Australian Year Book of International Law
Don Greig is a man of many parts. Alongside his distinguished contribution to public international law and the law of contract, the arts and soccer are very important to him. I have been privileged to be one of his colleagues on and off the soccer pitch. His advice and friendship have been equally valuable: his comments on one’s performance, whether as legal academic or footballer, are always as perceptive as they are trenchant. I offer this essay, on the relationship between our two fields, with gratitude and respect.
Many constitutions specify that customary international law, and reciprocal treaties which the state has ratified, are to form part of municipal law without the need for further state action. This monist model, in which municipal and international law form part of a single system, or at least directly related systems, situates the state in the community of nations. It makes clear the responsibility of state agencies for securing compliance with international law, including respect for human rights. It recognises the role of international law in defining the scope of a state’s authority, and entails constitutional rules establishing the status of rules of international law relative to other rules of the legal system. However, it restricts the sovereignty of the state, and may accordingly limit the capacity of the internal political process to oversee the extent of the state’s international obligations and domestic powers.
In the United Kingdom, the common law recognises customary international law as a direct source of rules in municipal law.
However, the law adopts a ‘dualist’ stance in respect of treaties<, including human rights treaties: they have no direct effect in national law in the absence of legislation to transform them into rules of the municipal legal order. This dualism is qualified by the fact that courts will often use treaties as aids in deciding questions of municipal law, albeit not as a source of law in their own right. Before the Human Rights Act 1998 came into force on 2 October 2000, treaties could be used in several different ways.Treaties are used as an aid to interpreting legislation that is intended to give effect to treaty obligations, and to resolve ambiguity in other legislation. The United Kingdom’s international treaty obligations are also relevant to the development of the common law. In administrative law, courts would have regard to relevant treaty obligations binding the state as part of the background against which the rationality of the actions or rules would be assessed. The more significant the right asserted, or the interference, the stronger would be the justification needed for interfering with it. Where the threatened right is fundamental, such as the right to life, judges have said that they would subject any interference with it to particularly anxious scrutiny. Furthermore, the Court of Appeal has in principle accepted that, as the High Court of Australia held in Minister for Immigration and Ethnic Affairs v Teoh, entering into a human rights treaty may give rise to a legitimate expectation, enforceable through judicial review, that state agencies will comply with the obligations imposed by the treaty. However, this will depend on how recently the treaty was entered into, and statutes enacted in the interim may negate the expectation. Finally, human rights might be directly enforceable so far as they give rise to general principles of European Community law that need to be given effect in litigation before national courts. This interpretative and developmental role for treaties in municipal law is justified by rebuttable presumptions that the legislature does not intend to legislate inconsistently with the state’s obligations in international law, and that the courts are agencies of the state and have their own responsibility for ensuring that the state complies with its international obligations.
Nevertheless, international and municipal law in the United Kingdom are generally seen as separate systems inhabiting different, though adjacent, spheres. The separation of international and municipal law can be justified by reference to two considerations. The first, derived from political theory, is concerned with self-determination, democracy, and accountability. It would be wrong for national law to be subject to alteration by events on the international plane over which the municipal legislature has no control. Dualism protects national self-determination and political processes against legislation by treaty. The second consideration derives from constitutionalism,  particularly the principle of the rule of law. While the content of the rule of law is controversial, as a minimum it demands that it should be possible to say with reasonable confidence whether or not a specified norm forms part of the municipal legal system. This should be achieved by principles laid down in the constitution, or by rules validly made for that purpose under the constitution.
The democratic justification for national sovereignty in law-making is powerful. The reference to international law presents a problem of legitimacy in national constitutional law. In a democracy, why should sources external to the state, over which the state and its electorate have no control or even much influence, help to shape the meaning of the Constitution for a free and sovereign people? Nevertheless, international and national factors may militate in favour of a limited relaxation of national domination over the content of municipal law. Some standards in international law (jus cogens) are regarded as sufficiently important to make it proper to enforce them without the need for national agreement, and notwithstanding any agreement to the contrary. Within a state, historical circumstances may make people feel a need to turn to international law as a source of standards of fairness and humanity to supplement local traditions which may be tainted with bias, discrimination or faction. South Africa in the 1990s was such a state, about which more will be said below. In the next section, it will be argued that allowing international law rules to operate directly as norms of municipal law may be legitimate under certain conditions. One is that the state’s constitