bophuthatswana.
Instead, we have a number of unreasoned and uninformed decisions,
Page 323
which take no account of the basic principles of international law that seek to limit the
law and philosophy of S v Ebrahim. [121]
The only solution is for the courts of post-apartheid South Africa to repudiate these
decisions as judicial aberrations applicable only to the Bantustan states and not to
relations between South Africa and foreign states.

3 Surrender to international criminal tribunals

Extradition is to be distinguished from surrender to international criminal tribunals. The

latter entails the surrender by a state of a person sought by an international criminal
tribunal to that tribunal. There is no requesting state, but a request by an international
criminal tribunal.
Under South African law, surrender to the International Criminal Court is regulated
by the Implementation of the Rome Statute of the International Criminal Court Act 27
of 2002. This is discussed fully in Chapter 10.

Extradition from South Africa to foreign states

The Extradition Act regulates the process for extradition from South Africa to another
state. The Act prescribes different processes for extradition depending on whether the
requesting state is a foreign state or an associated state.
The Act defines a foreign state as including any foreign territory. [122] An associated
state is then defined in s 6 of the Act. [123] There are two requirements for a foreign
state to be an associated state. First, the foreign state must be in Africa. Secondly,
there must be an extradition treaty between the Republic and that African state that
provides for the endorsement for warrants of arrest on a reciprocal basis. We first deal
below with the process for a foreign state, while the next section of this chapter deals
with the process for requests by an associated state.
The process of extraditing a person from South Africa to a foreign state can be
divided into three phases: the administrative phase, the judicial phase and the
executive phase. [124]
Page 324

1 The administrative phase

A foreign state requesting South Africa to extradite an alleged criminal to it must make
its request to the Minister of Justice through diplomatic channels. [125] The Act does not
prescribe any form or content for the request, and the Minister is not expected to
examine the merits of the request at this stage. [126]
In S v Thornhill [127] the Western Cape High Court accepted that the existence of
such a request is a jurisdictional fact that must be satisfied before a person can be
committed pending surrender. [128] Without a request in terms of s 4, a person is not
liable to be surrendered. But a magistrate, when conducting an inquiry under s 10(1), is
allowed to presume that such a request has been made unless there is evidence
suggesting otherwise. [129] On the facts of Thornhill, the Minister had countersigned the
President’s consent to surrender the sought person. [130] The Minister thus created the
impression that he had duly received a request in terms of s 4, and there was nothing
to indicate otherwise. The High Court therefore held that the magistrate was justified in
finding Mr Thornhill liable for surrender even though the extradition request sent to the
Minister via diplomatic channels was not before the magistrate.
If there is an extradition treaty between the requesting foreign state and the
Republic, then after receiving a request the Minister has the power to notify a
magistrate of such a request. [131] This notification empowers the magistrate to issue a
warrant for the arrest of the sought person, regardless of the sought person’s
whereabouts. [132]
If there is no extradition treaty between the requesting foreign state and the
Republic, then the Minister must first forward the request to the President. The
President must consent to the surrender of the person. Only thereafter can the Minister
notify the magistrate of the request—empowering the latter to issue a warrant of
arrest. [133]
Page 325
The issuing of the notice to the magistrate ‘sets in motion’ the domestic process of
extradition. [134] It is not a decision to extradite the sought person. The Minister’s
decision under s 11 to surrender the sought person, and his or her consideration of the
Act’s requirements for extradition, are done only after a committal order by a
magistrate. [135] To issue the notice, the Minister only needs to have received the
request through diplomatic channels by a person recognised by the Minister as a
diplomatic or consular representative. [136] Her decision to issue the notice therefore is
not reviewable if she does not consider the merits of the extradition request before
issuing such a notice. [137]
In certain circumstances the magistrate may issue a warrant for the arrest or further
detention of a person sought to be extradited without notification from the
Minister. [138] For this to occur there must be sufficient information before the
magistrate to demonstrate that the sought person was accused or convicted of an
extraditable offence committed within the jurisdiction of the foreign state. [139] This
evidence will be sufficient if it would justify the issuing of the warrant as if the sought
person committed an offence in the Republic. If the magistrate issues a warrant without
the Minister’s notification, then the magistrate must notify the Minister
immediately. [140]
Certain extradition treaties provide for the provisional arrest of a sought person. For
example, article 9 of the treaty between South Africa and the United States provides
that ‘in case of urgency, the Requesting State may, for the purpose of extradition,
request the provisional arrest of the person sought pending presentation of the
documents in support of the extradition request’. [141]
Once the magistrate issues an arrest warrant, the warrant is executed according to
the ordinary principles of criminal procedure. If an arrest warrant is issued or executed
unlawfully (for example, if the applying officer commits a material non-disclosure) then
the subsequent steps in the extradition process (including the hearing) could be
unlawful. [142]
Page 326

2 The judicial phase

An arrested person must be brought before a magistrate as soon as possible for the
purpose of an enquiry into the extradition of the person. [143] This enquiry is modelled
upon a preparatory examination. [144] But the enquiry is not a criminal
trial, [145] although the proceedings resemble criminal proceedings in many
respects. [146] Importantly, the magistrate can admit the arrested person to bail pending
the finalisation of the enquiry. [147]
Section 9(3) regulates the evidence admissible in the enquiry before the magistrate.
It prescribes various requirements pertaining to authentication of evidence from the
foreign state. These requirements are aimed at ensuring that the relevant evidence
submitted by the foreign state ‘is what it purports to be’. [148] These authentication
requirements apply to any foreign deposition, any statement on oath or affirmation,
any record of any conviction, and any warrant issued. They do not apply, for example,
to foreign statutes that may be included in the extradition request bundle. [149] At least
one of these requirements must be satisfied to authenticate a document. [150]
Section 10 regulates enquiries for extraditions to foreign states. [151] Section 10
empowers the magistrate to commit a sought person to detention pending the
Minister’s decision to surrender the sought person. That is, the magistrate does not
consider whether to extradite or surrender the person before him or her. [152] The
decision is only to commit or not to commit.
Page 327
Section 10(1) explains that a magistrate, on the consideration of the evidence before
her or him, must be satisfied that two conditions are both fulfilled before a committal
order can be made. First, the person must be liable to be surrendered to the foreign
state concerned. Secondly, in the case where such person is accused of an offence,
there must be sufficient evidence to warrant a prosecution for the offence in the foreign
state. [153]
The magistrate does not consider, for example, whether it is unjust or unreasonable
to surrender the applicant. This is because the magistrate is not considering surrender
to the foreign state but committal pending the final phase of the extradition. This is a
narrow and specific issue that does not involve deliberation on human rights issues (for
instance). In Chanthunya v S, [154] the North West High Court held that the question of
whether an assurance not to implement the death penalty had been secured is not a
matter to be determined by the magistrate under a s 10 enquiry.
If the magistrate finds that both requirements in s 10(1) are satisfied and issues an
order of committal, then the sought person has a right of appeal or review to the
provincial or local division of the High Court having jurisdiction before he is
surrendered. [155] The sought person can apply for bail pending this appeal. [156] Such a
bail application pending appeal is governed by s 65 of the Criminal Procedure Act, and
so an appeal against a magistrate’s refusal to grant bail can be heard by a single
judge. [157]
It is convenient to deal with the second s 10(1) requirement first. For the purpose of
satisfying herself or himself that there is sufficient evidence to warrant a prosecution in
the foreign state, the magistrate must accept as conclusive proof a certificate issued by
the appropriate authority in the requesting state, stating that it has sufficient evidence
at its disposal to warrant the prosecution of the person concerned. [158] This test,
introduced in 1996, represents an abandonment of the Anglo-American common-law
requirement that the requesting state make out a prima facie case of guilt against the
fugitive before extradition is granted. This was done to overcome difficulties
experienced by civil-law systems which do not employ the prima facie test and have
difficulty in satisfying the requirements inherent in the common-law test. In this
Page 328
respect, South Africa has followed the example of other common-law countries, such as
Britain and Australia. [159] As for the first requirement, liability for extradition is
regulated in s 3 of the Act. There are three kinds of persons who are liable to be
extradited:
(1)
Persons who are convicted or accused of an extraditable offence contained in an
extradition agreement committed within the jurisdiction of a foreign state. The
offence could have been committed before or after the commencement of the
Extradition Act or the extradition treaty. [160]
(2)
Persons who are accused or convicted of an extraditable offence committed
within the jurisdiction of a foreign state that is not party to an extradition treaty
with South Africa and the President has consented in writing to their
surrender. [161] The power of the President is not to determine whether the
person should be extradited but whether the extradition process should be
applied to the specific request. [162] The written consent contemplated in s 3(2) of
the Act takes the place of a duly concluded extradition treaty with the requesting
foreign state. [163] A High Court has held that the President’s decision under s
3(2) can be reviewed on the ground of rationality. [164] This means the President
cannot consent to an extradition while knowing that the extradition request is
spurious and in bad faith. [165] But this review should be launched after the
decision is taken. [166]
(3)
Persons who are accused or convicted of an extraditable offence committed
within the jurisdiction of a designated state. The designation of the state, as per
 s 2(1)(b), can occur before or after the offence was committed. [167] South Africa
has designated Ireland, Zimbabwe, Namibia and the United Kingdom. [168]
Page 329
All persons liable to be extradited must therefore be (a) accused or convicted of an
extraditable offence, and (b) committed within the jurisdiction of the foreign state.
These requirements are considered in turn.

(a) Extraditable offences and double criminality

The principle of double criminality requires that the conduct claimed to constitute an
extraditable crime should constitute a crime in both the requesting and the requested
state. It is not necessary that the offence should have the same name in both states,
provided that it is substantially similar. [169]
In the past it was common practice to list the offences in respect of which extradition
is to apply in the extradition treaty. Today the tendency is for parties to provide for
extradition in respect of crimes that are punishable in both the requesting state and the
requested state with a sentence above a particular severity without naming the crime.
For instance, the agreement between South Africa and the People’s Republic of
China [170] provides:
For the purpose of this Treaty, extradition shall be granted for conduct which constitutes an
offence under the laws of both Contracting States that is punishable by imprisonment for a
period of at least one year or by a more severe penalty.
The Extradition Act, as amended in 1996, approves this approach by providing:
‘[E]xtraditable offence’ means any offence which in terms of the law of the Republic and of the
foreign state concerned is punishable by a sentence of imprisonment or other form of
deprivation of liberty for a period of six months or more. [171]
The Extradition Act is silent on the question whether the crime in respect of which
extradition is requested must be a crime in South Africa at the time of the extradition
request or at the time the alleged
Page 330
offence was committed. The critical date is that of the extradition request and there is
no reason why the term ‘extraditable offence’ (above) should not be so interpreted.
Section 3, which extends the Act to offences committed before the Act or extradition
agreement came into operation, supports this interpretation. A South African High
Court [172] has, however, followed the interpretation of ‘extradition crime’ applied by the
House of Lords in the Pinochet Case, [173] in which it held that the former Chilean
dictator could not be extradited to Spain for acts of torture committed in Chile before
the United Kingdom enacted the 1984 Torture Convention into municipal law by reason
of the principle of double criminality, which it interpreted—in terms of its own statute—
to apply only to offences committed abroad that were punishable as crimes in the
United Kingdom at the time of their commission.
In Patel v National Director of Public Prosecutions, [174] the Supreme Court of Appeal
departed from the approach in Pinochet and settled the question of when the offence
underlying the extradition needs to be a crime under South African law. In Patel the
offences with which the sought person had been charged in the United States were
allegedly committed between 2005 and 2007. That conduct was not criminal in South
Africa until 2010, when s 28 of the Financial Intelligence Centre Act 38 of 2001 came
into operation. [175] The SCA held that the Act ‘unquestionably refers to the present: is
the offence an extraditable offence, ie now, at the date of the request for extradition?’
Because the sought person was accused of conduct that was criminal at the time of the
request by the United States, the requirement of double criminality was satisfied. The
Supreme Court of Appeal distinguished the approach in Pinochet because the House of
Lords based its conclusion on an interpretation of the specific provisions of the UK
Extradition Acts that are not mirrored in the South African Extradition Act. In doing so,
the Court endorsed the view of one of the authors of this chapter who had stressed that
an approach in terms of which the conduct date is decisive for the purpose of double
criminality would undermine mutual co-operation between the states and negate the
very purpose of a bilateral extradition treaty: the parties enter into reciprocal rights and
duties
Page 331
in order to bring to justice those who have committed serious crimes. [176]

(b) Within the jurisdiction of the foreign state

In Chapter 9, the various bases on which states can exercise extraterritorial jurisdiction
were discussed. The issue is if the Extradition Act extends liability for extradition to
instances when the foreign state exercises extraterritorial jurisdiction. For example, if a
Belgian national commits a crime in Venezuela, escapes to South Africa, and then is
accused by Belgium on the basis of active personality for the crime committed abroad,
is that person accused of an offence ‘committed within the jurisdiction’ of Belgium?
The Extradition Act is not entirely clear. The starting point will often be South Africa’s
international legal obligations. [177] Many extradition treaties expressly prevent a
requested state from refusing to extradite only because the requesting state exercises
extraterritorial jurisdiction. [178] In those instances, the question is whether the
Extradition Act can reasonably be interpreted to give effect to that international
obligation to extradite in response to extraterritorial exercises of jurisdiction. [179]
There is nothing in the words ‘within the jurisdiction’ and the structure of the Act to
prevent such an extraterritorial interpretation. On the contrary, given the increasingly
transnational nature of crimes, and South Africa’s exercise of extraterritorial jurisdiction
in respect of various crimes, [180] the Extradition Act can reasonably be
Page 332
interpreted to allow for extraditions in response to extraterritorial exercises of
jurisdiction.
The question of extraterritoriality in s 3(1) was touched upon in the High Court by
Gamble J and Donen AJ in Carolissen v Director of Public Prosecutions. [181] Mr
Carolissen was indicted by a grand jury in Maine in the United States for the
manufacturing and dissemination of child pornography. At all relevant times, Mr
Carolissen was based in Cape Town, South Africa. Yet he had emailed undercover US
police officers images and videos of child pornography. The police officers received this
pornography while in Maine. They subsequently caused Mr Carolissen’s indictment, and
the US requested his extradition to the US. Mr Carolissen was arrested and a
magistrate ordered his committal pending the Minister’s decision to surrender him in
terms of s 11. Mr Carolissen appealed this decision.
In the judgment, the High Court found that the United States had jurisdiction over
Mr Carolissen. But it required the United States to ‘show the necessary nexus to its
territorial jurisdiction’ before extradition could happen. [182] On the facts, ‘territorial
jurisdiction [was] contained in the assertion [that the agent] was present in Maine
when he “chatted” with [Carolissen] and subsequently received the pornographic
images which had been manufactured (or possibly sourced elsewhere) by the
appellant’. [183] The Court also held that the grand jury ‘was satisfied as to territorial
jurisdiction’ of the United States. [184] It concluded that ‘an applicant for extradition
must make out a prima facie case for its territorial jurisdiction, regardless as to whether
the offence relates to cybercrime or otherwise’. [185]
The Court thus, at times, seemed to accept that extradition can only occur if
territorial jurisdiction is established. But at other times, the Court made comments to
the contrary. It noted:
[I]t is established law that it is open to a sovereign state to enact legislation permitting it to
prosecute within its own jurisdiction suspects who have committed crimes elsewhere in the
world, where those crimes might ultimately have a deleterious effect in the territorial
jurisdiction of the requesting state. [186]
The Court cited various UK and Canadian authorities for the extraterritorial liability for
extradition. [187] Citing Geuking v President of the RSA, the Court also emphasised the
importance of effective
Page 333
extradition for transnational crimes. [188] The Court further made reference to the
extraterritorial application of various South African statutes. [189]
It seems then that the Court did not answer the question of extraterritorial
jurisdiction generally. It made obiter comments regarding extraterritorial jurisdiction
and the effectiveness of such jurisdiction for transnational crimes (such as cybercrime).
But its ratio is limited to the facts before it. Namely, there was a territorial link to the
United States because the child pornography had been sent there. That said, the
Court’s obiter recognition of extraterritorial jurisdiction suggests that s 3(1) would be
interpreted extraterritorially if no territorial link had been established.
Recently, the Constitutional Court took a broad interpretation to the
extraterritoriality of the Protection of Constitutional Democracy against Terrorist and
Related Activities Act 33 of 2004. [190] The Constitutional Court made the following
apposite finding at para 43 when interpreting the jurisdiction of the Act:
While it is true that territoriality has been the traditional basis on which courts establish
jurisdiction, international and South African jurisprudence recognise other methods of asserting
jurisdiction . . . This is particularly true when the crimes over which a court asserts jurisdiction
have an international dimension. We should not, through a narrow interpretation of section
15(1), mistakenly perpetuate an historical disinclination to extra-territoriality.
These comments suggest that courts will interpret ‘within the jurisdiction of the foreign
state’ to include the recognised bases of extraterritorial jurisdiction.

3 The executive phase

When a magistrate issues a committal order, the person is imprisoned pending the
Minister’s decision to surrender them under s 11. The magistrate cannot grant bail at
this stage, although the sought person can apply to the High Court to be admitted to
bail pending the s 11 decision. [191]
Under s 11, the Minister has two options. She can decide to surrender the sought
person. Or she can refuse to surrender the sought person for various reasons and on
certain conditions. [192] These reasons include considerations of fairness, justice and
reasonableness. It is at this stage of the extradition process that the merits of the
Page 334
extradition, including human rights concerns, must be considered by the Minister. [193]

The Minister’s decision is subject to review. [194]

The Constitutional Court has recently reiterated that if a person is a refugee under
South African law, that person cannot be extradited if he or she would be persecuted,
lose freedom or be physically harmed as a result of such extradition. [195] Conversely, if
South Africa is obliged to extradite someone before he or she is given refugee status,
South Africa cannot flout the duty to extradite and grant the sought person refugee
status. [196]
The Minister can decide to surrender a citizen of South Africa. Civil-law countries,
which exercise personal jurisdiction over their nationals for offences committed abroad,
favour the exemption of their own nationals from extradition. [197] Common-law
countries, which in most circumstances do not exercise extraterritorial jurisdiction over
their nationals, adopt a different approach and allow the extradition of their
nationals. [198] These divergent attitudes undermine the important principle of
reciprocity. The compromise is for a treaty to include a clause that gives either state
discretion to refuse to extradite its own nationals. [199] This allows civil-law countries to
refuse extradition of their nationals and to try such nationals themselves, while at the
same time permitting common-law countries to extradite their nationals for offences
committed abroad beyond their criminal jurisdiction.
Page 335
South Africa adheres to the common-law tradition. The Extradition Act of 1962
contains no exemption for nationals. South Africa’s attitude towards nationality
(citizenship) and extradition was examined by the Constitutional Court in Geuking v
President of the RSA, [200] in which it was argued that the President, in exercising his
power to surrender a person to the Federal Republic of Germany, under s 3(2) of the
Extradition Act, had failed to have regard to the fact that the person was a South
African national (citizen). The Constitutional Court held:
In the present case, the President stated in the affidavit he filed in the High Court that in
deciding whether to grant his consent under s 3(2) of the Act the citizenship of the appellant
would not have been a relevant consideration. I can find no constitutional ground for attacking
that policy decision. Unlike the [Federal Republic of Germany] and many other civil law
jurisdictions, South Africa does not ordinarily prosecute its citizens for crimes committed
beyond its borders. Criminal conduct would go unpunished if South African citizens were not
extradited to face prosecution in the country where the crime was committed. The President is
therefore entitled to adopt a policy that it is in the interests of the Republic to consent to a
request for extradition proceedings against a person, regardless of his or her citizenship. [201]
A crucial factor that should prevent the Minister from ordering surrender is non bis in
idem. A person may not be extradited in respect of an offence for which he has already
been acquitted, or convicted, by the requested state. This principle, which confirms the
principle of autrefois acquit (or autrefois convict), is not expressly included in the
Extradition Act but appears in most extradition agreements. [202]
If the Minister does decide to surrender the sought person, then that surrender is
regulated by the principle of speciality. According to the principle of speciality, an
extradited person may not be tried for an offence other than that for which he was
extradited, unless the extraditing state consents to such a prosecution. This principle is
confirmed by the Extradition Act [203] and is a common clause in extradition agreements.
In terms of the Act, a person may be tried for an offence other than that for which he
was extradited if the offence is
Page 336
one for which he might lawfully have been convicted on a charge of the offence for
which extradition was sought. [204] The Supreme Court of Appeal held in S v
Stokes [205] that the word ‘sought’ in s 19 of the Act should be read as ‘successfully
sought’ and the principle of speciality thus respected.
Finally, offences of a political character are of relevance at this stage of the
extradition.

(a) Offences of a political character [206]

Extradition law and practice exempt the political offender from extradition. This rule had
its origins in the 19th century, when the governments of the new liberal democracies
refused to return political dissidents to the despotic states of the ancien régime. The
principal justifications advanced for the rule are, first, that states should not intervene
in the internal political conflicts of other states by assisting in the rendition of political
opponents of the government, and, secondly, that political offenders, unlike ordinary
criminals, threaten only the criminal justice system of the state from which they have
fled and not that of the state granting asylum. Over the years, the romantic image of
the political dissident fighting for democracy has been tarnished by the political terrorist
fanatically determined to overthrow the regime of his home state by all means,
including hostage-taking and hijacking. As a result, the political offence exception has
become highly controversial and courts have sought to define the political offence in
such a way that it excludes the political terrorist but does not abandon the protection of
the genuine political dissident.
South African extradition law recognises the political offence exception. Section 15 of
the Extradition Act empowers the Minister of Justice to intervene at any stage during
extradition proceedings in order to release a fugitive if he is satisfied that the offence in
respect of which extradition is sought is ‘an offence of a political character’. Moreover,
extradition agreements to which South Africa is a party generally exempt the political
offender from extradition. The 1968 agreement with Swaziland, [207] for example,
provides in article 3:
A person claimed shall not be extradited if the offence for which his extradition is requested is
regarded by the requested Party as one of a political character, or if he satisfies the requested
Party that the request for his
Page 337
extradition has in fact been made with a view to try or punish him for an offence of a political
character.
Once an extradition agreement is incorporated into municipal law in accordance with s 2
of the Extradition Act, it becomes part of the law of the land, with the result that a
municipal court will be required to consider whether the offence is one of a ‘political
character’. [208]
Courts throughout the world have experienced great difficulty in deciding when an
offence is one of a political character. Clearly treason and sedition are political offences.
Problems arise in the case of ordinary crimes, such as murder or robbery, when they
are politically motivated. A South African extradition court has not yet been called upon
to examine this problem in any detail, but when the occasion arises there is little doubt
that guidance will be sought in English law [209] and in the guidelines on ‘political
offences’, laid down in the Promotion of National Unity and Reconciliation Act [210] in
order to determine who would qualify for amnesty following the abandonment of
apartheid.
English law and the political offence
English decisions have refused to lay down an exclusive definition of ‘political offence’.
In Schtraks v Government of Israel, Lord Radcliffe said: ‘[N]o definition has yet
emerged or by now is ever likely to. Indeed it has come to be regarded as something of
an advantage that there is to be no definition.’ [211] The decisions do, however, lay down
a number of tests or guidelines.
In re Castioni [212] involved a political uprising in Switzerland against the government
of one of the cantons in the course of which Castioni killed a member of the
government. He then fled to England. His extradition was refused on the ground that
the offence was incidental to and formed part of a political uprising and was therefore
an offence of a political character. This ‘incidence test’ has been approved by courts in
the United States. In a decision involving the extradition of a member of the Irish
Republican Army to the United Kingdom, the test was formulated in the following
language:
First there must be an uprising, a political disturbance related to the struggle of individuals to
alter or abolish the existing government in their country . . . Second, the charged offence must
have been committed in furtherance of the
Page 338
uprising; it must be related to the political struggle or be consequent to the uprising
activity. [213]
This test, premised on the presence of a political disturbance or uprising, is too
restrictive. In R v Governor of Brixton Prison, Ex p Kolczynski and Others, [214] the Court
dispensed with the need for a political disturbance where a number of Polish seamen
had hijacked a Polish trawler and assaulted a political commissar in order to effect an
escape from the oppressive Polish regime. In refusing extradition, Lord Goddard
declared that ‘reasons of humanity’ compelled ‘a wider and more generous meaning’ of
the term ‘political offence’—one that went beyond the limited incidence test. A ‘wider
and more generous’ meaning was also given to the term in Schtraks v Government of
Israel. [215] In this case, Lord Reid stated that political offences were not limited ‘to
cases of open insurrection’ or even to cases of attempts to overthrow a government:
The use of force, or it may be other means, to compel a sovereign to change his advisers, or to
compel a government to change its policy may be just as political in character as the use of
force to achieve a revolution. And I do not see why it should be necessary that the refugee’s
party should have been trying to achieve power in the state. It would be enough if they were
trying to make the government concede some measure of freedom but not attempting to
supplant it. [216]
In the same case, Lord Radcliffe declared:
In my opinion the idea that lies behind the phrase ‘offence of a political character’ is that the
fugitive is at odds with the state that applies for his extradition on some issue connected with
the political control or government of the country. [217]
English courts have stressed that not every politically motivated crime will qualify as a
political offence. In Cheng v Governor of Pentonville Prison, [218] Lord Diplock stated:
But if the accused had killed a dictator in the hope of changing the government of the country,
his object would be sufficiently immediate to justify the epithet ‘political’. For politics are about
government. ‘Political’ as descriptive of an object to be achieved must, in my view, be confined
to the object of overthrowing or changing the government of a state or inducing it to change
its policy or escaping from its territory the better so to do.
Page 339
Although T v Secretary of State for the Home Department [219] concerned an application
for asylum—by a member of the Front Islamique du Salut (FIS) responsible for placing
a bomb at Algiers Airport which killed ten people—the House of Lords made a thorough
examination of the political offence exception in extradition. The House of Lords held
that the applicant failed to qualify for asylum in terms of article 1F(b) of the Convention
Relating to the Status of Refugees of 1951, which denies the granting of asylum to a
person who has committed a ‘serious non-political crime outside the country of refuge’.
Lord Lloyd, for the majority, defined a ‘political crime’ in language that by necessary
implication excluded terrorism as a political offence. He stated:
A crime is a political crime for the purposes of article 1F(b) of the 1951 convention if, and only
if: (1) it is committed for a political purpose, that is to say, with the object of overthrowing or
subverting or changing the government of a state or inducing it to change its policy; and (2)
there is a sufficiently close and direct link between the crime and the alleged political purpose.
In determining whether such a link exists, the court will bear in mind the means used to
achieve the political end, and will have particular regard to whether the crime was aimed at a
military or governmental target, on the one hand, or a civilian target on the other, and in
either event whether it was likely to involve the indiscriminate killing or injuring of members of
the public. [220]

Amnesty and the political offence

The Promotion of National Unity and Reconciliation Act [221] established amnesty
committees which were empowered to grant amnesty to individuals who sought
amnesty for acts or ‘an act associated with a political objective committed in the course
of the conflicts of the past’ and who made full disclosure of all relevant facts. [222] The
criteria employed for deciding whether the act was one ‘associated with a political
objective’ were drawn from the principles used in extradition law for deciding whether
the offence in respect of which extradition was sought was a political offence. The
criteria included, inter alia, the motive of the offender; the context in which the act took
place and, in particular, whether it was committed ‘in the course of or as part of a
political uprising, disturbance or event’; the gravity of the act; the objective of the act
and, in particular, whether it was ‘primarily directed at a political opponent or state
property or personnel or against private property or individuals’; whether the act was
committed in execution of an order by an organisation; and the
Page 340
relationship between the act and the political objective pursued, and ‘in particular the
directness and proximity of the relationship and the proportionality of the act to the
objective pursued’. [223]
This statutory ‘codification’ of the political offence, and the interpretation of this
provision by courts and amnesty committees, may serve as a useful guide to South
African courts when they confront the political offence as a defence in extradition
proceedings.

International terrorism and the political offence

International terrorism presents a particular problem [224] for extradition, as most
transnational acts of terror are politically motivated and fall within the tests laid down
for the political offender. There is, however, an important difference between
Castioni [225] and the hijacker of an international flight. Castioni’s offence threatened the
legal order of one state only—Switzerland. Hijacking, on the other hand, threatens the
international order and demands an international co-operative response. At present,
this response takes the form of extradition or trial by the state with physical control
over the offender (aut dedere aut judicare) [226] which would be defeated if the political
offence exception were to be allowed. Consequently, there is a measure of consensus
that the political offence exception should not apply to the extradition of international
criminals.
Unfortunately, early multilateral treaties creating international crimes do not
expressly exclude the political offence exception. The Hague Convention of 1970 on
aerial hijacking, [227] for example, obliges the state with physical control over the
hijacker to extradite him, or to submit his case to its competent authorities for the
purpose of prosecution in the same manner that it would consider the prosecution ‘of
any ordinary offence of a serious nature’ under its own law. This does not prevent such
a state from either deciding that the offence is one of a political nature [228] or pursuing
a ‘friendly prosecution’ of the
Page 341
offender. [229] Most other conventions of this kind follow the precedent of the Hague
Convention. [230]
The European Convention on the Suppression of Terrorism of 1977 [231] adopts a
different approach in expressly providing that acts of international terror, including
hijacking and hostage-taking, shall not be treated as political offences for the purpose
of extradition. This is confirmed by the European Union Extradition Agreement of
1996. [232] The International Convention for the Suppression of Terrorist Bombings of
1998, and the International Convention for the Suppression of the Financing of
Terrorism of 1999, unlike other treaties criminalising acts of international terror,
expressly exclude the political offence as an obstacle to extradition. [233] Finally, it is
today not uncommon for an extradition agreement to exclude from the ambit of the
political offence acts declared to constitute an offence under a multilateral treaty which
imposes an obligation to prosecute or extradite. [234]
In 2004 the South African Extradition Act of 1962 was amended by the Protection of
Constitutional Democracy against Terrorist and Related Activities Act 33 of 2004 to
exclude the political offence defence to extradition when a person is charged with
‘terrorist activity’. [235]
Page 342

Extradition from South Africa to associated states

In order to facilitate extradition between South Africa and its neighbours in Africa, a
more expeditious procedure is followed. Here the extradition agreement provides that
the requesting state is not required to request extradition through diplomatic channels
and may instead submit a warrant for the arrest of the fugitive to the prosecutor having
jurisdiction, together with a statement providing details of the offence and prima
facie evidence of the fugitive’s guilt. [236] A magistrate may then simply endorse the
warrant for execution in South Africa. [237]
The enquiry before the magistrate then combines the judicial and executive phases
discussed above. The magistrate holds an enquiry into questions such as whether the
request complies with the terms of the extradition agreement; whether the fugitive will
be prosecuted or prejudiced in the associated state by reason of his ‘gender, race,
religion, nationality or political opinion’; or whether it would in all the circumstances of
the case be unjust or unreasonable to extradite the fugitive. [238] In such a case, the
magistrate is not required to find that there would be sufficient reason for putting the
fugitive on trial had the offence been committed in South Africa. [239] After this enquiry,
the magistrate may order the extradition of the fugitive. The Minister is not included in
this process. The magistrate decides to extradite—not the Minister. The fugitive has a
right of appeal to the provincial or local division of the High Court having jurisdiction
before he is surrendered. [240]

International co-operation in the suppression of crime by

means other than extradition

The internationalisation of crime has made national law enforcement authorities

increasingly dependent on international co-operation. Extradition is the oldest and best-
known form of such co-operation. However, there is a growing need for international
co-operation on the part of both national police forces and judicial authorities in
Page 343
respect of the collection of evidence, the tracing of the proceeds of crime, money-
laundering and the enforcement of judicial orders in criminal proceedings. [241] This has
resulted in states entering into both bilateral and multilateral mutual assistance
treaties. In 1990 the United Nations gave its approval to a UN Model Treaty on Mutual
Assistance in Criminal Matters; [242] there is a Commonwealth Scheme Relating to
Mutual Assistance in Criminal Matters; [243] and the 1988 Vienna Convention against
Illicit Traffic in Narcotic Drugs and Psychotropic Substances [244] contains a
comprehensive set of rules on mutual assistance. Assistance of this kind includes taking
evidence or statements from persons, effecting service of judicial documents, executing
searches and seizures, tracing the proceeds and instrumentalities of crime, and making
persons available to give evidence or assist in investigations.
During the isolation of South Africa under apartheid, there was little incentive to
make South African law accord with the requirements relating to mutual assistance in
criminal proceedings contained in bilateral treaties or multilateral treaties. Changed
circumstances, however, prompted the enactment of the International Co-operation in
Criminal Matters Act, [245] which aims to enable South Africa to become a party to such
treaties. [246]
This Act provides for the issue of a letter of request to a foreign state for the taking
of evidence in the foreign state when an investigation has been instigated into the
commission of a crime in South Africa. [247] Parties to the proceedings in South Africa
may submit interrogatories attached to the letter of request to the foreign state, or they
may appear at the examination to examine and cross-examine witnesses in the foreign
state. [248] To permit reciprocal
Page 344
treatment, a foreign state may request the collection or taking of evidence in South
Africa for use in criminal proceedings in the foreign state. [249] In Thatcher v Minister of
Justice, [250] Mark Thatcher, son of former British Prime Minister Margaret Thatcher,
unsuccessfully challenged a request from the government of Equatorial Guinea to the
South African government that he be compelled to respond to certain questions relating
to his alleged involvement in a failed coup in Equatorial Guinea. The Court dismissed
arguments that compelling Thatcher to answer the questions posed to him would
violate his right to silence and might have an adverse effect on him in any extradition
proceedings against him. [251] In Thint Holdings (Southern Africa) (Pty) Ltd v National
Director of Public Prosecutions; Zuma v National Director of Public Prosecutions [252] the
Constitutional Court, in interpreting s 2(2), held that a letter of request may be issued
and sent to a foreign state if the circumstances of the case allow. [253]
 The Act also provides for mutual assistance in the enforcement of orders arising from
criminal proceedings. These provisions allow a South African court to request a foreign
state to assist in the recovery of fines and compensatory orders, and in the
enforcement of confiscation and restraint orders. [254] Conversely, provision is made for
the enforcement of such orders in South Africa. [255] No provision is made for the
execution of foreign prison sentences in South Africa.
An example of a mutual legal assistance treaty entered into in terms of the
International Co-operation in Criminal Matters Act of 1996 is that between South Africa
and the People’s Republic of China of 2005, [256] which provides for mutual legal
assistance in matters such as the delivery of documents, the taking of evidence, the
locating of persons, the conducting of judicial inspections, the carrying out of searches,
the confiscation of the proceeds of crimes and the exchanging
Page 345
of information on law. [257] Parties may refuse to provide assistance if the offence is not
a crime in both states, the request relates to a political offence, or ‘the Requested State
is of the opinion that the execution of the request would impair its sovereignty,
security, public order or other essential public interests, or would be contrary to the
fundamental principles of its domestic law’. [258]
This Act must be read with the Prevention of Organised Crime Act, [259] which seeks
to suppress racketeering, the financing of terrorism and money-laundering committed
in South Africa and elsewhere, inter alia, by confiscation and restraint orders and by the
preservation and forfeiture of property used in the commission of such crimes.

[1] IA Shearer Extradition in International Law (1971); NJ Botha ‘Extradition’ in Law of South Africa vol
10, 2 ed (2008); C and I Stanbrook The Law and Practice of Extradition 2 ed (2000); J Dugard and C van
den Wyngaert (eds) International Criminal Law and Procedure (1996); G Gilbert Transnational Fugitive
Offenders in International Law: Extradition and Other Mechanisms (1998); C Nicholls, C Montgomery, J
Knowles, A Doobay and M Summers The Law of Extradition and Mutual Assistance 3 ed (2013); M Cherif
Bassiouni International Extradition: United States Law and Practice 5 ed (2007).
[2] R Jennings and A Watts (eds) Oppenheim’s International Law vol 1, 9 ed (1992) 948–9.
[3] Shearer (n 1) 237; N Botha ‘The basis of extradition: The South African perspective’ (1991–92)
17 SAYIL 117 at 131–3. But see Cherif Bassiouni (n 1) 25, who argues that there may be a duty in
customary international law to extradite for international crimes.
[4] For an examination of the different bases advanced for the granting of extradition, see Botha (n 3).
Botha concludes that ‘a duty to extradite arises only in the context of a treaty commitment to do so’
(147). In Attorney-General v Andreson 1897 Off Rep 287 the Court held that there could be no
extradition in the absence of an extradition agreement. Cf the dissent of Ameshoff J at 291–4. See
also Harksen v President of the Republic of South Africa 2000 (2) SA 825 (CC) para 4.
[5] These treaties, like any other, are regulated by the law of treaties as discussed in Chapter 20.
[6] Cherif Bassiouni (n 1) 24.
[7] For a list of these, including the relevant Government Gazettes in which they were published, see
the Department of Justice’s website at http://www.justice.gov.za/ilr/mla.html.
[8] Paris, 13.XII.1957.
[9] Other than state members of the Council of Europe, the Republic of Korea and Israel have signed
and ratified the Convention. See further N Botha ‘Strange bedfellows: South Africa and accession to the
European Convention on Extradition 1957’ (1998) 23 SAYIL 247 at 249–53 in which the author describes
the present state of South Africa’s extradition relations with European states.
[10] For a further discussion of this Convention, see Chapter 10.
[11] Article 3(1) of the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment
or Punishment. South Africa ratified this treaty on 10 December 1998. See Mohamed v President of the
RSA 2001 (3) SA 893 (CC) para 60.
[12] Resolution 45/116 adopted on 14 December 1990: (1991) 30 ILM 1407. See further B Swart
‘Refusal of extradition and the UN Model Treaty on Extradition’ (1992) 23 Netherlands Yearbook on
International Law 175.
[13] Harksen (n 4); Geuking v President of the Republic of South Africa 2003 (3) SA 34 (CC) para
2; President of the Republic of South Africa v Quagliani, and Two Similar Cases 2009 (2) SA 466 (CC)
para 41.
[14] Cherif Bassiouni (n 1) 25.
[15] Ibid 25–6.
[16] 44 & 45 Vict c 69.
[17] 33 & 34 Vict c 52 (1870), 36 & 37 Vict c 60 (1873), 58 & 59 Vict c 33 (1895), 6 Edw VII c 15
(1906). (The Extradition Act 2003 now governs extradition to and from the United Kingdom.)
[18] Section 7 of Act 70 of 1934.
[19] Section 2(4) of the Extradition Act 67 of 1962. See Chapter 20 on succession to extradition
agreements. See, too, Botha (n 9).
[20] Section 2 of Act 67 of 1962.
[21] See Botha (n 3) 118, fn 6.
[22] This was not always possible. In 1968, Denmark terminated its agreement with South Africa: Proc
157 GG 2101 of 21 June 1968.
[23] Swaziland (Proc R292 GGE 2179 of 4 October 1968 (Reg Gaz 1026)); Botswana (Proc
R118 GG 2376 of 2 May 1969 (Reg Gaz 1128)); and Malawi (Proc 67 GG 3424 of 24 March 1972).
[24] Proclamation R14 GGE 6362 of 5 February 1960 (Reg Gaz 6), amended by Proc R184 GG 5283 of
10 September 1976 (Reg Gaz 2367).
[25] Proclamation 83 GG 11316 of 24 May 1988.
[26] This Scheme, drawn up by Commonwealth law ministers, based on an earlier scheme of 1966
(HMSO, London Cmnd 3008) replaces the Fugitive Offenders Act of 1881 (n 16). The text can be found in
Annexure F to the South African Law Commission’s Report on International Co-operation in Criminal
Prosecutions (Project 98, December 1995). See further D McClean International Judicial
Assistance (1992); N Botha ‘The Commonwealth Extradition Scheme and the Law Commission Working
Paper 56’ (1995) 20 SAYIL 40; and D van Zyl Smit ‘Developments in criminal law and criminal justice:
Re-entering the international community—South Africa and extradition’ (1995) 6 Crim LF 369.
[27] Although the power to ‘designate’ a state for the purposes of extradition is primarily intended for
Commonwealth countries, it is envisaged that ‘designation’ may also be extended to a non-
Commonwealth country that has the appropriate extradition legislation in place: see the South African
Law Commission’s Report on International Co-operation in Criminal Prosecutions (n 26) at 171 (para
6.105).
[28] Section 2(1)(b) of Act 67 of 1962, inserted by Act 77 of 1996.
[29] Act 77 of 1996.
[30] GNR 188 GG 18663 of 13 February 1998.
[31] GNR 593 GG 22430 of 29 June 2001 (Reg Gaz 7100), GG 24872 of 13 May 2003, respectively. See
further HB van Heerden ‘Extradition treaties negotiated since 1994’ (2005) 30 SAYIL 173.
[32] See GNR 719 GG 22542 of 1 August 2001 (Australia); GNR 391 GG 22284 of 18 May 2001
(Canada); GNR 312 GG 28680 of 7 April 2006 (India); GNR 33 GG 27168 of 21 January 2005 (China).
[33] States parties include Angola, Botswana, Lesotho, Malawi, Mauritius, Mozambique, Namibia,
Seychelles, Swaziland, Tanzania, Zambia and Zimbabwe. GNR 405 GG 35368 of 25 May 2012.
[34] See J Dugard and C van den Wyngaert ‘Reconciling extradition with human rights’ (1998)
92 AJIL 187; GN Barrie ‘Human rights and extradition proceedings: Changing the traditional landscape’
1998 TSAR 125; M du Plessis ‘The extra-territorial application of the South African Constitution’ (2003)
120 SALJ 796 at 800–13; JMT Labuschagne and M Olivier ‘Extradition, human rights and the death
penalty: Observations on the process of the internationalization of criminal justice values’ (2003)
28 SAYIL 130; C Pyle Extradition, Politics and Human Rights (2001).
[35] Geuking (n 13) para 1.
[36] Quagliani (n 13) para 39.
[37] Irish courts have refused extradition where it would violate rights guaranteed by the Constitution.
See Finucane v McMahon [1990] 1 IR 165 (HCt & SC); Magee v O’Dea [1994] 1 IR 500 (HCt).
[38] This formula, adopted from article 11 of the European Convention on Extradition of 1957 ((1960)
359 UNTS 273), appears in article 5 of the US–South African extradition treaty (n 31). In United States v
Burns [2001] 1 SCR 283, (2001) 40 ILM 1034, the Canadian Supreme Court held that there was an
obligation on the Canadian government to seek an assurance from the requesting state (if it applied the
death penalty) that the death penalty would not be imposed.
[39] 2001 (3) SA 893 (CC).
[40] 2012 (5) SA 467 (CC) para 67. In Makwakwa v S [2011] ZAFSHC 27 (11 February 2011) the
request for extradition by Lesotho in respect of conduct that included conspiracy to kill the Prime Minister
of the Kingdom of Lesotho included a written assurance that the death penalty would not be carried out
should it be imposed.
[41] See article 3(2) of the European Convention on Extradition (n 31); article 4(3) of the US–South
African Extradition Treaty (n 31).
[42] Sections 11(b)(iv) and 12(2)(c)(ii) of Act 67 of 1962.
[43] Article 3(f) (n 12).
[44] Article 3 of the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or
Punishment of 1984, (1984) 24 ILM 535. For a South African domestic example where the treaty was
applied, see Tantoush v Refugee Appeal Board 2008 (1) SA 232 (T).
[45] Article 3(f) (n 12) prohibits extradition if the extradited person would not receive the minimum
guarantees for a fair trial contained in article 14 of the International Covenant on Civil and Political
Rights.
[46] ECHR Series A no 161 (judgment of 7 July 1989), (1989) 28 ILM 1063. Cited with approval by
Gubbay CJ in Catholic Commission for Justice and Peace in Zimbabwe v Attorney General 1993 (4) SA
239 (ZS) 261–4. See further C van den Wyngaert ‘Applying the European Convention on Human Rights to
extradition: Opening Pandora’s box?’ (1990) 39 ICLQ 757; R Lillich ‘The Soering Case’ (1991)
85 AJIL 128.
[47] 98 ILR 479.
[48] Dugard and Van den Wyngaert (n 34) 200–1. The Court in S v Williams 1988 (4) SA 49 (W)
seemed to recognise this possibility when it stated that it would not extradite a fugitive to a state likely to
impose a sentence that ‘is wholly inappropriate or unconscionable’ (53F–G, 54E–I).
[49] Dugard and Van den Wyngaert (n 34) 189–91.
[50] Ibid 206–8. See Robinson v Minister of Justice and Constitutional Development 2006 (6) SA 214
(C) 230F–G.
[51] 2005 (4) SA 235 (CC). For critical discussion, see S Pete and M du Plessis, ‘South African nationals
abroad and their right to diplomatic protection: Lessons from the “Mercenaries Case”’ (2006)
22 SAJHR 439.
[52] Subsequently, the applicants were tried and sentenced to short jail terms in Zimbabwe. No
attempt was made to extradite them to Equatorial Guinea.
[53] Kaunda (n 51) para 50.
[54] Ibid paras 126–7.
[55] Ibid para 144(6). See further M du Plessis ‘The Thatcher Case and the supposed delicacies of
foreign affairs: A plea for a principled (and realistic) approach to the duty of government to ensure that
South Africans abroad are not exposed to the death penalty’ (2007) 20 SACJ 143.
[56] See n 32.
[57] Tsebe (n 40) para 43.
[58] Ibid para 66.
[59] Ibid para 31.
[60] Ibid paras 34 and 64.
[61] Tsebe (n 40) paras 52, 57 and 59. Botswana requested South Africa to try the two accused in
South Africa. But South Africa could not accede to this request because there is no legislation authorising
South Africa to prosecute people for committing murders outside of the territory of South Africa.
Arguably, this could constitute a breach of South Africa’s duty under article 5(c) of the SADC Protocol, but
the Court did not consider this point.
[62] Ironically this power seems to have been used more frequently since 1990 than before: Botha (n
3) at 137 lists three cases in which this power was used before 1990. Since then it has been used in a
number of reported cases: Hirantner v Minister of Law and Order 1992 (1) SACR 414 (W); S v Bell [1997]
2 All SA 692 (E); S v Thornhill 1997 (2) SACR 626 (C); Harksen v President of the Republic of South
Africa 2000 (2) SA 825 (CC) at 829E–F; Geuking v President of the Republic of South Africa 2003 (3) SA
34 (CC).
[63] 2000 (2) SA 825 (CC). For comments on this case and the decision of the Cape Provincial Division
in S v Harksen; Harksen v President of the RSA; Harksen v Wagner 2000 (1) SA 1185 (C), see J Dugard
and G Abraham ‘Public international law’ 2000 Annual Survey 103 at 114; N Botha ‘Further questions and
answers on extradition in the South African context’ (2002) 27 SAYIL 311; I Southwood ‘Constitutionality
of the extradition process’ (2000) 25 SAYIL 260.
[64] 2000 (2) SA 825 (CC) para 21.
[65] Ibid para 22.
[66] Harksen 2000 (2) SA 825 (CC) 835–6.
[67] 2003 (3) SA 34 (CC) (‘Geuking v President of the RSA’).
[68] Ibid 44–5.
[69] Section 2(3)(a) of Act 67 of 1962.
[70] Section 2(3)ter.
[71] See s 239 of the Constitution 1996 for the definition of ‘national legislation’, which does not
include government notices.
[72] This is clear from the report of the South African Law Commission on International Co-operation in
Criminal Prosecutions (n 26) 167 (para 6.90).
[73] 2009 (2) SA 466 (CC).
[74] Paragraph 46.
[75] Du Toit et al Commentary on the Criminal Procedure Act (vol 2) Appendix B16–B17; N Botha
‘Rewriting the Constitution: The “strange alchemy” of Justice Sachs, indeed!: South African judicial
decisions’ (2009) 34 SAYIL 253–67. See further Chapter 4.
[76] Minister of Justice v Additional Magistrate, Cape Town 2001 (2) SACR 49 (C) 61.
[77] 2008 (1) SACR 298 (SCA) para 15.
[78] See Fedsure Life Assurance Ltd v Greater Johannesburg Transitional Metropolitan Council 1999 (1)
SA 374 (CC) para 56; President of the Republic of South Africa v South African Rugby Football
Union 2000 (1) SA 1 (CC) para 148; Pharmaceutical Manufacturers Association of South Africa: In re Ex
parte President of the Republic of South Africa 2000 (2) SA 674 (CC) para 80.
[79] Fedsure Life Assurance (n 78) para 56.
[80] Shearer (n 1) 76–91; C van den Wyngaert The Political Offence Exception: The Delicate Problem of
Balancing the Rights of the Individual and International Public Order (1980) 52–63; MG Cowling
‘Unmasking “disguised extradition”—Some glimmer of hope’ (1992) 109 SALJ 241. See, too, the
judgment of the European Court of Human Rights in the Bozano Case ECHR Series A vol 111 (judgment
of 18 December 1986) discussed by Van den Wyngaert (n 46) 774.
[81] In 1962, Britain deported Dr Soblen, a citizen of the United States, to the United States to face a
charge of espionage. Had the extradition agreement between the two countries been invoked, Dr Soblen
would have been able to rely on the political offence exception. See R v Brixton Prison (Governor): Ex p
Soblen [1962] 3 All ER 641 (CA); P O’Higgins ‘Disguised extradition: The Soblen Case’ (1964) 24 Mod LR
521.
[82] Supra (n 39). See, on this case, M du Plessis ‘The extra-territorial application of the South African
Constitution’ (2003) 120 SALJ 797; N Botha ‘Deportation, extradition and the role of the state’ (2001)
26 SAYIL 227; J Dugard and G Abraham ‘Public international law’ 2001 Annual Survey 128 at 142.
[83] Mohamed (n 39) paras 44 and 52–3.
[84] Ibid 906–9.
[85] Here, the Court relied on Soering v UK (1989) 11 EHRR 439 and Minister of Justice v Burns [2001]
1 SCR 283, (2001) 40 ILM 234.
[86] Mohamed (n 39) para 58. The United Nations Human Rights Committee reached a similar
conclusion in Judge v Canada (2003) 42 ILM 1214, where Canada deported a person to the United States
without first ensuring that a sentence of death would not be carried out.
[87] For example, Mr Rashid’s removal was declared unlawful by the Supreme Court of Appeal
in Jeebhai v Minister of Home Affairs 2009 (5) SA 54 (SCA). For a discussion of the facts and law, see M
du Plessis ‘Removals, terrorism and human rights—Reflections on Rashid’ (2009) 25 SAJHR 353, 360–2.
See also the case of Samotse v Minister of Home Affairs [2014] ZAGPPHC 1001 (23 December 2014).
Most recently, see the decision of the European Court of Human Rights in Abu Zubaydah v
Lithuania (Application No 46454/11, 31 May 2018) in which the Court strongly condemned Lithuania for
its role in the CIA rendition and torture programme.
[88] 1994 (2) SACR 441 (C).
[89] [1982] 75 App R 24. This decision, involving Rhodesia—Zimbabwe—was not followed in R v
Plymouth Magistrates’ Court, Ex parte Driver [1985] 2 All ER 681 DC. This decision was later itself
overruled in Bennett v Horseferry Road Magistrates’ Court [1993] 3 All ER 138 (HL).
[90] [1993] 3 All ER 138 (HL).
[91] See the criticisms of Rosslee in J Dugard ‘Public international law’ 1994 Annual Survey 98 at 110–
11.
[92] See S v Beahan 1992 (1) SACR 307 (ZS).
[93] For the approach of the European Court of Human Rights to abduction, see Öçalan v
Turkey (2003) 42 ILM 257 at 270–97; A Künzli ‘Öçalan v Turkey: Some comments’ (2004) 17 Leiden
Journal of International Law 141.
[94] R v Officer Commanding Depot Battalion RASC, Colchester: Ex p Elliott [1949] 1 All ER 373 (KB)
376G–H; S v Ramotse (1970) cited in Ex parte Ebrahim: In re Maseko 1988 (1) SA 991 (T) 1003D. In
1962, an out-of-court settlement was made by the state in a civil action for damages for unlawful arrest
instituted by Anderson Ganyile. For a description of this arrest, see E Kahn ‘Constitutional and
administrative law’ 1962 Annual Survey 1 at 52, and Ganyile v Minister of Justice 1962 (1) SA 647 (E).
See, too, Ebrahim v Minister of Law and Order 1993 (2) SA 559 (T) in which the plaintiff successfully
claimed damages for his abduction by South African governmental agents from Swaziland (commented
on in (1992/3) 18 SAYIL 142). In Minister of Law and Order v Thandani 1991 (4) SA 862 (A) the plaintiff
successfully sued the South African police for handing him over to the Ciskei police in breach of the
extradition agreement between South Africa and Ciskei.
[95] This was acknowledged by Steyn JA in S v Ebrahim 1991 (2) SA 553 (A) 576E. See, too,
the dictum of Lord Griffiths in Bennett (n 90) 151B.
[96] Shearer (n 1) 72–6; Van den Wyngaert (n 80) 50–63.
[97] On the pre-1993 position in England, see R v Officer Commanding Depot Battalion RASC,
Colchester (n 94); R v Plymouth Magistrates’ Court, Ex parte Driver [1985] 2 All ER 681 (QB); F
Morgenstern ‘Jurisdiction in seizures effected in violation of international law’ (1952) 29 BYIL 265; P
O’Higgins ‘Unlawful seizure and irregular extradition’ (1960) 36 BYIL 279. English law has now changed
dramatically as a result of Bennett v Horseferry Magistrates’ Court (n 90) (discussed below). Cf the
decision of the European Court of Human Rights in the Stocké Case ECHR Series A no 199 (judgment of
19 March 1991).
[98] Government of Israel v Eichmann 36 ILR 18 (1961).
[99] Ganyile (n 94) 652F–H; Abrahams v Minister of Justice 1963 (4) SA 542 (C); S v Ramotse,
reported in J Dugard ‘Foreign affairs and public international law’ 1970 Annual Survey 69 at 80–
2; Ndhlovu v Minister of Justice 1976 (4) SA 250 (N); Nduli v Minister of Justice 1978 (1) SA 893 (A); Ex
parte Ebrahim: In re S v Maseko 1988 (1) SA 991 (T). The Rhodesian courts followed the South African
decisions: S v Ndhlovu 1977 (4) SA 125 (RA).
[100] Ex parte Ebrahim: In re S v Maseko 1988 (1) SA 991 (T).
[101] Supra (n 95); (1992) 31 ILM 888 (English translation). See on this decision J Dugard ‘No
jurisdiction over abducted persons in Roman-Dutch law: Male captus, male detentus’ (1991)
7 SAJHR 199; MG Cowling ‘S v Ebrahim’ (1991) SACJ 384; JHT Labuschagne ‘Die volkeregtelike dimensie
van staatlike regsnorming of wetsuitleg’ (1992) 55 THRHR 155.
[102] At 579F–G.
[103] At 568H–J. Here the Court distinguished the Appellate Division decision in Nduli v Minister of
Justice (n 99).
[104] S v Ebrahim (n 95) at 569A–B.
[105] At 582C–E.
[106] 1992 (1) SACR 307 (ZS). See, too, the judgment of the Court a quo in S v Beahan 1990 (2)
SACR 44 (Z).
[107] At 317D–F.
[108] (1992) 31 ILM 900.
[109] At 917–18. Justices Blackmun and O’Connor concurred in this dissent. For comments on this
case, see ‘Agora: International kidnapping’ in (1992) 86 AJIL 736; R Rayfuse ‘International abduction and
the United States Supreme Court: The law of the jungle reigns’ (1993) 42 ICLQ 882.
[110] Supra (n 90). Sed contra, see the opinion of the Lord Justice-General of Scotland in Bennett,
Petitioner 1994 SCCR 902.
[111] Bennett (n 90) 150–1.
[112] Bennett (n 90) 155F–I. See, too, the dictum of Lord Lowry at 163.
[113] 1991 (4) SA 862 (A).
[114] 1993 (2) SACR 18 (E).
[115] 1994 (1) SACR 530 (O).
[116] S v Mahala 1992 (2) SACR 305 (E); S v Mofokeng 1993 (2) SACR 697 (NC); S v Mabena 1993
(2) SACR 295 (B); S v Mahoko 1993 (2) SACR 509 (B).
[117] S v Mahala 1994 (1) SACR 510 (A); S v December 1995 (1) SACR 438 (A).
[118] 1994 (1) SACR 510 (A).
[119] 1995 (1) SACR 438 (A).
[120] 1927 PCIJ Reports Series A no 10. See Chapter 9.
[121] See the criticisms of decisions such as S v Mahala and S v December in J Dugard ‘Abduction:
Does the Appellate Division care about international law?’ (1996) 12 SAJHR 24; GN Barrie ‘The friendly
posse and the disregard for territorial jurisdiction’ (1996) 113 SALJ 576; HA Strydom ‘Abductions on
foreign soil—again: S v Mahala’ (1993) 9 SAJHR 308; N Botha ‘Aspects of extradition and deportation’
(1993/94) 19 SAYIL 163; JT Schoombie ‘A licence for unlawful arrests across the borders’ (1984)
101 SALJ 713.
[122] Section 1 definition of ‘foreign state’.
[123] Section 1 definition of ‘associated state’.
[124] Carolissen v Director of Public Prosecutions 2016 (2) SACR 171 (WCC) para 69.
[125] Section 4 of Act 67 of 1962.
[126] Abel v Minister of Justice 2000 (2) SACR 333 (C) para 55; Geuking v President of the Republic of
South Africa 2002 (1) SA 204 (C).
[127] 1997 (2) SACR 626 (C).
[128] As per s 10(1). See below.
[129] At 635E–G. This is because of the presumption omnia praesumuntur rite esse acta. This
presumes in favour of validity on the basis that the necessary formalities for the exercise of power by a
functionary had been complied with.
[130] As per s 3(2). The UK had not been designated per s 3(3) at the time, so the President had to
consent to the extradition of Mr Thornhill. See below.
[131] The Minister must do so if the extradition treaty obliges South Africa to extradite the sought
person. Abel v Minister of Justice (n 126) para 66.
[132] Section 5(1)(a).
[133] Abel v Minister of Justice (n 126) para 15; Moti v President of the Republic of South Africa [2017]
ZAGPPHC 501 (18 August 2017) para 26.
[134] Ibid para 38.
[135] This is the executive phase of extraditions. See below.
[136] The request must also state that the person in question is accused (or convicted) of an
extraditable offence, or alternatively an offence included in an existing extradition treaty, committed
within the jurisdiction of the requesting state. Abel v Minister of Justice (n 126) para 60.4.
[137] Abel v Minister of Justice (n 126) paras 56–60.
[138] Sections 5(1)(b) and 7.
[139] See below for a discussion on extraditable offences.
[140] Section 8.
[141] For example, as in Carolissen v Director of Public Prosecutions (n 124).
[142] See McCarthy v Additional Magistrate, Johannesburg 2000 (2) SACR 542 (SCA) and Saliu v
S [2015] ZAGPJHC 179 (25 August 2015). In both cases, there was no material non-disclosure. But both
Courts seem to assume that if there had been, then the subsequent extradition hearing would have been
unlawful.
[143] Section 9(1).
[144] Section 9(2). Garrido v Director of Public Prosecutions, Witwatersrand Local Division [2004] 4 All
SA 110 (SCA) para 24; S v Mlotshwa [2009] ZAGPPH 64 (4 March 2009) para 10; Abel v Minister of
Justice (n 126) para 45. Preparatory examinations are regulated in Chapter 20 of the Criminal Procedure
Act 51 of 1977. For a fuller discussion of preparatory examinations, see A Kruger Hiemstra’s Criminal
Procedure (2017) 20-1 et seq.
[145] S v Bell [1997] 2 All SA 692 (E) 698F; Geuking v President of the RSA (n 67) para
42(a); Director of Public Prosecutions, Cape of Good Hope v Robinson 2005 (4) SA 1 (CC) para 33.
[146] See the remarks of Howie JA in his dissenting opinion in S v McCarthy 1995 (3) SA 731 (A) 741–
2.
[147] Section 65 of the Criminal Procedure Act.
[148] Benjamin v Additional Magistrate Cape Town [2014] ZAWCHC 115 (1 August 2014) para 25.
[149] Ibid para 47. They also do not apply to the s 10(2) certificate discussed below. See Van Rooyen v
S [2014] ZAGPJHC 177 (13 March 2014) para 20.
[150] The section joins the requirement disjunctively (‘or’). Van Rooyen v S (n 149) para 10; S
v Bell (n 145) 697.
[151] Section 9(4)(a) of the Extradition Act.
[152] Robinson (n 145) para 5.
[153] Geuking v President of the RSA (n 67) para 15.
[154] [2013] ZANWHC 45 (30 May 2013).
[155] Section 13(1). The grounds for appealing or reviewing the decision are those in s 22 of the
Superior Courts Act 10 of 2013. Abel v Additional Magistrate, Cape Town; S v Abel 2002 (2) SACR 83 (C).
[156] Section 13(3).
[157] S v Tucker 2018 (1) SACR 616 (WCC).
[158] Section 10(2). Geuking v President of the RSA (n 67).
[159] See Report of the South African Law Commission on International Co-operation in Criminal
Prosecutions (n 26) 153–5, 168–70.
[160] Section 3(1).
[161] Section 3(2).
[162] Moti v President of the Republic of South Africa (n 133) para 8.
[163] Ibid para 14.
[164] Ibid para 18.
[165] In Moti v President of the Republic of South Africa (n 133) the applicants were accused by
Lebanon of fraud committed there, even though, on the common cause facts before the Court, the
applicants had never set foot in Lebanon. The applicants approached the Court in an attempt to interdict
the President from giving his consent to a request for extradition from Lebanon should Lebanon submit
one. The Court accepted that the extradition request, should it be made, would be spurious, and that
consenting to it would be irrational. But because no request had been made yet, the application for an
interdict was premature.
[166] Ibid paras 19 and 20.
[167] Section 3(3).
[168] GNR 188 GG 18663 of 13 February 1998.
[169] See article 7(2) of UN Model Treaty on Extradition (n 12); articles 2(3) and (5) of the extradition
agreement between South Africa and the Peoples’ Republic of China (n 32). See, too, Geuking v President
of the RSA (n 67) para 45; Abel v Additional Magistrate, Cape Town 2002 (2) SACR 83 (C) 92; S v
Bell [1997] 2 All SA 692 (E) 699B–C; S v Thornhill 1997 (2) SACR 626 (C) 636E; Harksen v President of
the RSA 1998 (2) SA 1011 (C) 1038H–I; Palazzolo v Minister of Justice and Constitutional
Development [2010] ZAWCHC 422 (14 June 2010). See further M v Federal Department of Justice (1979)
75 ILR 197 (involving an extradition agreement between Switzerland and South Africa); R v Governor of
Pentonville Prison; Ex p Budlong [1980] 1 WLR 1110; Re Nielsen [1984] AC 606 (HL); Riley v
Commonwealth of Australia 159 CLR 1, particularly at 15–20. In S v Bell (supra) 699–700 the Court held
that the principle applied in respect of crimes that had prescribed under South African law.
[170] See n 32, article 2(1).
[171] Section 1 of Act 67 of 1962.
[172] Palazzolo (n 169) para 34.
[173] R v Bow Street Metropolitan Stipendiary Magistrate; Ex p Pinochet Ugarte (No 3) [1999] 2 All ER
97 (HL). See, on this decision, A O’Shea ‘Pinochet and beyond: The international implications of amnesty’
(2000) 16 SAJHR 642 at 653–56; M du Plessis ‘The Pinochet cases and South African extradition law’
(2000) 16 SAJHR 669 at 680; C Warbrick ‘The extradition law aspects of Pinochet’ (1999) 48 ICLQ 958.
[174] 2017 (1) SACR 456 (SCA).
[175] Ibid para 17.
[176] Patel (n 174) paras 30 and 37 (fn 26) endorsing the views expressed in M du Plessis
‘The Pinochet cases and South African extradition law’ (2000) 16 SAJHR 669 at fn 19.
[177] For a full analysis of the problem, see Cherif Bassiouni (n 1) 349–484. For an analysis of the
problem under the former UK Extradition Acts, see A Jones Extradition and Mutual Assistance (2001) Ch
8.
[178] For example, article 7(2) of the European Convention on Extradition. See also article 2(6) of the
treaty with Argentina (GeN 519 GG 40978 of 14 July 2017); article 2(5) of the treaty with Canada (n 31);
article 2(4) of the treaty with the United States (n 30); article 3(4) of the SADC Protocol on Extradition of
3 October 2002; article 2(4) of the treaty with Australia (n 31); article 2(4) of the treaty with China (n
30); article 2(4) of the treaty with Egypt (GNR 774 GG 26497 of 2 July 2004).
[179] As per s 233 of the Constitution. See Chapter 4.
[180] For example, s 4 of the Implementation Act, s 30A of the Films and Publications Act 65 of 1996, s
12(1)(a) of the Prevention and Combating of Trafficking in Persons Act 7 of 2013, s 61 of the Criminal
Law (Sexual Offences and Related Matters) Amendment Act 32 of 2007, s 11 of the Prohibition of
Mercenary Activities and Regulation of Certain Activities in Country of Armed Conflict Act 27 of 2006, s 6
of the Prevention and Combating of Torture of Persons Act 13 of 2013, and s 15 of the Protection of
Constitutional Democracy against Terrorist and Related Activities Act 33 of 2004.
[181] 2016 (2) SACR 171 (WCC).
[182] Paragraph 48.
[183] Ibid.
[184] Paragraph 49.
[185] Paragraph 50.
[186] Paragraph 37.
[187] Paragraphs 37–41.
[188] Paragraph 42.
[189] Paragraph 46.
[190] S v Okah 2018 (1) SACR 492 (CC).
[191] Ex parte Graham: In re United States of America v Graham 1987 (1) SA 368 (T).
[192] Section 11(b) of the Act.
[193] The Minister may refuse to surrender the fugitive where proceedings against him are pending in
the Republic, where such surrender would in all the circumstances of the case be ‘unjust or
unreasonable’, or where she is satisfied that the person concerned will be prosecuted or prejudiced at his
trial in the requesting state by reason of his ‘gender, race, religion, nationality or political opinion’.
[194] See, for example, Tsebe (n 40) and Robinson (n 145).
[195] Saidi v Minister of Home Affairs 2018 (4) SA 333 (CC) para 27.
[196] Mail and Guardian Media Ltd v Chipu NO 2013 (6) SA 367 (CC) para 23.
[197] The European Convention on Extradition of 1957 ((1960) 359 UNTS 273) recognises the right of
a signatory ‘to refuse extradition of its nationals’ (article 1). The UN Model Treaty on Extradition (n 12)
recognises nationality as an optional ground for the refusal of extradition (article 4(a)).
[198] See Abel v Minister of Justice (n 126).
[199] See, for example, the Extradition Agreement between South Africa and the People’s Republic of
China, which provides in article 5: ‘(1) A Contracting State shall have the right to refuse to extradite its
own nationals. (2) If extradition is refused solely on the basis of the nationality of the person sought, the
Requested State shall, at the request of the Contracting State, submit the case to its prosecuting
authorities.’ See GNR 34 GG 27168 of 21 January 2005 (Reg Gaz 8132). In S v Pirzenthal 1969 (2) SA
224 (T) the Court held that the discretion to refuse to extradite a national is one that ‘rests with the
Minister’ and not the Court (225B).
[200] Supra (n 67). See, too, Geuking v President of the RSA 2002 (1) SA 204 (C).
[201] Geuking v President of the RSA (n 67) para 28. See, too, Abel v Minister of Justice (n 126) at
342A–C. Cf Robinson (n 145) para 20.
[202] See, for example, the Agreement with the United States (n 32) article 6. According to the Law
Commission, it is a basic principle of law that must be applied by a magistrate despite its omission from
the Act: Report of the SA Law Commission on International Co-operation in Criminal Prosecutions (n 26)
159 (para 6.67).
[203] Sections 2(3)(c) and 19 of Act 67 of 1962. In Harksen v President of the RSA 1998 (2) SA 1011
(C), the Court ‘assumed’ without deciding that the principle is part of customary international law (1039–
40). In Harksen v Minister of Justice and Constitutional Development of the RSA [2002] 4 All SA 642 (C)
para 40, the High Court considered speciality to be a principle of customary international law.
[204] Sections 23(c) and 19 of Act 67 of 1962.
[205] 2008 (5) SA 644 (SCA) para 10.
[206] Van den Wyngaert (n 80); Shearer (n 1) 166–93; DP King ‘The political offence exception in
international extradition’ (1980) 13 CILSA 247; D Dörfling ‘Die “politieke misdryf”—Uitsondering
(verweer) in aansoeke om uitlewering’ 1996 TSAR 475.
[207] Proclamation R292 GGE 2179 of 4 October 1968 (Reg Gaz 1026).
[208] S v Bull 1967 (2) SA 636 (T) at 642E–G.
[209] Ex parte Rolff 26 SC 433 at 436, 439; S v Devoy 1971 (1) SA 359 (N) 363E–H; S v Sibanda 1965
(4) SA 241 (SRA) 243–4.
[210] 34 of 1995.
[211] [1964] AC 556 (HL) 589.
[212] [1891] 1 QB 149 at 166 (Hawkins J) and 156 (Denman J). See, too, Re Meunier [1894] 2 QB 415
at 419.
[213] Quinn v Robinson 783 F 2d 776 (9th Cir 1989). See, too, Eain v Wilkes 641 F 2d 504 (7th Cir
1981) 518–23, involving the extradition of a member of the PLO to Israel.
[214] [1955] 1 QB 540 at 551.
[215] Supra (n 211).
[216] At 583.
[217] At 591.
[218] [1973] 2 All ER 204 (HL) 209. Cf Re Gross, Ex p Treasury Solicitor [1968] 3 All ER 804 (QB)
where Chapman J formulated the test as being whether the offender could claim political asylum with any
prospect of success (810D).
[219] [1996] 2 All ER 865 (HL).
[220] T v Secretary of State (n 219) 899. See, too, the dictum of Lord Mustill at 878–86.
[221] 34 of 1995.
[222] Section 20(1). See Stopforth & Veenendal v Minister of Justice 2000 (1) SA 113 (SCA) in which
the Supreme Court of Appeal held that an amnesty committee had no power to grant amnesty for an
offence committed in Namibia.
[223] Section 20(1).
[224] See, on this subject, Van den Wyngaert (n 80) 139–58; Colloquium ‘Terrorism as an international
crime’ (1989) 19 Israel Yearbook of International Law; J Dugard ‘Terrorism and international law:
Consensus at last?’ in E Yakpo and T Boumedra (eds) Liber Amicorum Mohammed Bedjaoui (1999) 159; J
Dugard ‘The problem of the definition of terrorism in international law’ in P Eden and T O’Donnell
(eds) September 11, 2001: A Turning Point in International and Domestic Law? 187; ‘Symposium on
terrorism’ in (2011) 24 Leiden Journal of International Law 651–700.
[225] In re Castioni (n 212).
[226] See Chapter 9.
[227] Convention for the Suppression of Unlawful Seizure of Aircraft (1971) 10 ILM 133, article 7.
[228] In 1975 the Chambre d’ Accusation of the Court d’ Appel of Paris refused a request from the
United States to extradite Holder and Kerkow on hijacking charges on the ground of political motive. See
McDowell Digest of United States Practice in International Law (1976) 168.
[229] Ibid. Arguably, the prosecution in 1982 of the hijackers of an Air India flight following an abortive
coup against the government of the Seychelles fell into this category: S v Hoare and Others 1982 (4) SA
865 (N).
[230] See, for example, the Convention on the Prevention and Punishment of Crimes against
Internationally Protected Persons including Diplomatic Agents of 1973—(1974) 13 ILM 43, article 7; the
International Convention Against the Taking of Hostages—(1979) 18 ILM 1456, articles 8 and 9.
[231] (1976) 15 ILM 1272, article 1. The impact of this Convention is weakened by article 13, which
permits a state to reserve the right to refuse extradition in respect of a political offence after due
consideration of a number of factors relating to the offence. See C van den Wyngaert ‘The political
offence exception to extradition: How to plug the “terrorists’ loophole” without departing from
fundamental human rights’ (1989) 19 Israel Yearbook on Human Rights 297.
[232] Official Journal of the European Communities No C313/12 of 23 October 1996. This Agreement
goes further than the European Convention on the Suppression of Terrorism as it does not permit states
to attach reservations to the obligation to extradite those suspected of offences constituting international
terrorism: article 5.
[233] (1998) 37 ILM 249; article 11; (2000) 39 ILM 268, article 14.
[234] See the extradition treaty between South Africa and the United States of 2001, which excludes
not only such crimes from the political offence but also murder and offences against the head of state:
GN R593 GG 22430 of 29 June 2001 (Reg Gaz 7100), article 4.
[235] The Schedule to Act 33 of 2004 amends the Extradition Act by the insertion of s 22, which only
excludes the political offence defence in the case of violations of ss 4 and 5 of Act 33 of 2004, dealing
with the financing of terrorism and offences relating to explosive or lethal devices. Section 1(5) of Act 33
of 2004, however, goes further and excludes the political offence defence in all crimes involving terrorist
activity.
[236] Malawi and Botswana qualify as associated states (S v Bull (n 208) 640H; S v Williams 1988 (4)
SA 49 (W) 51B) but not, so it has been held, Swaziland (Minister of Justice v Bagattini 1975 (4) SA 252
(T) 256C). Although, compare (regarding Botswana) S v Khanyisile [2012] ZANWHC 35 (19 July 2012).
The SADC Protocol does not envisage such an endorsement of warrants. See article 6.
[237] Section 6 of Act 67 of 1962.
[238] Section 12. See S v Bull (n 208) at 642–3.
[239] S v Bull (n 208) 642D, 643B–C.
[240] Section 13.
[241] See D McClean International Co-operation in Civil and Criminal Matters (2002). See further N
Boister ‘Global trends towards the simplification of extradition procedures’ in T Maluwa, M du Plessis and
D Tladi (eds) The Pursuit of a Brave New World in International Law: Essays in Honour of John
Dugard (2017).
[242] General Assembly Resolution 45/117 of 14 December 1990, (1991) 30 ILM 1421.
[243] McClean (n 241); 1990 Commonwealth Law Bulletin 1043.
[244] (1989) 28 ILM 493. See further N Boister ‘International legal regulation of drug production,
distribution and consumption’ (1996) 29 CILSA 1; N Boister Penal Aspects of the UN Drug
Conventions (2001).
[245] 75 of 1996. Reuters Group plc v Viljoen and Others NNO 2001 (12) BCLR 1265 (C) 1276–7. See,
too, HA Strydom and S du Toit ‘Transnational crime: The Southern African response’ (1998)
23 SAYIL 116; J D’Oliveira ‘International co-operation in criminal matters: The South African contribution’
(2003) 16 SACJ 323; G Kemp ‘Foreign relations, international co-operation in criminal matters and the
position of the individual’ (2003) 16 SACJ 370.
[246] Section 27.
[247] Section 2. See Kolbatschenko v King NO 2001 (4) SA 336 (C).
[248] Section 3.
[249] Sections 7 and 8. In Beheersmaaatschappij Helling I NV v The Magistrate, Cape Town 2007 (1)
SACR 99 (C), the Court held that the authorities had acted irregularly in carrying out searches of
premises in South Africa on behalf of the Netherlands. In particular, they had failed to follow the
procedure laid down in s 7 of the Act.
[250] 2005 (4) SA 543 (C).
 [251] For criticism, see M du Plessis ‘The Thatcher case and the supposed delicacies of foreign affairs:
A plea for a principled (and realistic) approach to the duty of government to ensure that South Africans
abroad are not exposed to the death penalty’ (2007) 20 SACJ 143.
[252] 2009 (1) SA 1 (CC).
[253] See Mudaly v Gwala 2011 (1) SACR 302 (KZD).
[254] Sections 13, 19, 23.
[255] Sections 15, 20, 24. Falk v National Director of Public Prosecutions 2011 (1) SACR 105 (SCA).
[256] GNR 33 GG 27168 of 21 January 2005 (Reg Gaz 8132). See, too, the mutual legal assistance
treaties with France (R224 GG 27371 of 18 March 2005) and Egypt (R775 GG 26497 of 2 July 2004).
[257] Article 1.
[258] Article 3.
[259] 121 of 1998, as amended by the Schedule to the Protection of Constitutional Democracy against
Terrorist and Related Activities Act 33 of 2004.

Page 346

Chapter 12
Immunity from Jurisdiction

John Dugard
A state has jurisdiction over all persons within its territory and over all acts that take
place within its territory. [1] In certain circumstances, however, it will not exercise its
territorial jurisdiction. This occurs where a foreign sovereign, its property or its agents
are involved. Although such persons or property are not exempt from legal liability or
immune from the observance of the local law, [2] international law exempts them from
the exercise of territorial jurisdiction. The non-assertion of jurisdiction in such a case
may be ascribed to international comity or to the argument that because ‘all sovereigns
[are] equal no one of them can be subjected to the jurisdiction of another without
surrendering a fundamental right’. [3] In Liebowitz v Schwartz, [4] Nicholas J gave his
approval to both these explanations when he observed that ‘the courts of a country will
not by their process make a foreign state a party to legal proceedings against its will’,
and stated that this principle was ‘founded on grave and weighty considerations of
public policy, international law and comity’. [5]
The immunity accorded to foreign sovereigns takes two forms: first, sovereign or
state immunity, which involves the immunity of the head of a foreign state or
government, the government of a foreign state or its senior officials or a department of
such a government; secondly, diplomatic and consular immunity, which deals with the
immunities and privileges granted to foreign diplomats and consuls.
Page 347

Sovereign or state immunity

Sovereign immunity, which is today more frequently termed ‘state immunity’, is a ‘rule
of international law which serves to preclude a state or its representatives from being
sued or prosecuted in foreign courts’. [6] It has its origin in the immunity of the person
of the foreign sovereign from the jurisdiction of municipal courts. [7] Later, the
personification of the sovereign was replaced by the abstraction of the state and its
organs. Until the emergence of the socialist state after the Russian revolution in 1917,
neither the sovereign nor her government engaged in trade or commercial activities to
any appreciable degree. Consequently, states were prepared to grant immunity to all
the acts of foreign sovereigns and their governments, including those of their armed
forces and state-owned vessels. [8] Sovereign immunity was absolute. The advent of the
socialist state and the emergence of state-owned trading corporations altered the
situation. Today, many states support a doctrine of restricted or qualified immunity,
according to which immunity from the jurisdiction of municipal courts will be granted in
respect of acts jure imperii (ie governmental public activities) and not in respect of
acts jure gestionis (ie commercial activities). The reason for this change in attitude is
that ‘a foreign government which enters into an ordinary commercial transaction with a
trader . . . must honour its obligations like other traders: and if it fails to do so, it
[should] be subject to the same laws and amenable to the same tribunals as they’. [9]
The doctrine of restricted or qualified immunity in respect of the commercial
activities of states has probably acquired the status of customary international law. This
appears from the adoption by the General Assembly of the United Nations in 2004 of a
United Nations Convention on Jurisdictional Immunities of States and their Property
prepared by the International Law Commission. It approves restricted immunity in
respect of commercial activities and asserts in its preamble ‘that the jurisdictional
immunities of States and their
Page 348
property are generally accepted as a principle of customary international law’. [10]

Among ‘Western’ states, a restrictive approach has been practised for many years.
The United Kingdom was, however, slow to abandon the absolute approach [11] out of
respect for the doctrine of stare decisis. It was only in 1976 that Lord Denning,
in Trendtex Trading Corporation v Central Bank of Nigeria, [12] approved the restrictive
approach, holding that judicial precedent ‘as to what was the ruling of international law
50 or 60 years ago’ was no longer binding, as ‘international law knows no rule of stare
decisis’. Since then approval has been given to the restrictive approach by both the
House of Lords [13] and Parliament. [14]
The influence of English decisions on South African courts in the field of international
law is no more evident than in the field of sovereign immunity, in which South African
courts slavishly followed English decisions upholding the absolute doctrine [15] until the
abandonment of this approach in Trendtex. Thereafter, both the courts and the
legislature endorsed the restrictive approach.
Inter-Science Research and Development Services (Pty) Ltd v Republica
Page 349
Popular de Moçambique, [16] decided in 1979, raised the question whether the
government of Mozambique could plead sovereign immunity in respect of the
commercial activities of a corporation owned by the government. Margo J, delivering
the judgment of the Court, acknowledged that there was ‘an abundance of South
African judicial authority . . . in support of the absolute doctrine’. [17] On the other hand,
he stated, ‘there is good reason to believe that the rule of sovereign immunity has
undergone an important change, and that the old doctrine of absolute immunity has
yielded to the restrictive doctrine’. [18] In order to demonstrate this change in
international law, Margo J examined the movement away from the absolute approach in
English law culminating in Trendtex and the State Immunity Act 1978; the adoption of
the restrictive approach by the United States, Canada and other countries; and the
support for the restrictive view on the part of modern writers in South Africa and
abroad. On this evidence, he concluded that the restrictive doctrine was a general rule
of international law and that a South African court was obliged to apply this rule, in the
absence of any statute or principle of South African law in conflict with the doctrine.
This left South African precedent, premised on English precedent, as the only obstacle
in the way of the application of the restrictive approach. On this subject, Margo J held:
Were the matter res nova, there would be no difficulty in applying the restrictive doctrine. The
only remaining question is whether or not, on the principle of stare decisis, we should follow
the earlier South African decisions. Lord Denning’s view in the Trendtex case . . . is that
international law knows no rule of stare decisis, but it does not appear to me to be necessary
in the present case to adopt that proposition. In South Africa the earlier cases are all founded
on the English decisions which laid down and reaffirmed the absolute doctrine of sovereign
immunity. . .. However, the rule stated in the earlier English decisions no longer represents the
rule of international law, and the ratio of the earlier South African cases is therefore no longer
applicable. To apply the restrictive doctrine would therefore not involve any criticism of or
dissent from the earlier South African decisions. [19]
This judgment was endorsed by Eksteen J in the Eastern Cape Division in the following
year in Kaffraria Property v Government of the Republic of Zambia. [20] Again, the Court
directed its attention to the problem caused by precedent in support of the absolute
doctrine,
Page 350
particularly De Howorth v The SS India, [21] in which Gardiner J had upheld the plea of
sovereign immunity in respect of a merchant ship owned by the Portuguese
government on the ground that ‘any use of a vessel for the purpose of obtaining
revenue for the state is a public purpose’. [22] In refusing to follow this precedent,
Eksteen J held:
When that case was decided, however, Gardiner J was merely stating and applying the rules of
international law as they existed at the time—as, in fact, we are bound to do today. Customary
international law, depending as it does on ‘universal recognition by civilized states’, is bound to
and does change from time to time as a result of changing circumstances, international
agreements or treaties, or even by virtue of the force of public opinion; and when it does so
change, as it has done on the principle of sovereign immunity, it is the duty of our courts to
ascertain the nature and extent of such change and to apply it in appropriate circumstances.
Lord Denning has expressed this principle in the Trendtex case by his dictum that ‘international
law knows no rule of stare decisis’ . . .. I therefore see no incongruity in declining today to
apply the principles enunciated in the SS ‘India’ case, without in any way reflecting on the
correctness of that decision. [23]
Shortly after these judgments were delivered, the legislature gave its approval to the
restrictive approach in the Foreign States Immunities Act, [24] which is modelled on the
United Kingdom’s State Immunity Act. [25]

1 Foreign States Immunities Act 87 of 1981 [26]

The Act starts by asserting a general immunity on the part of foreign states (including
heads of state, governments and government departments) from the jurisdiction of
South African courts in s 2(1). It then proceeds to itemise the circumstances in which
sovereign immunity will not prevail in civil cases [27] and, in so doing, gives
Page 351
approval to the restrictive approach. [28] Immunity will not be granted in the following
cases.
(a) Waiver [29]

A foreign state will have no immunity where it has expressly waived immunity after the
dispute has arisen or where it has done so by prior written agreement. A provision in an
agreement that it is to be governed by the law of South Africa shall not be regarded as
a waiver. A state is deemed to have waived immunity where it has instituted
proceedings itself. The waiver extends to any appeal or counter-claim arising out of the
proceedings.

(b) Commercial transactions

A foreign state will not be immune from the jurisdiction of municipal courts in
proceedings relating to ‘a commercial transaction’ entered into by the state.
‘Commercial transaction’, in terms of s 4 of the Foreign States Immunities Act, means:
(i)
any contract for the supply of services or goods;
(ii)
any loan or other transaction for the provision of finance and any guarantee or
indemnity in respect of any such loan or other transaction or of any other
financial obligation; and
(iii)
any other transaction or activity of a commercial, industrial, financial,
professional or other similar character into which a foreign state enters or in
which it engages otherwise than in the exercise of sovereign authority,
but does not include a contract of employment between a foreign state and an
individual.
In order to determine whether a transaction is commercial, it is therefore necessary
to consider its nature [30] and not its purpose. [31] Although most commercial
undertakings involving foreign states are covered by this definition, there are still areas
of uncertainty in which it will be necessary to have recourse to case law dealing with
the
Page 352
distinction between acta jure imperii and acta jure gestionis. As Lord Dyson MR
observed in Benkharbouche v Embassy of the Republic of Sudan, [32] the ‘distinction
between sovereign acts and non-sovereign acts is easy to state but notoriously difficult
to apply in practice’. The most helpful guidelines are probably those enunciated
in Victory Transport Inc v Comisaria General de Abastecimientos Y
Transportes. [33] According to this decision, acta jure imperii are limited to:
(1)
internal administrative acts, such as expulsion of an alien;
(2)
legislative acts, such as nationalisation;
(3)
acts concerning the armed forces;
(4)
acts concerning diplomatic activity;
(5)
public loans. [34]
Acts concerning the armed forces and diplomatic activity cause particular difficulties.
Suggestions [35] that the present law makes it clear that a contract for the purchase of
boots for the army or of a battleship for the navy [36] falls clearly within the definition of
a commercial transaction, as the purpose of the purchase is no longer relevant, fail to
have regard to the fact that courts are likely to be wary in asserting jurisdiction over
any matter relating to the armed forces. This is illustrated by Aerotrade v Republic of
Haiti, [37] in which a United States court granted immunity in respect of a claim for
payment for military equipment supplied to the Republic of Haiti for use by its armed
forces and for services rendered in connection with the supply of such equipment. The
difficulties relating to diplomatic activity are illustrated by two conflicting decisions.
In Prentice, Shaw & Scheiss Incorporated v Government of the Republic of
Bolivia, [38] on the one hand, Goldstone J classified a contract for the erection of an
embassy as ‘undoubtedly an actus jure imperii’. An English court in Planmount Ltd v
Republic of Zaire, [39] on the other hand, held that ‘it is hard to imagine a clearer case of
an act or transaction of a private or commercial nature than the repairs to the
ambassador’s residence’. [40]
The following cases illustrate the complexities of the distinction between acts jure
imperii and acts jure gestionis. In Littrell v USA (No
Page 353
2), [41] the English Court of Appeal held that a claim in tort brought by a United States
soldier stationed in the United Kingdom for negligent medical treatment at an American
military hospital was ‘clearly on the jure imperii side of the line’. In Kuwait Airways Corp
v Iraqi Airways Co [42] the House of Lords considered the question whether Iraqi Airways
was entitled to claim immunity in respect of the seizure, retention and use of aircraft
belonging to Kuwait Airways following Iraq’s invasion of Kuwait in 1990. While it
accepted that the initial seizure was jure imperii, it held, by three to two, that the
subsequent retention and use of the aircraft by Iraqi Airways was jure gestionis. The
ultimate test, for the majority, as to what constitutes an act jure imperii is whether the
act in question is of its own character a governmental act or an act that a private
citizen can perform. In Koo Golden East Mongolia v Bank of Nova Scotia, the English
Court of Appeal held that a contract by which a gold mining company deposited gold
with the Mongolian Central Bank was entered into ‘in the exercise of sovereign
authority’ as its purpose was to increase Mongolia’s currency resources. [43] In CGM
Industrial (Pty) Ltd v KPMG, the Transvaal Provincial Division held that the seizure of
documents by the government of Lesotho for the purpose of criminal investigations was
clearly an activity jure imperii. [44] The Court held in Minister of Water Affairs and
Forestry v Swissborough Diamond Mines (Pty) Ltd [45] that the South African
government might not be compelled to produce documentation relating to litigation in
Lesotho on the ground that the acts in question were acts jure imperii.

(c) Contracts of employment

A South African court will have jurisdiction, in terms of s 5(1) of the Foreign States
Immunities Act, in proceedings relating to a contract of employment between a foreign
state and an individual provided:
(a)
the contract was entered into in the Republic or the work is to be performed
wholly or partly in the Republic; and
(b)
at the time when the contract was entered into the individual was a South African
citizen or was ordinarily resident in the Republic; and
(c)
 at the time when the proceedings are brought the individual is not a citizen of the
foreign state.
Page 354
This does not apply to proceedings relating to the employment of diplomatic, consular,
administrative, technical or service staff of a foreign diplomatic mission or consular
post. Nor does it apply to a case in which the parties to the contract have agreed in
writing that any dispute arising out of the contract of employment is to be justiciable by
the courts of a foreign state. [46]

(d) Personal injury and damage to property

In terms of s 6 of the Foreign States Immunities Act, a foreign state is not immune
from the jurisdiction of a municipal court in proceedings relating to the death or injury
of any person, or to damage to tangible property caused by its act or omission in the
Republic. [47] It has been suggested [48] that one of the purposes of this provision is to
permit the victim of a traffic accident to sue the state of the foreign diplomat
responsible for the accident, despite the fact that the diplomat could not be sued in
person because of the personal immunity from civil actions conferred on diplomats by
the Vienna Convention on Diplomatic Relations.
This exception to immunity is wide enough to cover the political or governmental
acts of a foreign state. In Letelier v Republic of Chile [49] the Court dismissed the
argument that Chile could plead sovereign immunity to a claim for damages arising out
of the assassination of an opponent of the Chilean government committed by agents of
the Chilean government in the United States.
Section 6 makes no exception for the acts of armed forces in providing that a foreign
state shall not enjoy immunity in proceedings relating to ‘the death or injury of any
person’ or damage to property occurring in the territory of South Africa. In this respect,
it follows the legislation of the United States, the United Kingdom, Australia, Canada,
Singapore, Argentina, Israel and Japan, and article 12 of the United Nations Convention
on Jurisdictional Immunities of States and their Property. In Jurisdictional Immunities of
the State (Germany v Italy; Greece Intervening) [50] the International Court of Justice
was asked by Italy to find that these instruments provided evidence of a customary rule
that armed forces engaged in acts jure imperii were not entitled to immunity and hence
that Germany was responsible for the acts of its armed forces in Italy in World War II.
The Court refused to accept this argument and, in an unconvincing
Page 355
judgment, found that ‘customary international law continues to require that a state be
accorded immunity in proceedings for torts allegedly committed on the territory of
another state by its armed forces and other organs of the state in the course of
conducting an armed conflict’. [51] This decision does not reflect the law of South Africa
as s 6 continues to provide that there shall be no immunity for armed forces in such
circumstances.
In Democratic Alliance v Minister of International Relations and Co-operation; Engels
v Minister of International Relations and Co-operation [52] the Gauteng High Court held
that s 6 precluded Grace Mugabe, wife of the then President of Zimbabwe, Robert
Mugabe, from claiming immunity in a case involving an assault carried out in
Johannesburg. As Robert Mugabe himself would not have been able to succeed in a
claim for immunity on account of s 6, it followed that his wife—if she was entitled to
immunity as his spouse—would not have been able to claim immunity either. [53]
(e) Miscellaneous
The Foreign States Immunities Act further provides that a foreign state will not enjoy
immunity in respect of proceedings relating to immovable property (except where the
property is used for a diplomatic mission or a consular post); [54] patents or trade
marks; [55] membership of an association; [56] arbitration (where the foreign state has
agreed to submit a dispute to arbitration); [57] ships and cargo used for commercial
purposes; [58] sales tax, customs or excise duty; and rates in respect of premises used
for commercial purposes. [59]
A foreign government may intervene to protect its interests in property that may be
affected by the judgment in an action. In such a case, the foreign government must
prove that its claim is not illusory or founded on a title that is manifestly
defective. [60] It may plead immunity and this will stop the proceedings. Where a foreign
state
Page 356
intervenes in proceedings, this may be done by an official of that state other than the
head of its diplomatic mission. [61]
Any document required to be served for the purpose of instituting proceedings
against a foreign state shall be served by transmission through the Department of
International Relations and Co-operation to the ministry of foreign affairs of the
state. [62]

2 Enforcement [63]

In terms of s 14 of the Foreign States Immunities Act, the property of a foreign state
may not be subjected to any process for the enforcement of a judgment or an
arbitration award, unless the state gives its written consent to such a process or the
property in question is ‘in use or intended for use for commercial purposes’. [64]
The equivalent provision in the United Kingdom’s State Immunity Act of 1978 allows
the head of a state’s diplomatic mission to certify that any property is not in use for
commercial purposes and this is to be accepted as sufficient evidence of such a fact
unless the contrary is proved. [65] In practice, it is extremely difficult for a judgment
creditor to discharge the onus of proving that the property is in use for commercial
purposes. [66] Fortunately, the South African statute contains no such provision and it is
therefore left to the courts to decide this issue. [67]
Waiver of immunity from the jurisdiction of a municipal court by a foreign state does
not include consent to the enforcement of an adverse judgment. [68] A separate waiver
of immunity for the purpose of judgment is required.
Section 14 was subjected to thorough scrutiny in The Akademik Fyodorov:
Government of the Russian Federation v Marine Expeditions Inc. [69] Here, the Court
held that the warrant for the arrest of a ship for the purpose of providing security for a
claim in arbitration constituted a process for the enforcement of an arbitration award.
Consequently, the warrant of arrest was to be set aside unless the property
Page 357
was ‘in use or intended for use for commercial purposes’. The definition in s 4 of a
‘commercial transaction’ does not apply to s 14(3), which uses the term ‘commercial
purposes’, said Rose Innes J,
 since the former poses an objective criterion based upon the nature or character of a particular
transaction, contract or activity without reference to the purpose with which it was concluded
or engaged in, while the latter poses a criterion of the purpose for which property was used or
for which it was intended to be used, which introduces a subjective test relating to the
purposes or intentions of the foreign state. [70]
The Court then examined the activities in which the ship was engaged and concluded
that the polar scientific research for which it was used was a public governmental
activity. As it was not used for a commercial purpose, the Court found that it could not
be subjected to a warrant for its arrest.
In Abbott v South Africa [71] the Spanish Constitutional Court held that bank accounts
held by South Africa in Spain to be used for ordinary diplomatic and consular activities
were immune from attachment or execution despite the fact that the funds were also
used for commercial purposes.
In Rootman v President of the Republic of South Africa [72] the Supreme Court of
Appeal held that s 14, providing for the execution of a judgment in respect of
commercial property belonging to a foreign state (the Democratic Republic of the
Congo) could not be re-enforced by an order of the Court directing the South African
government to ensure compliance with the execution of the judgment debt.

3 The state, the government and separate entities

For the purposes of the Foreign States Immunities Act, a foreign state includes the
head of state, the government and any government department of the state. It does
not include a constituent part of a federal foreign state or a ‘separate entity’, ie ‘any
entity which is distinct from the executive organs of the government of that foreign
state and capable of suing or being sued’. [73] A certificate by the Minister of
International Relations and Co-operation will be conclusive proof as to whether any
territory is a constituent part of a
Page 358
federation and whether any person is to be regarded as the head of state or
government of a foreign state. [74]
When the absolute approach to sovereign immunity prevailed, much of the litigation
in this field was aimed at establishing that state-owned corporations or quasi-
government departments were not to be assimilated with the state, and were therefore
not entitled to immunity. [75] The adoption of the restrictive approach has substantially
altered the situation, as today a state-owned corporation will not enjoy immunity for a
commercial transaction—even if it can prove that it is to be assimilated with the state.
The Foreign States Immunities Act distinguishes between states and separate
entities for the purpose of immunity. A foreign state has immunity in all matters other
than the exceptional cases described in the Act. A ‘separate entity’, however, has
immunity only if the proceedings relate ‘to anything done by the separate entity in the
exercise of sovereign authority’ and ‘the circumstances are such that a foreign state
would have been so immune’. [76] The onus of proof is upon the separate entity to prove
that it is entitled to immunity. [77]
In deciding whether an entity qualifies as a government department or ‘separate
entity’, South African courts will be guided by the judgment of Goldstone J in Banco de
Mocambique v Inter-Science Research and Development Services. [78] Here, the judge
held that in order to determine the nature of the relationship between a foreign state
and a central bank, it was necessary to have regard to the principles of both South
African law and the law of the state in question, namely Mozambique:
South African law must decide upon the qualities which are necessary or sufficient to confer
upon the applicant the status of a department or organ of government. However, Mozambican
law is relevant to the issue of whether the applicant in fact possesses those qualities. [79]
Central banks will generally qualify as ‘separate entities’. However, they are given
preferential treatment in respect of the attachment of their property for the purpose of
enforcing judgment debts.
Page 359
The property of a central bank shall not be regarded as being in use for commercial
purposes, with the result that its property will always be immune from execution unless
written consent to execution is given. [80]
The Shipping Corporation of India Ltd v Evdomon Corporation [81] concerned a
related but different issue—whether a ship owned by a private shipping company, which
was a wholly owned subsidiary of the Government of India, could be attached to found
jurisdiction by a private company with a claim against the Indian government. Although
Corbett CJ considered the case law relating to sovereign immunity, he preferred to see
the two situations as ‘entirely different’. [82] He held:
[G]enerally, it is of cardinal importance to keep distinct the property rights of a company and
those of its shareholders, even where the latter is a single entity . . .. And in this regard it
should not make any difference whether the shares be held by a holding company or by a
government. [83]
Accordingly, he held that there was no reason to pierce the corporate veil in the present
case and that the property of the Shipping Corporation of India could not be attached to
found jurisdiction.
No reasons are advanced for refusing to distinguish between a non-governmental
holding company and a government in the characterisation of the status of a company,
despite the fact that sound policy considerations might require a different approach to
be adopted. If foreign governments are to be held accountable for their commercial
transactions, it is surely desirable that every effort should be made to provide the non-
governmental plaintiff with a remedy to enforce its claims—if necessary, by attaching
the property of wholly owned subsidiaries of that government in order to found
jurisdiction.

4 Indirect impleading and state immunity

The question whether Morocco was indirectly impleaded in proceedings arose

in Saharawi Arab Democratic Republic v Owners and Charterers of the Cherry
Blossom. [84] Here, an application was brought by the Saharawi Arab Democratic
Republic for an order restraining the respondents from removing phosphate from the
jurisdiction of the Court on the ground that the phosphate had been illegally mined in
Western Sahara, which had been unlawfully annexed by Morocco.
Page 360
The respondents opposed the interdict by arguing that the application indirectly
impleaded Morocco, under whose laws the phosphate had been mined. They cited
the Monetary Gold [85] principle expounded by the International Court of Justice to the
effect that a court may not exercise jurisdiction in a dispute where the lawfulness of the
conduct of a party not before the court (in casu Morocco) was in issue. The Court
rejected this argument, holding that an order that the phosphate may not be removed
from the jurisdiction of the court ‘cannot in any legal sense affect the rights of Morocco
at international law’. [86] It added that any effect that its order might have on the
interests of Morocco ‘falls within the realm of political or moral interests and cannot
have legal effect’. [87] This meant that the claim to state immunity could not be upheld.

Immunity, human rights and international crimes

Contemporary international law no longer accepts that a state may treat its nationals as
it pleases. Conventions and custom prescribe a wide range of human rights obligations
with which states must comply. Moreover, some human rights norms enjoy such a high
status that their violation, even by state officials, constitutes an international crime.
The doctrine of immunity cannot stand aloof from these developments. International
commerce has destroyed the absoluteness of state immunity in respect of commercial
transactions. International human rights law and international criminal law may now be
poised to weaken it still further. [88]
The conflict between human rights and immunity arises in two situations. First, it
arises in criminal proceedings, where an incumbent or former head of state or
government or senior government official is brought before an international criminal
tribunal or domestic court charged with a serious international crime and raises the plea
of sovereign immunity as a defence or as bar to the jurisdiction of the court to try him.
This also occurs where such a person is present in a foreign state and an international
criminal tribunal, or a state with jurisdiction to try him for such a crime, requests the
former state to assist in the surrender of such a person to the international criminal
Page 361
tribunal, or state with jurisdiction, for the purpose of bringing him to trial. Secondly, it
arises in civil proceedings, where a government or government agent is sued in civil
proceedings in the courts of a foreign state for compensation resulting from an
international crime or serious human rights violation, usually committed outside the
forum state.

1 Criminal proceedings

The question of the immunity of heads of state or government or senior state officials
for serious international crimes such as war crimes, crimes against humanity, torture or
genocide is one of the most contested issues of contemporary international law. It is
one that has divided judges of both international and national courts, led to fractious
debates in the International Law Commission [89] and spawned an acrimonious debate
among scholars. [90]
The orthodox formulation of the law of immunity of heads of state or government
and of senior state officials, largely inspired by the Arrest Warrant case [91] (of which
more later), is that there is no immunity for such persons before international criminal
tribunals but there is immunity before national courts. [92] A head of state or
government or a senior state official does, however, lose immunity after he ceases to
hold such office in respect of acts committed in a private capacity before, during or
after he assumed office. [93] Immunity persists only in respect of official acts (acta
ratione materiae) [94] but there is a strong body of opinion that holds that such
immunity does not extend to serious international crimes. [95] This formulation, which
takes no account of the distinction between immunity as a defence before a court of law
or as a bar to the jurisdiction of a court and
Page 362
immunity in respect of judicial assistance, requires careful consideration, particularly as
this distinction was in issue in the al-Bashir case before the Supreme Court of
Appeal. [96]
The examination of this branch of state immunity that follows describes the manner
in which the present law of immunity has developed. At the same time, it argues that
an exposition of the law that takes account of the distinction between the different
types of immunity described above accords more with contemporary international law’s
demand that there should be no impunity for international crimes. [97]
The attempt of the Allied powers to try the German Kaiser for international crimes
committed during World War I served notice that heads of state were no longer to be
seen as immune from prosecution for international crimes. [98] This was confirmed after
World War II by the Charters of both the Nuremberg and Tokyo tribunals, which
specifically excluded immunities as a defence or as a bar to jurisdiction for those to be
tried before the tribunals for war crimes and crimes against humanity. The Nuremberg
Tribunal stated unambiguously that ‘[t]he principle of international law, which under
certain circumstances protects the representatives of a state, cannot be applied to acts
which are condemned as criminal by international law’. [99] National prosecutions of Nazi
war criminals followed a similar course. In 1950 the General Assembly adopted a
resolution endorsing the Nuremberg Principles. [100] Principle III declared that ‘[t]he fact
that a person who committed an act which constitutes a crime under international law
acted as Head of State or responsible government official does not relieve him from
responsibility under international law’. These principles were approved by the Supreme
Court of Israel in the trial of Adolf Eichmann [101] and incorporated into the International
Law Commission’s 1996 Draft Code of Crimes against the Peace and Security of
Mankind. [102] The statutes of ad hoc international criminal tribunals for the Former
Page 363
Yugoslavia [103] and Rwanda [104] and the special court for Sierra Leone [105] likewise
exclude immunities for heads of state or senior state officials, as does the Rome Statute
of the International Criminal Court. [106] In excluding the immunity of heads of state and
government for the commission of serious international crimes, the Charters of the
Nuremberg and Tokyo tribunals and the statutes of contemporary international criminal
tribunals, the jurisprudence of these tribunals, the General Assembly’s Principles and
the International Law Commission’s Draft Code do not in any way suggest that this
exclusion of immunities is to be confined to international criminal tribunals. It is the fact
that the charges pertain to international crimes which results in the loss of immunities
as a defence or bar to jurisdiction and not the character of the court (national or
international) in which the person is to be tried. From all of this, it may be concluded
that customary international law before the Arrest Warrant case of 2002 appeared to
recognise a rule that heads of state and government and senior officials were precluded
from raising immunity as a defence in their trials or as a bar to jurisdiction, whatever
the character of the court before which such person was to be tried and regardless of
whether the person acted in a private or official capacity. [107] These developments did
not, however, exclude immunity in the case of requests for judicial assistance
emanating from another jurisdiction, national or international.
 The Pinochet case, [108] heard by the House of Lords in the late 1990s, is confusing as
six Law Lords gave separate opinions. Nevertheless, it may be seen as giving support to
the above view. Although this case involved a request for judicial assistance—the
extradition of the former Chilean dictator Pinochet from Britain, where he was receiving
medical attention, to Spain for the crime of torture committed in Chile—the House of
Lords made it clear that they viewed torture, an international crime, as one not subject
to immunity before a national
Page 364
court. The six separate opinions make it difficult to discern the ratio decidendi for the
refusal of immunity. However, Lord Phillips MR, one of the Law Lords in
the Pinochet case, later stated that several judges held that the torture alleged against
Pinochet could not constitute an official act as required for a former head of state to
succeed in a plea of immunity, but that the majority held that he could not assert
immunity ‘in relation to a criminal prosecution for torture in as much as torture is a
breach of jus cogens under international law’. [109] In other words, it was the nature of
the crime which was to determine the question of immunity. Here, the fact that the
matter was before a national court as opposed to an international court did not feature
in the reasoning of the Law Lords.
The law of immunity for heads of state or government and foreign ministers was
shaken by the Arrest Warrant case. [110] Here, the International Court of Justice held
that Belgium had violated international law by issuing a warrant for the arrest of the
Minister of Foreign Affairs (Mr Yerodia) of the Democratic Republic of Congo (DRC) on
charges of crimes against humanity and war crimes committed in the DRC in that it
failed to respect the immunity from criminal jurisdiction which the Minister enjoyed
under international law before national courts. The Court found that customary
international law precluded national courts from trying a Minister of Foreign Affairs, and
by implication other senior government officials required to travel in the course of their
duties. Although the Court could not find any state practice to support this conclusion, it
insisted
Page 365
that the function of a Minister is to travel abroad and engage foreign governments, and
this gave rise to a customary rule of immunity. [111]
Even more surprising was the finding of the Court that the Minister was immune
from prosecution before national courts for international crimes, including crimes
against humanity and war crimes. [112] It acknowledged, however, that this immunity
would not apply once he had ceased to hold office ‘in respect of acts committed prior or
subsequent to his or her period of office, as well as in respect of acts committed during
that period of office in a private capacity’. [113] Nor, said the Court, did immunity from
prosecution for international crimes extend to international criminal courts. [114]
This decision, premised on a rule of customary international law with little practice to
support it, was strongly criticised by Judge ad hoc Van den Wyngaert [115] and Judge Al-
Khasawneh [116] as a setback for the law against impunity for the commission of
international crimes.
There are several strange features of this decision. The failure of the Court to
provide evidence of state practice in support of its decision, the extension of immunity
to a foreign minister and the apparent absence of concern for impunity for international
crimes have attracted the most criticism. But, in addition, the Court made no attempt
to distinguish between immunity as a defence or bar to jurisdiction and immunity in
respect of a request for judicial assistance. This case involved a request for judicial
assistance for the arrest and extradition of Mr Yerodia to Belgium. It did not concern a
plea of immunity as a defence to a charge before a Belgian court by Mr Yerodia when
he was present in Belgium. Nor did it involve a plea by Mr Yerodia to bar the jurisdiction
of a Belgian court when he was present in Belgium. This was absolutely clear. The
Court, however, conflated all three types of immunity in holding that Belgium’s
circulation of an arrest warrant by Interpol ‘constituted a violation of an obligation of
Belgium towards the Congo, in that it failed to respect the immunity of an incumbent
Minister of Foreign Affairs of the Congo and, more particularly, infringed the immunity
Page 366
from criminal jurisdiction then enjoyed by him under international law’. [117]

The Court made it clear that Mr Yerodia’s all-encompassing immunity extended only
to national courts. In so finding, the Court flatly rejected Belgium’s argument that
provisions excluding immunity in the charters and statutes of international criminal
tribunals and the decisions of these tribunals had created a customary rule excluding
immunity in all courts, both national and international. [118] The Arrest Warrant case is
therefore seen as authority for the proposition that there is a dichotomy between
immunity for international crimes before international courts and national courts. This
bizarre dichotomy has been seized upon by some scholars and courts to claim that
heads of state or government and senior officials are immune from the jurisdiction of
national courts in respect of serious international crimes. [119]
The Arrest Warrant case has also resulted in a distinction between the acts of senior
state officials that are carried out in a private capacity and in an official capacity for the
purposes of immunity, and between immunities for incumbent officials and former
officials. [120] The latter category distinguishes between immunity ratione personae and
immunity ratione materiae. The former immunity attaches to a person because of his
status or office, while the latter form of immunity relates to acts performed in an official
capacity.
Immunity ratione personae attaches to senior state officials, such as heads of state
or government or Ministers of Foreign Affairs, while they are in office. This immunity
applies even to international crimes, as held by national courts in cases involving
Gaddafi, [121] Castro, [122] Sharon, [123] Mofaz [124] and Mugabe [125] while they were still in
office. According to Dapo Akande, writing in the wake of the Arrest Warrant case:
Judicial opinion and state practice on this point are unanimous and no case can be found in
which it was held that a state official possessing immunity ratione personae is subject to the
criminal jurisdiction of a foreign state when it
Page 367
is alleged that he or she has committed an international crime. [126]

Immunity ratione materiae, which attaches to official acts, may be invoked not only by
serving state officials in respect of their official acts, but also by former officials in
respect of official acts performed while they were in office. However, national courts
have held that such immunity does not exist when a person is charged with an
international crime, either because such acts can never be ‘official’ or because they
violate norms of jus cogens and such peremptory norms prevail over immunity. This
has recently been confirmed by the International Law Commission. [127]
As the Arrest Warrant case is seen as the authority for the distinction between
immunity before national courts and international courts and between official acts and
private acts, it is necessary to examine the facts of the decision more closely.
 ‘In law context is everything,’ said Lord Steyn in R v Secretary of State for the Home
Department, Ex parte Daly. [128] This means it is necessary to revert to the context of
the Arrest Warrant case. It concerned a request for judicial assistance by Belgium, that
is, a request for the arrest and extradition to Belgium of Mr Yerodia. It did not concern
a plea by Mr Yerodia for immunity before a Belgian court or a plea that the court lacked
jurisdiction. At no time was Mr Yerodia in Belgium before a Belgian court. This means
that the Arrest Warrant case is authority only for the proposition that there is immunity
in respect of a request for judicial assistance. This is emphasised by the fact that the
Court ordered Belgium to cancel the arrest warrant for Mr Yerodia. [129]
The issue of immunity in a request for judicial assistance came before the Supreme
Court of Appeal in Minister of Justice v SALC, [130] which concerned a visit to South
Africa by President al-Bashir of Sudan in 2015 to attend an African Union Summit
meeting. President al-Bashir had been indicted by the International Criminal Court on
charges of war crimes, crimes against humanity and genocide, which had resulted in a
call to South Africa for his arrest and surrender to the ICC in terms of the Rome
Statute. The ICC request was confirmed by a decision of the Gauteng High Court, which
ordered the government to arrest President al-Bashir and to ensure that he did not
leave the country. Despite this order, President al-Bashir was
Page 368
allowed to leave South Africa. By way of explanation and defence, the South African
government argued that it had been faced with two conflicting obligations—the
obligation to provide immunity from arrest to all heads of state attending the African
Union summit and its obligation to arrest President al-Bashir and surrender him to the
ICC.
In essence, this dispute concerned two provisions of the Rome Statute of the ICC
dealing with immunities—articles 27 and 98—which ex facie appear to contradict each
other. Article 27 restates the customary rule espoused in this chapter that immunity is
excluded as a defence or bar to jurisdiction, in that the official capacity of a person as
head of state or government or other senior official ‘shall in no case exempt a person
from criminal responsibility’; and immunities that attach to a person under national or
international law ‘shall not bar the Court from exercising its jurisdiction over such a
person’. Article 98 in somewhat convoluted language deals with the question of
immunities in the case of judicial assistance. It provides that the Court may not proceed
with a request for the surrender of a suspect which would require the requested state
to act inconsistently with its obligations under international law with respect to the
immunity of a person of a third state unless the Court can first obtain the co-operation
of that third state for the waiver of immunity. In effect, article 98 is a blocking
mechanism that sets a limit to the ability of the ICC to demand co-operation from a
state in relation to the surrender of a suspect where such co-operation would affect the
immunities of a third state. This means that the ICC Statute does not give state parties
express or implied authority to disregard the immunities of another state. The
determination as to whether a request for surrender or assistance from the ICC raises a
problem under article 98 is to be made by the state concerned and not the ICC.
Minister of Justice v SALC concerned the question of immunity in a case involving a
request for the surrender of President al-Bashir to the ICC, that is, a matter of judicial
assistance. It did not involve immunity as a defence or bar to jurisdiction. In this
respect, it was like the Arrest Warrant case. There was, however, an important
difference. In the Arrest Warrant case the request for an arrest came from a domestic
jurisdiction—Belgium—whereas here the request came from an international tribunal.
The Supreme Court of Appeal was acutely aware of this distinction when it stated that
there was a difference between saying that an international tribunal with a provision
such as article 27 of the ICC Statute excluding immunity ‘provides an exception to the
customary international law rule that heads of state enjoy immunity ratione
personae, and saying that a
Page 369
national court asked to provide assistance to that international tribunal is likewise not
bound by the customary international law rule. It is the latter proposition with which we
are concerned.’ [131]
The Supreme Court of Appeal, relying on the Arrest Warrant case, rejected the
notion that customary international law today recognises an all-encompassing exception
to immunities in the case of international crimes. [132] This did not mean, however, that
President al-Bashir was entitled to immunity while he was in South Africa. [133] In the
case of a request for judicial assistance from the ICC, as was the case here, it was
necessary to have regard to domestic law, particularly s 4(2)(a) the Implementation of
the Rome Statute of the International Criminal Court Act 27 of 2002, which provides
that ‘despite any other law to the contrary, including customary and conventional
international law’, the fact that a person is a head of state or government or senior
state official is not ‘a defence to a crime’. That South Africa was determined not to allow
immunity to be raised in its dealings with the ICC was confirmed by ss 8–10 dealing
with the procedure for surrendering a suspect to the ICC, which make no mention of
immunity. On this basis Wallis JA, speaking for the majority of the Court, [134] declared
that when South Africa decided to implement its obligations under the Rome Statute by
the Implementation Act ‘it did so on the basis that all forms of immunity, including
head-of-state immunity, would not constitute a bar to the prosecution of international
crimes in this country or to South Africa cooperating with the ICC by way of the arrest
and surrender of persons charged with such crimes before the ICC, where an arrest
warrant has been issued and a request for cooperation made’. [135]
Wallis JA continued that in adopting this approach, ‘South Africa was taking a step
that many other nations have not yet taken. If that puts this country in the vanguard of
attempts to prevent international crimes . . . that seems to me to be a matter for
national pride rather than concern. It is wholly consistent with our commitment to
human rights, both at a national and at an international level.’ The Court therefore
found that the government’s failure to arrest President al-Bashir and detain him for
surrender to the ICC was
Page 370
inconsistent with its obligations under the Rome Statute and the Implementation Act
and was unlawful. [136]
It is difficult to draw clear conclusions on the present state of international law on
the immunities of heads of state or government and of senior state officials. Before
the Arrest Warrant case there was support for the view that customary international
law did not permit immunity to be raised by a head of state or government or senior
state official as a defence or bar to jurisdiction when such a person was charged with a
serious international crime, such as genocide, crime against humanity, war crime or
torture. This portrayal of the law was based largely on the experience of Nuremberg
and Tokyo and the instruments which succeeded these trials. The concern of the
international community for a world order in which impunity for international crimes no
longer prevailed gave support to the notion that immunity had been abolished before
both international and national courts where it was raised as a defence or bar to
jurisdiction of the court. The idealism of earlier years has been shaken by the Arrest
Warrant case, which asserts the pre-World War II notion of absolute immunity before
national courts. If this decision is confined to its context—judicial assistance—this would
go a long way towards restoring faith in this idealism. It is unlikely that the
International Law Commission will do this as it is wedded to support for the Arrest
Warrant case. Whether the International Court of Justice will confine this decision to
judicial assistance in the pending advisory opinion on immunities requested by the
African Union remains to be seen.
The law governing immunity in respect of requests for judicial assistance is,
however, clear. Customary international law does not at this stage exclude immunity in
respect of a request for judicial assistance in the arrest and surrender of a head of state
or government or senior state official by either the government of another state or an
international tribunal, including the ICC. In such a case, it is for the requested state to
decide on the basis of its own law and international legal obligations whether to accept
a plea of immunity. This is precisely what the Supreme Court of appeal decided
in Minister of Justice v SALC.

2 Civil proceedings

Courts, both international and national, have held that there is nothing incongruous in
denying immunity to senior state officials in criminal proceedings but allowing immunity
in civil proceedings against a government or senior state officials for similar conduct.
Thus, several judges in Pinochet held that Pinochet could successfully
Page 371
have claimed immunity if sued in civil proceedings. [137] This distinction between
immunity in criminal and civil proceedings was accepted by the International Court of
Justice in Jurisdictional Immunities of the State (Germany v Italy) [138] and by the
Supreme Court of Appeal in the al-Bashir case. [139] In practice, little turns on this
distinction as both national [140] and international courts have accepted the broad
formulation of immunity expounded in the Arrest Warrant case. Indeed, in Jurisdictional
Immunities of the State the International Court of Justice stated that the reasoning of
the Arrest Warrant case ‘is applicable to the application of the customary international
law regarding the immunity of one State from proceedings in the courts of
another’. [141] The blind acceptance of immunity in civil proceedings concerning serious
international crimes by both national and international courts on the basis of the Arrest
Warrant case is unfortunate. In so doing, courts have failed to distinguish between
immunity as a defence, bar to jurisdiction or obstruction to judicial assistance. Had they
drawn this distinction they might well have reached a different conclusion on the broad
scope of immunity in civil proceedings.
Be that as it may, international law now accepts that a plea of immunity will succeed
in civil proceedings before a foreign court when a state is sued for damages on the
basis that its agents committed international crimes in violation of jus cogens norms,
such as torture, genocide, crimes against humanity or war crimes. Decisions of both
international and national courts attest to this. [142]
In Jones v Ministry of the Interior Al-Mamlaka Al-Arabiya as Saudiya (the Kingdom of
Saudi Arabia) the House of Lords held that developments in respect of norms of jus
cogens did not provide an exception to the rule of immunity contained in the State
Immunity Act 1978. [143] In so doing, the Court reaffirmed the distinction between civil
and criminal proceedings in respect of immunity. [144] And, in so finding, the House of
Lords reversed a decision of the English Court of Appeal, [145] which held that ‘it can no
longer be appropriate to give
Page 372
blanket effect to a foreign state’s claim to state immunity ratione materiae in respect of
a state official alleged to have committed acts of systematic torture’. [146]
American courts have also applied immunity in civil claims arising out of a breach of
a norm of jus cogens. In Siderman de Blake v Republic of Argentina [147] and Saudia
Arabia v Nelson, [148] immunity was extended to foreign governments in respect of
claims based on torture. In Princz v Federal Republic of Germany, [149] Germany
succeeded in a plea of immunity in respect of a suit brought by a Holocaust survivor to
recover damages for injuries he had suffered in Nazi concentration camps.
The European Court of Human Rights has likewise rejected the argument that
immunity does not apply in cases involving the violation of a norm of jus cogens. In Al-
Adsani v United Kingdom [150] the Court held that an English court did not violate article
6 of the European Convention on Human Rights in granting immunity to the
government of Kuwait in respect of a claim alleging torture on the part of that
government. The Court stated that:
Notwithstanding the special character of the prohibition of torture in international law, the
Court is unable to discern in the international instruments, judicial authorities or other
materials before it any firm basis for concluding that, as a matter of international law, a State
no longer enjoys immunity from civil suit in the courts of another State where acts of torture
are alleged. [151]
This decision was reached by the narrow margin of nine votes to eight. Six judges in a
strong dissenting opinion stated:
The distinction made by the majority between civil and criminal proceedings, concerning the
effect of the rule of the prohibition of torture, is not consonant with the very essence of the
operation of jus cogens rules. It is not the nature of the proceedings which determines the
effects that a jus cogens rule has upon another rule of international law, but the character of
the rule as a peremptory norm and its interaction with a hierarchically lower rule. The
prohibition of torture, being a rule of jus cogens, acts in the international sphere and deprives
the rule of sovereign immunity of all its legal effects in that sphere. The criminal or civil nature
of the domestic proceedings is immaterial. [152]
Page 373
Support for the above view is to be found in Prosecutor v Furundzija [153] in which the
International Criminal Tribunal for the Former Yugoslavia held that the prohibition on
torture is a peremptory norm which would allow the victim to ‘bring a suit for damage
in a foreign court’.
The European Court of Human Rights followed its ruling in Al-Adsani by upholding a
plea of sovereign immunity with regard to civil proceedings for reparation sought by
victims of Nazi atrocities in Greece during World War II. [154]
In 2004, in Ferrini v Federal Government of Germany, the Italian Court of Cassation
held that the Federal Republic of Germany could not succeed in a plea of immunity in a
delictual action brought by an Italian citizen for deportation and forced labour during
World War II. The Court held that international crimes constituting peremptory norms
take priority over immunity. [155]
The Ferrini decision resulted in an application by Germany to the International Court
of Justice in which it asked the Court to declare that Italy had violated its international
law obligations in respect of jurisdictional immunity by allowing such civil claims.
In Jurisdictional Immunities of the State [156] the International Court held that Italy had
violated its obligation to respect the immunity of Germany by allowing civil claims to be
brought against it based on violations of international humanitarian law. In so finding, it
stated that customary international law does not deprive a state of immunity by reason
of the fact that it was accused of serious violations of international human rights or
humanitarian law or of jus cogens rules. [157]
It is strange that, while courts were prepared to respond to the absolute immunity
rule in respect of commercial transactions by fashioning a new rule of customary
international law that restricts such immunity, they have been restrained from so doing
in respect of the absolute approach to immunity where serious violations of human
rights law and international humanitarian law are concerned. Not only have judges
refrained from so doing but they have sought to justify their inaction by claiming that it
is not the role of national judges to develop new rules of customary international law,
pretending
Page 374
to be blissfully unaware of the manner in which national courts developed a new rule of
international law in respect of commercial transactions.
In Jones, Lord Hoffmann acknowledged that the Italian Court in Ferrini had ‘given
priority to the values embodied in the prohibition of torture over the values and policies
of the rules of immunity’. He added that if the case had been concerned with domestic
law, it
might have been regarded by some as ‘activist’ but would have been well within the judicial
function . . .. But the same approach cannot be adopted in international law, which is based
upon the common consent of nations. It is not for a national court to ‘develop’ international
law by unilaterally adopting a version of that law which, however desirable, forward-looking
and reflective of values it may be, is simply not accepted by other states. [158]
This view was endorsed by Wallis JA in the al-Bashir case when he stated that
‘[h]owever tempting it may be to a domestic court to seek to expand the boundaries of
customary international law by domestic judicial decision, it is not in my view
permissible to do so’. [159]
These statements are simply wrong. The restrictive approach to sovereign immunity
in the case of commercial acts did not come about as a result of treaty or ‘the common
consent of nations’ but as a result of national court decisions from many countries that
culminated in a customary law rule. Only much later was the rule converted into
national law by legislation, and recognised by treaty in the Convention on Jurisdictional
Immunities of States and their Property. Here, national court decisions rejecting
absolute immunity provided the necessary evidence of state practice for the formation
of a new customary rule.
Foreign states are not immune from litigation in respect of international crimes by
virtue of any fundamental sovereign right, but because states, for reasons of policy and
comity, decline to exercise jurisdiction. [160] Absolute immunity in respect of commercial
transactions has given way to a restrictive approach to accord with changed
international expectations and policy. It is not unlikely that absolute immunity in
respect of the violation of norms of jus cogens will in time change because international
policy towards human rights and international crimes has undergone major changes in
recent decades. As Judge Yusuf stated in Jurisdictional Immunities of
Page 375
the State, ‘[i]mmunity is not an immutable value in international law’. [161]

Diplomatic and consular immunity [162]

The principles governing diplomatic immunity are probably the oldest of all the
principles of international law. [163] As all states have an interest in the exchange and
protection of diplomats, the rules of diplomatic protection are well settled and strictly
observed. This explains the almost universal condemnation of Iran when it held
members of the United States embassy in Tehran as hostages from 1979 to 1981,
following the admission of the deposed Shah of Iran into the United States for medical
treatment. In finding that the government of Iran had violated its obligations under
international law, the International Court of Justice declared:
[T]his case is unique and of very particular gravity because here it is not only private
individuals or groups of individuals that have disregarded and set at naught the inviolability of
a foreign embassy, but the government of the receiving state itself . . .. Such events cannot
fail to undermine the edifice of law carefully constructed by mankind over a period of centuries,
the maintenance of which is vital for the security and well-being of the complex international
community of the present day. [164]
Political relations between states are conducted by diplomatic missions, comprising
ambassadors and diplomats. Trade relations, on the other hand, are managed by
consular officials. Today, the strict distinction between diplomatic and consular services
is often blurred in the larger missions, which integrate their political and trade
representatives. [165] As the functions of diplomats and consuls differ, special rules of
immunity apply to each. Consequently, the law on diplomatic and consular immunities
is today contained in two multilateral treaties, both of which are largely declaratory of
international law and were drafted by the International Law Commission.
Page 376
These treaties are the Vienna Convention on Diplomatic Relations of 1961 [166] and the
Vienna Convention on Consular Relations of 1963. [167]
In 1951 South Africa enacted the Diplomatic Privileges Act, [168] which largely
accorded with customary international law. [169] This legislation remained in force,
subject to a number of amendments, until 1989 when South Africa finally
acceded [170] to the Vienna Conventions on both diplomatic and consular relations and
enacted legislation to give effect to its obligations under these treaties.

1 The Diplomatic Immunities and Privileges Act (excluding

the Conventions in the schedules)

The Diplomatic Immunities and Privileges Act of 1989 [171] was ambivalent as to
whether it incorporated the Vienna Conventions on diplomatic and consular relations in
full or whether it incorporated only parts of these conventions. [172] In order to remove
this uncertainty, a new Diplomatic Immunities and Privileges Act was enacted in
2001, [173] which makes it clear that these two Conventions, together with the
Convention on the Privileges and Immunities of the United Nations of 1946 and the
Convention on the Privileges and Immunities of Specialized Agencies of 1947, are,
subject to the provisions of the Act, to ‘have the force of law in the Republic’. [174]
The Act provides that the Vienna Convention on Diplomatic Relations set out in
Schedule 1 to the Act, and the Vienna Convention on Consular Relations, set out in
Schedule 2 to the Act, are to apply to all diplomatic and consular missions and their
members in the Republic. [175]
The Diplomatic Immunities and Privileges Act prescribes the immunities accorded to
diplomats, the representatives of international institutions and participants at
intergovernmental conferences held in the Republic. Visiting heads of state and special
envoys are granted immunity from the criminal and civil jurisdiction of municipal
Page 377
courts and enjoy such privileges as ‘heads of state enjoy in accordance with the rules of
customary international law’. This provision is subject to s 4(2) of the Implementation
of the Rome Statute of the International Criminal Court Act 27 of 2002, which provides:
Despite any other law to the contrary, including customary and conventional international law,
the fact that a person—
(a)
is or was a head of State or government, a member of a government or parliament, an elected
representative or a government official;
...
is neither—

(i)

a defence to a crime; nor

(ii)

a ground for a possible reduction of sentence once a person has been convicted of a crime.
This means that a head of state will not be able to plead immunity in respect of the
crimes recognised by the Rome Statute—genocide, crimes against humanity and war
crimes. This matter is discussed above in the context of Minister of Justice v SALC. [176]
The question whether the spouse of a head of state is automatically entitled to
immunity under customary international law was considered in Democratic Alliance v
Minister of International Relations and Co-operation, [177] which concerned a claim to
immunity on the part of Grace Mugabe, wife of the former president of Zimbabwe,
Robert Mugabe. Vally J held that while customary international law extends
immunity ratione personae to heads of state and government and foreign Ministers in
respect of both their official and private acts, there is no such customary rule in favour
of the spouse of such a person. After a thorough examination of the contradictory
decisions of national courts on this subject, he concluded that where immunity was
granted to a spouse, it ‘was on the basis of international comity rather than on the
basis of a finding that it is a principle of international customary law’. [178]
The Diplomatic Immunities and Privileges Act extends immunity to the United
Nations, specialised agencies and their officials in the Republic in terms of the
Convention on the Privileges and Immunities of the United Nations of 1946 and the
Convention on the Privileges and Immunities of the Specialized Agencies of
1947. [179] Intergovernmental organisations recognised by the Minister of International
Relations and Co-operation, and their officials, may be
Page 378
accorded immunities by means of agreement. [180] Immunities may also be extended to
participants in intergovernmental conferences held in the Republic. [181] Agreements are
to be published by notice in the Government Gazette. [182] In addition, the Minister of
International Relations and Co-operation may, in terms of s 7(2), grant immunities and
privileges to foreign representatives by notice in the Government Gazette in the
absence of an agreement. [183] It was under this provision that the Minister conferred
immunity, retrospectively, on Grace Mugabe in 2017 in respect of charges of assault on
the ground that such immunity was in the interest of South Africa. The Court, however,
set aside this decision on the ground that it was inconsistent with the Constitution. [184]
The Minister of International Relations and Co-operation is required to keep a
register of all persons entitled to immunity from the civil and criminal jurisdiction of the
courts of the Republic. [185] A complete list of such persons is to be published at least
once a year in the Government Gazette and is kept on the website of the Department of
International Relations and Co-operation. [186] If there is any dispute about the
entitlement of any person to immunity in legal proceedings in a municipal court, a
certificate from the Director-General of International Relations and Co-operation will
be prima facie evidence on this subject. [187] A waiver of immunity on behalf of any
person entitled to immunity is to be express and in writing. [188]
If it appears to the Minister of International Relations and Co-operation that the
immunities or privileges accorded to a mission of the Republic in the territory of any
state are less than those conferred in the Republic on the mission of that state, the
Minister may withdraw so much of the immunities or privileges accorded to that mission
as appears to be proper. [189]
All foreign missions or consular posts, the United Nations, and all specialised
agencies must submit a written request to the Director-General of International
Relations and Co-operation for acquiring, constructing, relocating, renovating,
replacing, extending or leasing
Page 379
immovable property in the name of such mission or its representatives. [190]

As the immunity of diplomats from the jurisdiction of municipal courts extends to

motor accidents, a South African citizen may find himself without any legal remedy
against a diplomat responsible for such an accident. In order to overcome this problem,
the Minister is empowered to compel persons entitled to immunity to insure themselves
adequately against claims which may result from motor accidents. [191]
Anyone (including an attorney) who ‘wilfully or without the exercise of reasonable
care’ sues a person entitled to immunity under the Diplomatic Immunities and
Privileges Act is guilty of an offence for which the punishment is a fine or imprisonment
not exceeding three years. [192]

2 The Vienna Convention on Diplomatic Relations 1961

(Schedule 1)

According to the Diplomatic Immunities and Privileges Act of 2001, the Vienna
Convention on Diplomatic Relations is to have ‘the force of law in the Republic’ and to
apply to ‘all diplomatic missions and members of such missions in the Republic’. [193] It
is therefore necessary to examine its provisions in detail.
There is no right on the part of a state to enter into diplomatic relations with another
state and there is no duty to maintain diplomatic relations once they have been entered
into. [194] Although the sending state selects its ambassador and diplomatic staff, the
receiving state has the right to declare any diplomat to be unacceptable, ie persona non
grata, before or after he assumes his duties. [195] No explanation is required for such a
decision by the receiving state.
The head of a diplomatic mission assumes his duties after he has presented his
credentials to the head of the receiving state. [196] Normally the head of a mission will
be an ambassador accredited to the head of the receiving state. However, a small
mission or one that wishes to adopt a low profile for political reasons may be headed by
Page 380
a chargé d’affaires accredited to the Minister of Foreign Affairs of the receiving
state. [197]
Article 3 describes the functions of a diplomatic mission as, inter alia, representing
the sending state, protecting its interests and nationals in the receiving state,
negotiating with the government of the receiving state, reporting on conditions in the
receiving state, and promoting friendly relations between the sending and receiving
state. A diplomatic mission may also perform consular functions.

(a) Inviolability
In order to enable a diplomatic mission to carry out its functions freely, article 22
provides that:
1.
The premises of the mission shall be inviolable. The agents of the receiving state may not
enter them, except with the consent of the head of the mission.
2.
The receiving state is under a special duty to take all appropriate steps to protect the premises
of the mission against any intrusion or damage and to prevent any disturbance of the peace of
the mission or impairment of its dignity.
3.
The premises of the mission, their furnishings and other property thereon and the means of
transport of the mission shall be immune from search, requisition, attachment or execution.
Other articles provide that the archives [198] and official correspondence [199] of a mission
are inviolable and that a mission may freely communicate with its own government and
other missions and consulates of the sending state. [200] The failure of the government
of Iran to observe these fundamental principles of diplomatic law in 1979, when it
endorsed the seizure of the United States embassy in Tehran by militants, was strongly
condemned by the International Court of Justice. [201] In Democratic Republic of the
Congo v Uganda, the International Court of Justice held that attacks by Congolese
armed forces on the Ugandan Embassy in Kinshasa violated article 22. [202]
Diplomatic missions use a diplomatic bag to transport diplomatic
Page 381
documents or articles intended for official use. [203] This bag may ‘not be opened or
detained’ by the receiving state. [204]
In recent times there have been a number of instances of serious abuse of
diplomatic premises and the diplomatic bag by missions. In 1984 persons inside the
Libyan mission in London fired shots at demonstrators outside the mission, killing a
policewoman. As a result, Britain terminated diplomatic relations with Libya. Diplomatic
bags leaving the mission were not searched or scanned. Later a search by the British
authorities, accompanied by a representative of the Saudi Arabian embassy, revealed a
number of firearms. Although the diplomatic bags were not electronically scanned, the
British government took the view that such scanning is permissible as it does not
constitute ‘opening’ or ‘detention’ of the bag. [205] In the same year an attempt was
made to transport a Nigerian political exile in London back to Nigeria in a crate labelled
as ‘diplomatic baggage’. As the crate was not properly sealed it was opened by customs
officials. Two members of the Nigerian High Commission were expelled from Britain as a
result of this abuse. [206] These two incidents suggest that the only sanctions for non-
observance of the rules of diplomatic law are termination of diplomatic relations and the
expulsion of diplomats. It has been suggested, however, that receiving states have the
right to enter missions forcibly and to open diplomatic bags in the exercise of the right
of self-defence. [207] Considerations of this kind led Pakistan in 1973 to forcibly search
the Iraqi embassy, where it discovered large consignments of arms. [208]

(b) Extraterritoriality of missions

The inviolability of the diplomatic mission led to suggestions that it was accorded a
special status by the receiving state because it was an extension of the territory of the
sending state—that it was extraterritorial. This explanation for the inviolability of the
mission is no
Page 382
longer accepted. [209] Instead, it is generally agreed that the inviolability of a mission is
based on functional necessity; that such inviolability is necessary to enable the mission
to perform its functions properly. [210]
The extraterritoriality theory was firmly repudiated by Grosskopf J in Santos v
Santos [211] when he held that a marriage solemnised in a foreign embassy or consulate
by a person (in this case the Portuguese vice-consul) who was not a recognised
marriage officer under South African law was invalid. The judge accepted the view of
Michael Akehurst [212] that ‘diplomatic premises are not extraterritorial; acts occurring
there are regarded as taking place on the territory of the receiving state, not on that of
the sending state’. [213]

(c) Diplomatic immunities

Diplomats are granted extensive protection and immunities. Article 29 states:
The person of a diplomatic agent shall be inviolable. He shall not be liable to any form of arrest
or detention. The receiving state shall treat him with due respect and shall take all appropriate
steps to prevent any attack on his person, freedom or dignity.
The private residence of a diplomat enjoys the same inviolability as the premises of the
mission. [214] Article 31 provides:
1.
A diplomatic agent shall enjoy immunity from the criminal jurisdiction of the receiving state.
He shall also enjoy immunity from its civil and administrative jurisdiction, except in the case
of:

(a)

a real action relating to private immovable property situated in the territory of the receiving
state, unless he holds it on behalf of the sending state for the purposes of the mission;

(b)

an action relating to succession in which the diplomatic agent is involved as executor,

administrator, heir or legatee as a private person and not on behalf of the sending state;
(c)

an action relating to any professional or commercial activity exercised by the diplomatic agent
in the receiving state outside his official functions.
2.
A diplomatic agent is not obliged to give evidence as a witness.
Page 383
In Portion 20 of Plot 15 Athol (Pty) Ltd v Rodrigues [215] the Court was faced with a
claim to immunity in respect of a real action relating to private property. In finding that
the Angolan ambassador did not hold the property in question ‘on behalf of the sending
state for the purposes of the mission’ as required by article 31(1)(a), Hussein J stated:
[W]hat is contemplated in the Act is that in order to attract immunity the property must be
used for the professional, diplomatic purposes of a mission and did not include a diplomatic
agent’s private residence albeit that the diplomatic agent occasionally carried out official social
obligations there. [216]
Members of a diplomat’s family enjoy the same immunities, [217] but members of the
administrative and technical staff of the mission enjoy immunity from civil jurisdiction
only in respect of acts performed within the course of their duties. [218] A diplomat who
is a national or permanent resident of South Africa will enjoy immunity only ‘in respect
of official acts performed in the exercise of his functions’. [219]
As diplomats are not immune from legal liability but only immune from being
prosecuted or sued, [220] the sending state may waive immunity on behalf of its
diplomat. Such waiver must be express. As in the case of sovereign immunity, a
separate waiver is necessary in respect of the execution of a civil judgment. [221]
There is South African authority, albeit in the form of an obiter dictum, for the
proposition that a diplomat may be arrested and detained for acts which endanger the
security of the state. This statement by Smuts J in Nkondo v Minister of Police [222] is
premised on writings published before the adoption of the Vienna Convention and, in
any event, was made before South Africa acceded to the Vienna Convention. Article 29
of the Convention makes it clear that a diplomat is exempt from ‘any form of arrest or
detention’—presumably including an arrest for an act endangering the security of the
state. In such a case, the receiving state’s remedy is not arrest but expulsion after
declaring the diplomat to be persona non grata.
Page 384
The special status of diplomats does not entitle them to attend legal proceedings
from which the public has been excluded. In S v Mothopeng, [223] Curlewis J held that
diplomats might not attend an in camera hearing in the course of a political trial.

3 The Vienna Convention on Consular Relations 1963

(Schedule 2) [224]

The functions of a consul are varied and include the promotion of trade between
sending and receiving state, the protection of nationals of the sending state who find
themselves in difficulty in the receiving state, the issue of passports to nationals of the
sending state, and the issue of visas to non-nationals who wish to travel to the sending
state. In port cities consuls deal with the interests of their ships and resolve any
disputes that may arise between master and crew. [225] Whereas a state will maintain
only one diplomatic mission in a country, it may maintain several consular offices. In
South Africa foreign embassies are situated in Pretoria (and Cape Town during the
parliamentary session) while consular offices are to be found in Johannesburg, Durban,
Cape Town and Port Elizabeth.
As consuls are not responsible for political relations with the receiving state, they
enjoy a lower degree of immunity than diplomats. [226] Consular premises are inviolable
and may not be entered by the authorities of the receiving state except with the
consent of the head of the consular post. [227] The consular archives, correspondence
and bag are also protected. [228] A consular officer may not be arrested or detained,
‘except in the case of a grave crime’, [229] and is immune from the jurisdiction of local
courts ‘in respect of acts performed in the exercise of consular functions’. [230] This
immunity may be waived by the sending state. [231]
Where a state has few interests in another state it may prefer to appoint a local
businessman, who may or may not be one of its own nationals, to represent its
interests in that state on a part-time basis. Such a representative is known as an
honorary consul who, in terms of the Vienna Convention on Consular Relations, has
fewer privileges
Page 385
than a career consul. His immunities are strictly limited to his official functions. [232]

Asylum in diplomatic and consular premises [233]

In 2012 Julian Assange sought and was granted asylum in the Ecuadorian embassy in
London. Assange, the founder of WikiLeaks, who had published thousands of classified
US cables and documents relating to the US role in armed conflicts in Iraq and
Afghanistan, feared that a request for questioning in Sweden in connection with sexual
assaults was a pretext for extraditing him to the United States for the publication of
classified material. Six years later he remains in the Ecuadorian embassy, unable to
leave without being arrested by the British police. This incident starkly illustrates the
complexity of the legal problems raised by such asylum. A foreign embassy is inviolable
and may not be entered without the consent of the sending state. However,
international law does not recognise a right of safe passage out of the country of refuge
to the asylum seeker.
Assange’s case is not unique. History provides many examples of instances in which
the opponents of a regime have sought asylum in foreign embassies and consulates in
times of political turmoil. In 1936, thousands of political refugees were granted asylum
by diplomatic missions in Madrid during the Spanish Civil War. In 1973, 25 diplomatic
missions in Santiago granted asylum to some 8 000 people following the overthrow of
the Allende government in Chile. Cardinal Mindszenty was granted refuge in the US
embassy in Budapest for 14 years, following the failure of the Hungarian uprising in
1956. In Latin America no less than ten people who at some time in their lives were
heads of state found refuge in diplomatic missions. [234]
In the last years of the apartheid era, South Africa added to this historical record.
(a)
In 1984, six members of the United Democratic Front threatened with detention
under the Internal Security Act took refuge in the British Consulate in
Durban. [235] Despite demands for their
Page 386
surrender by the South African authorities, the British government refused to
compel them to leave. In response to Britain’s refusal to surrender the six men,
 the South African government reneged on an undertaking to return four South
Africans to Britain to stand trial on charges arising out of violations of the arms
embargo, claiming that this was a lawful act of reprisal. [236]
(b)
In 1985, Klaas de Jonge, a Dutch national detained on suspicion of gun-running
for the outlawed African National Congress, escaped from police custody and
managed to enter the Netherlands embassy in Pretoria. He was immediately re-
arrested by the police but released and allowed to return to the Netherlands
embassy, following a complaint from the Dutch government that the inviolability
of its embassy had been violated by the arrest. [237] There he remained, despite
protests from the South African government, until he was allowed to return to
the Netherlands as part of an international prisoner exchange in 1987. [238]
(c)
In 1988, three prominent leaders of the United Democratic Front—Murphy
Morobe, Mohammed Valli Moosa and Vusi Khanyile—detained under the
emergency regulations, escaped and took refuge in the American consulate in
Johannesburg. [239]
The lawfulness of the granting of asylum in these cases was challenged by the South
African government on the following grounds:
(a)
The granting of asylum violated South Africa’s territorial sovereignty as
diplomatic and consular missions are not extraterritorial.
(b)
Customary international law does not recognise diplomatic asylum, as evidenced
by the Asylum Case [240] in which the International Court of Justice denied the
right of the Colombian embassy in Peru to grant asylum to a leading political
opponent of the government, Haya de la Torre.
Page 387
(c)
The Vienna Convention on Diplomatic Relations does not recognise the granting
of asylum as a permissible diplomatic function and article 41(3) prohibits the use
of the mission ‘in any manner incompatible with the functions of the mission’.
(d)
The Vienna Convention on Consular Relations likewise fails to recognise asylum
as a consular function and not only prohibits the use of consular premises for
purposes incompatible with its functions [241] but also obliges consular officials to
refrain from interfering in the internal affairs of the receiving state. [242]
Although these arguments are persuasive, they fail to take account of the growing body
of support for diplomatic asylum on humanitarian grounds. In all the South African
incidents, the fugitives had escaped from detention under arbitrary laws that fell short
of minimum standards of criminal justice. It is hardly surprising, therefore, that the
foreign states in question refused to surrender the fugitives to the South African
authorities. Considerations of humanity were seen to qualify strict compliance with the
Vienna Conventions.
The practice of granting diplomatic asylum to political fugitives (as opposed to
ordinary criminals) is bound to continue. Although diplomatic and consular premises are
not extraterritorial, they are inviolable and may not be entered by the authorities of the
receiving state without the consent of the head of the mission. In most cases, receiving
states will respect this rule rather than risk an unpleasant diplomatic incident—as
illustrated by the De Jonge affair. The receiving state therefore must rely on the
government of the sending state to surrender the fugitive, which is unlikely to occur if
the receiving state has a poor human rights record and there is popular support for the
fugitive in the sending state. Customary law may be evolving in the direction of a rule
in favour of diplomatic asylum for fugitives from oppressive regimes. Already there is
considerable evidence of usus in support of such a rule. Protests from host states
against the granting of asylum suggest, however, that the necessary opinio juris is still
absent.
The position of the post-apartheid South African government is contradictory. In
1999 the South African High Commissioner in Lusaka handed over Zambian secessionist
leader Imasiku Mutangelwa to the Zambian authorities after he had spent a week in the
High Commission seeking refuge. [243] However, in 2006, the South African government
reversed its position when it granted asylum in
Page 388
the embassy in Kinshasa to Jean-Pierre Bemba, the former Vice-President of the
Democratic Republic of the Congo. Deputy Minister of Foreign Affairs Aziz Pahad was
reported to have said: ‘Bemba is still taking refuge in South African property. He will
remain until he sees it fit for him to leave . . . South Africa is bound by international
conventions to give him protection.’ [244] It is unfortunate that the Deputy Minister failed
to specify the ‘international conventions’ to which he was referring. The Vienna
Convention on Diplomatic Relations of 1961 contains no such obligation, but it is
possible that the Deputy Minister wished to give South Africa’s support to the evolving
customary rule of granting asylum to political fugitives in foreign embassies.

[1] See, generally, C Ryngaert Jurisdiction in International Law (2008). See, too, Abdi v Minister of
Home Affairs 2011 (3) SA 37 (SCA) 50–1.
[2] In Dickinson v Del Solar [1930] 1 KB 379, the Court held that, as diplomats are not immune from
legal liability, but merely exempt from being sued where they plead diplomatic immunity, an insurance
company could not escape liability where the insured, a diplomat, had chosen not to plead immunity to a
claim arising out of a motor accident. This principle was approved in Portion 20 of Plot 15 Athol (Pty) Ltd
v Rodrigues 2001 (1) SA 1285 (W) 1293 H–I.
[3] DP O’Connell International Law 2 ed (1970) vol 2 at 842.
[4] 1974 (2) SA 661 (T).
[5] At 661H, 662A.
[6] Saharawi Arab Democratic Republic v Owners and Charterers of the Cherry Blossom 2017 (5) SA
105 (ECP) para 60. See, too, Minister of Justice and Constitutional Development v Southern Africa
Litigation Centre 2016 (3) SA 317 (SCA) (‘Minister of Justice v SALC’) para 66. See further H Fox and P
Webb The Law of State Immunity 3 ed (2015); X Yang State Immunity in International Law (2012); C
Schreuer State Immunity: Some Recent Developments (1988); GN Barrie ‘Sovereign immunity of states:
Acts iure imperii and acts iure gestionis—What is the distinction?’ (2001) 26 SAYIL 156.
[7] In Mighell v Sultan of Johore [1894] 1 QB 149, a foreign sovereign was granted immunity in a
breach of promise suit.
[8] See the judgment of the United States Supreme Court in The Schooner Exchange v McFaddon 7
Cranch 116 (1812).
[9] Per Lord Denning in Thai-Europe Tapioca Service Ltd v Government of Pakistan, The
Harmattan [1975] 1 WLR 1485 at 1491F.
[10] Resolution 59/38 of 16 December 2004, (2005) 44 ILM 801. See further H Fox ‘In defence of state
immunity: Why the UN Convention on State Immunity is important’ (2006) 55 ICLQ 399; R O’Keefe and C
Tams (eds) The United Nations Convention on Jurisdictional Immunities of States and their
Property (2013).
[11] The Parlement Belge [1880] 5 PD 197 at 205; The Porto Alexandre [1920] P 30; The
Cristina [1938] AC 485; Krajina v Tass News Agency [1949] 2 All ER 274 (CA); Baccus SRL v Servicio
Nacional del Trigo [1957] 1 QB 438 (CA).
[12] [1977] QB 529 (CA) 554G–H.
[13] I Congreso del Partido [1983] 1 AC 244 (HL); Alcom Ltd v Republic of Colombia [1984] AC 580
(HL).
[14] State Immunity Act 1978, c 33.
[15] De Howorth v The SS India 1921 CPD 451; Ex Parte Sulman 1942 CPD 407; Kavloukis v
Bulgaris 1943 NPD 190; Parkin v Government of the Republique Democratique du Congo 1971 (1) SA 259
(W) 262; Liebowitz v Schwarz 1974 (2) SA 661 (T); Lendalease Finance Co Ltd v Corporacion de
Mercadeo Agricola 1975 (4) SA 397 (C) (criticised in J Dugard ‘Public international law’ 1975 Annual
Survey 27 at 31; H Booysen ‘State immunity’ (1975) 1 SAYIL 141). In Lendalease Finance Co Ltd v
Corporacion de Mercadeo Agricola 1976 (4) SA 464 (A) the Appellate Division left open the question
whether the restrictive approach had replaced the absolute approach (499D). See, too, Prentice, Shaw &
Scheiss Incorporated v Government of the Republic of Bolivia 1978 (3) SA 938 (T) in which Goldstone AJ
(as he then was) likewise left this matter undecided (940H). (See further J Dugard ‘Public international
law’ 1978 Annual Survey 59 at 75; NJ Botha ‘State immunity lingers on’ (1978) 4 SAYIL 179. For a
discussion of these South African decisions, see J Dugard ‘The “purist” legal method, international law
and sovereign immunity’ in JJ Gauntlett (ed) JC Noster: ’n Feesbundel (1979) 36; BJ Cartoon ‘Sovereign
immunity in international law: A review of the more important trends and their place in South Africa’
(1978) 11 CILSA 168.
[16] 1980 (2) SA 111 (T). See further J Dugard ‘International law in South Africa: The restrictive
approach to immunity approved’ (1980) 97 SALJ 357.
[17] At 119B–C.
[18] At 120C.
[19] At 125G–H.
[20] 1980 (2) SA 709 (E).
[21] 1921 CPD 451.
[22] At 464.
[23] Kaffraria Property (n 20) 715B–D.
[24] 87 of 1981.
[25] 1978, c 33. The Australian Foreign States Immunities Act 196 of 1985 is also modelled on the
British statute.
[26] See W Bray and M Beukes ‘Recent trends in the development of state immunity in South African
law’ (1981) 7 SAYIL 13; G Erasmus ‘Proceedings against foreign states—The South African Foreign States
Immunities Act’ (1982) 8 SAYIL 92; N Botha ‘Some comments on the Foreign States Immunities Act 87 of
1981’ (1982) 15 CILSA 334; H Booysen ‘Procedural and jurisdictional uncertainties in the Foreign States
Immunities Act’ (1987–8) 13 SAYIL 139.
[27] Section 2(3) makes it clear that the Act is not to be construed as subjecting a foreign state to the
criminal jurisdiction of South African courts.
[28] A foreign state does not enjoy immunity if any one of the statutory exceptions contained in ss 3–
12 applies to it: The Akamedik Fyodorov: Government of the Russian Federation v Marine Expeditions
Inc 1996 (4) SA 442 (C) 443B–C.
[29] Section 3.
[30] The Akademik Fyodorov (n 28) 447F–H.
[31] The Act repudiates the ‘public purpose’ test adopted in earlier South African decisions,
particularly De Howorth v The SS India 1921 CPD 451 at 464 and Lendalease (1975) (n 15) 404C–F.
According to the UN Convention on Jurisdictional Immunities (n 10): ‘In determining whether a contract
or transaction is a “commercial transaction” . . . reference should be made primarily to the nature of the
contract or transaction, but its purpose should also be taken into account if the parties have so agreed, or
if, in the practice of the State of the forum, that purpose is relevant to determining the non-commercial
character of the contract or transaction’ (article 2(2)).
[32] [2015] EWCA Civ 33, para 21.
[33] 336F 2d 354 (2nd Cir 1964), 35 ILR 110. See, too, Republic of Argentina v Weltover Inc 504 US
607 (1992).
[34] At 360.
[35] See the judgment of Lord Denning in Trendtex (n 12) 558E; FA Mann ‘The State Immunity Act
1978’ (1979) 50 BYIL 43 at 52.
[36] The Akademik Fyodorov (n 28) 447H–I; KJ International v MV Oscar Jupiter 1998 (2) SA 130 (D)
136B–C.
[37] 376 F Supp 1281 (1974), 63 ILR 41.
[38] 1978 (3) SA 938 (W) 940H.
[39] [1981] 1 All ER 1110 (QB).
[40] At 1114F.
[41] [1995] 1 WLR (CA) 82. See, too, Holland v Lampen-Wolfe [2001] 1 WLR 1573 (HL).
[42] [1995] 1 WLR (HL) 1147.
[43] [2007] EWCA Civ 1443, (2007) 78 BYIL 582.
[44] 1998 (3) SA 738 (T) 744D–G, noted in JMT Labuschagne ‘Scope of foreign state immunity’ (1998)
23 SAYIL 262.
[45] 1999 (2) SA 345 (T) 352–4.
[46] Section 5(2).
[47] Section 6.
[48] Letelier v Republic of Chile 488 F Supp 665 (1980) 671, which deals with the equivalent section in
US legislation on this subject.
[49] Ibid 671–3.
[50] 2012 ICJ Reports 99.
[51] 2012 ICJ Reports 99 para 78.
[52] [2018] ZAGPPHC 534 (30 July 2018).
[53] Ibid para 40.
[54] Section 7.
[55] Section 8.
[56] Section 9.
[57] Section 10. For an examination of the meaning of this section, see The Akademik Fyodorov (n 28)
443–5.
[58] Section 11.
[59] Section 12.
[60] CGM Industrial (Pty) Ltd v KPMG (n 44) 745C–E.
[61] CGM Industrial (n 44) 742F–H.
[62] Section 13. See Portion 20 of Plot 15 Athol (Pty) Ltd v Rodrigues (n 2) 1290–1.
[63] J Crawford ‘Execution of judgments and foreign sovereign immunity’ (1981) 75 AJIL 820.
[64] Section 14(3). See Republica Popular de Mocambique v Main Spares Acc (Pty) Ltd 1986 (4) SA 929
(W).
[65] Section 13(5). Cf the Australian Foreign States Immunities Act 196 of 1985, s 41, which does not
give such a certificate the same evidential weight.
[66] Alcom Ltd v Republic of Colombia [1984] 1 AC 580 (HL).
[67] For an example of such a judicial enquiry by a US court, see Letelier v Republic of Chile 748 F 2d
790 (1984).
[68] Section 14(2).
[69] 1996 (4) SA 422 (C).
[70] At 447F–G. This dictum was approved in KJ International v MV Oscar Jupiter 1998 (2) SA 130 (D)
136B–C. However, on the facts of this case, the Court held that the ship was used for ‘commercial
purposes’ and that the plea of immunity could not succeed. See further EC Schlemmer ‘The immunity of
state-owned ships’ (2002) 27 SAYIL 248.
[71] 113 ILR 411 at 423–4.
[72] Oxford Reports on International Law ILDC 469 (ZA 2006) and [2006] JOL 17547 (SCA). See, too,
the decision of the High Court in Rootman ILDC 153 (ZA 2005).
[73] Section 1(2).
[74] Section 17.
[75] Lendalease(n 15) 1975 403; Krajina v Tass Agency [1949] 2 All ER 274 (CA); Baccus SRL v
Servicio Nacional Del Trigo [1957] 1 QB 438 (CA).
[76] Section 15. See Minister of Water Affairs and Forestry v Swissborough Diamond Mines 1999 (2) SA
345 (T) 353–4.
[77] H Booysen Volkereg en sy Verhouding tot die Suid-Afrikaanse Reg (1989) 302–3. Conversely, if a
party seeks to prove that an entity is a separate entity distinct from the government, the onus of proof is
on the party so alleging. See The Akademik Fyodorov (n 28) 440C–D.
[78] 1982 (3) SA 330 (T), discussed in J Dugard ‘Public international law’ 1982 Annual Survey 57 at 62.
[79] At 335E.
[80] Section 15(3). For criticism of this provision, see Erasmus (n 26) 103–4; Mann (n 35) 62.
[81] 1994 (1) SA 550 (A).
[82] At 565I.
[83] At 566C–E.
[84] 2017 (5) SA 105 (ECP). The facts of this case are fully described in Chapter 4.
[85] Monetary Gold Removed from Rome in 1943, 1954 ICJ Reports 19.
[86] Cherry Blossom (n 84) para 84.
[87] Ibid para 85.
[88] See J Foakes The Position of Heads of State and Senior Officials in International Law (2014); R van
Alebeek Immunities of States and their Officials in International Criminal Law and International Rights
Law (2008); Y Naqvi Impediments to Exercising Jurisdiction over International Crimes (2010); R O’Keefe
‘State immunity and human rights: Heads and walls, hearts and minds’ (2011) 44 Vanderbilt Journal of
Transnational Law 999; H Fox and P Webb The Law of State Immunity 3 ed (2013).
[89] See International Law Commission: Sixth Report on Immunity of State Officials from Foreign
Criminal Jurisdiction by Concepcion Escobar Hernandez, GAOR A/CN.4/722 of 12 June 2018, paras 8–17.
See further on the 2017 debate of the International Law Commission, SD Murphy ‘Crimes against
humanity and other topics: The sixty-ninth session of the International Law Commission’ (2017)
111 AJIL 970 at 981–8.
[90] See n 88.
[91] Case Concerning the Arrest Warrant of 11 April 2000 (Democratic Republic of the Congo v
Belgium) 2002 ICJ Reports 3.
[92] Ibid para 58.
[93] Ibid para 61
[94] Ibid para 61.
[95] R v Bow Street Metropolitan Stipendiary Magistrate: Ex parte Pinochet Ugarte (No 3) [1999] 2 All
ER 97(HL). For the work of the International Law Commission and its decision to exclude
immunity ratione materiae in respect of serious international crimes, see Murphy (n 89) 981–8, especially
at 984.
[96] Minister of Justice v SALC (n 6).
[97] For a full exposition of this approach, see G Mettraux, J Dugard and M du Plessis ‘Heads of state
immunities, international crimes and President Bashir’s visit to South Africa’ (2018) 18 International
Criminal Law Review 1.
[98] Ibid 8–10.
[99] Trial of the Major War Criminals before the International Military Tribunal vol I (Nuremberg 1947)
223.
[100] Resolution 95(1).
[101] See Attorney-General of Israel v Adolf Eichmann 36 ILR 277 at 311.
[102] See article 7 of the Draft Code of Crimes against the Peace and Security of Mankind in Yearbook
of the International Law Commission (1996) vol II, part 2.
[103] Article 7(2).
[104] Article 6(2).
[105] Article 6(2).
[106] Article 27.
[107] Article 46Abis of the 2014 Malabo Protocol of the African Union, which provides for the immunity
of heads of state and government and senior officials, is therefore contrary to customary international
law. (This Protocol is discussed in Chapter 26.)
[108] In fact, the House of Lords gave two judgments in this case as the first was set aside because
one of the judges (Lord Hoffmann) had failed to disclose his close links with one of the parties to the
proceedings. See R v Bow Street Metropolitan Stipendiary Magistrate: Ex parte Pinochet Ugarte [1998] 3
WLR 1456 (HL), [1998] 4 All ER 897 (HL); R v Bow Street Metropolitan Stipendiary Magistrate: Ex parte
Pinochet Ugarte (No 2) [1999] 2 WLR 272 (HL), [1999] 1 All ER 577 (HL) (setting aside the former
decision); R v Bow Street Metropolitan Stipendiary Magistrate: Ex parte Pinochet Ugarte (No 3) [1999] 2
WLR 872 (HL), [1999] 2 All ER 97 (HL). See, too, Khurts Bat v German Federal Court [2011] EWHC 2029
(Admin).
[109] Jones v Ministry of the Interior Al-Mamlaka Al-Arabiya as Saudiya (The Kingdom of Saudi
Arabia) [2004] EWCA Civil 1394 paras 123–4. Cf D Akande ‘International law immunities and the
International Criminal Court’ (2004) 98 AJIL 407 at 415; D Akande and S Shah ‘Immunities of state
officials, international crimes and foreign domestic courts’ (2011) 21 EJIL 815; R van Alebeek
‘The Pinochet case: International human rights on trial’ (2000) 71 BYIL 29; A Bianchi ‘Immunity versus
human rights: The Pinochet case’ (1999) 10 EJIL 237; CH Powell and A Pillay ‘Revisiting Pinochet: The
development of customary international criminal law’ (2001) 17 SAJHR 477.
[110] Case Concerning the Arrest Warrant of 11 April 2000 (Democratic Republic of Congo v
Belgium) 2002 ICJ Reports 3, (2002) 41 ILM 536. For comments on this case, see G Erasmus and G
Kemp ‘The application of international criminal law before domestic courts in the light of recent
developments in international and constitutional law’ (2002) 27 SAYIL 64; K Hopkins ‘The International
Court of Justice and sovereign immunity: Why the Yerodia Case is an unfortunate ruling for the
development of public international law’ (2002) 27 SAYIL 256; M du Plessis and S Bosch ‘Immunities and
universal jurisdiction—The world court steps in (or on?)’ (2003) 28 SAYIL 246. See, too, Case Concerning
Certain Criminal Proceedings in France (Republic of Congo v France), Provisional Measures 2003 ICJ
Reports 102, (2003) 42 ILM 852 (discontinued 2010); Case Concerning Certain Questions of Mutual
Assistance in Criminal Matters (Djibouti v France) 2008 ICJ Reports 177.
[111] Arrest Warrant case (n 110) paras 52–4.
[112] Ibid paras 56–8.
[113] Ibid para 61. The Court fails to define acts committed in ‘a private capacity’. Would acts of
torture authorised by an official entitled to immunity while in office be subject to prosecution after he
leaves office, as held in R v Bow Street Metropolitan Stipendiary Magistrate: Ex parte Pinochet Ugarte (No
3) [1999] 2 All ER 97 (HL)?
[114] Arrest Warrant case (n 110) para 61.
[115] Ibid paras 24–38.
[116] Ibid para 7.
[117] Arrest Warrant case (n 110) para 71.
[118] Ibid para 58.
[119] See the cases described in Mettraux, Dugard and Du Plessis (n 97) 8, fn 19.
[120] Arrest Warrant case (n 110) para 58.
[121] 125 ILR 456 (France). See S Zappala ‘Do heads of state in office enjoy immunity from jurisdiction
for international crimes? The Ghaddafi Case before the French Cour de Cassation’ (2001) 12 EJIL 595.
[122] A Cassese International Criminal Law (2003) 272 (Spain).
[123] (2003) 42 ILM 596 (Belgium).
[124] (2004) 53 ICLQ 771 (UK).
[125] (2004) 53 ICLQ 769 (UK).
[126] ‘International law immunities and the International Criminal Court’ (2004) 98 AJIL 407 at 411.
See, too, D Akande and S Shah ‘Immunities of state officials, international crimes and foreign domestic
courts’ (2011) 21 EJIL 815.
[127] Murphy (n 89) 981–8.
[128] [2001] 2 AC 732.
[129] Arrest Warrant case (n 110) para 78.
[130] Supra (n 6).
[131] Minister of Justice v SALC (n 6) para 77.
[132] Ibid para 84. For the reasoning of the Court, see paras 70–84.
[133] Ibid para 85.
[134] A minority comprising Ponnan and Lewis JJA disagreed with the majority’s finding that customary
international law did not recognise an exception to immunity in the case of international crimes (ibid para
115).
[135] Minister of Justice v SALC (n 6) para 103 (emphasis added).
[136] Minister of Justice v SALC (n 6) para 113.
[137] [1999] 2 All ER 97 (HL) 157 (Lord Hutton), 179 (Lord Millett) and 182 (Lord Phillips).
[138] 2012 ICJ Reports 99 paras 87, 91.
[139] Minister of Justice v SALC (n 6) para 73.
[140] See the decisions cited by the International Court of Justice in Jurisdictional Immunities of the
State (n 138) paras 85, 96.
[141] Jurisdictional Immunities of the State (n 138) para 95.
[142] Ibid paras 85, 96.
[143] Jones v Ministry of the Interior Al-Mamlaka Al-Arabiya as Saudiya (Saudi Arabia) [2006] UKHL 26
paras 26–8.
[144] Paragraph 71.
[145] Jones v Ministry of the Interior Al-Mamlaka Al-Arabiya as Saudiya (The Kingdom of Saudi
Arabia) [2004] EWCA Civil 1394.
[146] Jones v Ministry of the Interior Al-Mamlaka (n 145) paras 92 (Mance LJ) and 131 (Lord Phillips
MR).
[147] 965 F 2d 699 (9th Cir 1992); see, too, Argentine Republic v Amerada Hess Shipping Corp 488 US
428 (1989).
[148] 100 ILR 544.
[149] 26 F 3d 1166 (DC Cir 1994).
[150] (2001) 34 EHRR 273, 123 ILR 24. See E Voyiakis ‘Access to court v state immunity’ (2003)
52 ICLQ 297.
[151] Paragraph 61. See, too, paras 65–6.
[152] Joint dissenting opinion of Judges Rozakis and Caflisch, joined by Judges Wildhaber, Costa,
Cabral Barreto and Vajic (Al-Adsani (n 150) para 4).
[153] (1998) 38 ILM 317 at 349–50 para 155.
[154] Kalogeropoulou v Greece and Germany ECHR Application No 59021/00 (judgment of 12
December 2002), (2003) 42 ILM 1030. Here, the European Court of Human Rights overruled a decision of
the Greek Supreme Court in Prefecture of Volotia v Federal Republic of Germany, noted by M Gavouneli
and I Bantekas in ‘Prefecture of Voiotia v federal Republic of Germany’ (2001) 95 AJIL 198.
[155] Ferrini v Federal Republic of Germany, noted by A Bianchi ‘Ferrini v Federal Republic of Germany’
(2005) 99 AJIL 242.
[156] Supra (n 138).
[157] Ibid paras 91, 97.
[158] Jones (n 143) para 63.
[159] Minister of Justice v SALC (n 6) para 74.
[160] See LM Caplan ‘State immunity, human rights and jus cogens: A critique of the normative
hierarchy theory’ (2003) 97 AJIL 741 at 781. See, too, JMT Labuschagne ‘Immunity of the head of state
for human rights violations in international criminal law’ (2001) 26 SAYIL 180.
[161] Supra (n 138) para 35, dissenting opinion of Judge Yusuf.
[162] See, generally, C Lewis State and Diplomatic Immunity 3 ed (1990); I Roberts (ed) Satow’s
Guide to Diplomatic Practice 7 ed (2017); B Sen A Diplomat’s Handbook of International Law and
Practice 3 ed (1988); E Denza Diplomatic Law 4 ed (2016); J Brown ‘Diplomatic immunity: State practice
under the Vienna Convention on Diplomatic Relations’ (1988) 37 ICLQ 53; P Behrens Diplomatic
Interference and the Law (2016).
[163] See the remarks of Schutz J in S v Muchindu 1995 (2) SA 36 (W) 37–8.
[164] Case Concerning United States Diplomatic and Consular Staff in Tehran 1980 ICJ Reports 3 at
42–3. The decision in this case is endorsed by Schutz J in S v Muchindu (n 163) 37–8. This is one of the
few occasions on which a South African court has cited a judgment of the International Court of Justice.
[165] DP O’Connell International Law 2 ed (1970) vol 2 at 914.
[166] 500 UNTS 95.
[167] 596 UNTS 261.
[168] 71 of 1951.
[169] S v Penrose 1966 (1) SA 5 (N).
[170] On 21 August 1989. See further on this accession the statement by the Deputy Minister of
Foreign Affairs, Mr JWH Meiring, in Debates of Parliament cols 10196–200 (23 May 1989).
[171] Act 74 of 1989.
[172] See J Dugard and G Abraham ‘Public international law’ 2001 Annual Survey 128 at 135–6. See,
too, Portion 20 of Plot 15 Athol (Pty) Ltd v Rodrigues (n 2) 1292–3.
[173] Act 37 of 2001.
[174] Sections 1 and 2.
[175] Section 3. Members of a diplomat’s family are also entitled to immunity. See s 2(b) of Act 37 of
2001 as amended by Act 35 of 2008.
[176] Supra (n 6) paras 49–53,92–5,102–3, 116–24.
[177] Supra (n 52).
[178] Paragraph 35.
[179] Section 5(1) and (2).
[180] Section 5(3).
[181] Section 6.
[182] Section 7.
[183] Section 7(2). See Proc 41 GG 14809 of 12 May 1993, for an example of such a notice.
[184] Democratic Alliance v Minister of International Relations and Co-operation (n 52) paras 43, 46.
[185] Section 9(1).
[186] Section 9(2) as amended by Act 35 of 2008.
[187] Section 9(3).
[188] Section 8.
[189] Section 10.
[190] Section 12. In Portion 20 of Plot 15 Athol (Pty) Ltd v Rodrigues (n 2) the Court held that a plea of
diplomatic immunity must fail in respect of property that had been acquired by a diplomat without the
submission of a written request to the Director-General.
[191] Section 13.
[192] Section 15.
[193] Sections 2(1) and 3(1).
[194] Article 2 of the Vienna Convention on Diplomatic Relations provides that the establishment of
diplomatic relations ‘takes place by mutual consent’.
[195] Article 9.
[196] Article 13.
[197] Article 14.
[198] Article 24.
[199] Article 27(2).
[200] Article 27(1).
[201] Case Concerning United States Diplomatic and Consular Staff in Tehran (n 164) at 30–1, 42.
[202] 2005 ICJ Reports 168 at 227–8, paras 337–8, 340.
[203] During the apartheid era, the diplomatic bag was apparently used by opponents of the regime to
transmit sensitive material to persons abroad: see A Wiebalck ‘Abuse of the immunity of diplomatic mail’
(1984) 10 SAYIL 175.
[204] Article 27(3) and (4).
[205] See R Higgins ‘The abuse of diplomatic privileges and immunities: Recent United Kingdom
experience’ (1985) 79 AJIL 641. In 1989, the International Law Commission adopted a set of Draft
Articles on the Status of the Diplomatic Courier and the Diplomatic Bag Not Accompanied by Diplomatic
Courier: Yearbook of the International Law Commission (1989) vol II, part 2 at 43. Article 28 provides
that the diplomatic bag ‘shall be exempt from examination directly or through electronic or other
technical devices’.
[206] DJ Harris Cases and Materials on International Law 7 ed (2010) 309.
[207] Brown (n 162) 86.
[208] Denza (n 162) 84.
[209] See WJ van der Merwe ‘Die grondgebied van ambassadepersele’ (1987)
28(1) Codicillus 20; Radwan v Radwan [1972] 3 All ER 967 (Fam) 971; S v Mharapara 1986 (1) SA 556
(ZS) 558G–559B.
[210] Portion 20 of Plot 15 Athol (Pty) Ltd v Rodrigues (n 2) 1293E.
[211] 1987 (4) SA 150 (W). Discussed by P van Warmelo ‘Statum personale, lex loci regi actum en die
beginsel van ekstraterritorialiteit’ (1988) 51 THRHR 102.
[212] At 152F–G.
[213] A Modern Introduction to International Law 4 ed (1982) 115.
[214] Article 30.
[215] 2001 (1) SA 1285 (W). See further on this case G Abraham ‘Portion 20 of Plot 15 Athol—“Some
corner of a foreign field that is forever . . . Angola?”’ (2001) 118 SALJ 441; JMT Labuschagne ‘Diplomatic
immunity: A jurisdictional or substantive-law defence?’ (2002) 27 SAYIL 291.
[216] At 1296C–D.
[217] Article 37(1).
[218] Article 37(2).
[219] Article 38.
[220] Dickinson (n 2).
[221] Article 32.
[222] 1980 (2) SA 894 (O) 900–2.
[223] 1979 (4) SA 367 (T).
[224] LT Lee and J Quigley Consular Law and Practice 3 ed (2008).
[225] Article 5 of the Vienna Convention on Consular Relations.
[226] S v Penrose 1966 (1) SA 5 (N) 11. See further J Dugard ‘Consular immunity’ (1966)
83 SALJ 126.
[227] Article 31.
[228] Articles 33 and 35.
[229] Article 41.
[230] Article 43. Customary law likewise adopts a functional approach to consular immunity: see
Dugard (n 226) 128–30.
[231] Article 45.
[232] Article 58. In S v Penrose (n 226) the Court held that the honorary consul for Colombia could not
raise a plea of immunity to a charge of negligent driving.
[233] See P Behrens Diplomatic Interference and the Law (2016) Ch 11; SP Sinha Asylum in
International Law (1971); A Jeffery ‘Diplomatic asylum: Its problems and potential as a means of
protecting human rights’ (1985) 1 SAJHR 10; F Morgenstern ‘Extra-territorial asylum’ (1948)
25 BYIL 236; R Jennings and A Watts (eds) Oppenheim’s International Law 9 ed (1992) vol 1 at 1082.
[234] Jeffery (n 233) 23.
[235] For accounts of this asylum, see Jeffery (n 233); J Dugard ‘Public international law’ 1984 Annual
Survey 68 at 72–5; W Booyse ‘South Africa’s foreign policy and international practice as reflected in
speeches, parliamentary statements and replies—1984’ (1984) 10 SAYIL 300–3.
[236] JC Heunis The Coventry Four (1985); GN Barrie ‘The Durban Six and the Coventry Four: Asylum
and reprisal in international law’ (1984) 10 SAYIL 138. Cf G Erasmus ‘Afdwinging van die volkereg deur
weerwraak: Wanneer geoorloof?’ (1986) 49 THRHR 38; JD van der Vyver ‘The Coventry Four: Another
perspective’ (1985–6) 11 SAYIL 157.
[237] See D Basson ‘Die volkeregtelike beginsels insake diplomatieke asiel en die De Jonge debakel’
(1985) 18 De Jure 300; J Dugard ‘Public international law’ 1985 Annual Survey 69 at 73.
[238] J Dugard ‘Public international law’ 1987 Annual Survey 74 at 78.
[239] J Dugard ‘Public international law’ 1988 Annual Survey 68 at 72–3.
[240] 1950 ICJ Reports 266.
[241] Article 5(m).
[242] Article 55. See Booyse (n 235) 301.
[243] Saturday Star 7 and 14 August 1999.
[244] Legalbrief Today 29 March 2007; Mail & Guardian 30 March–4 April 2007 at 16.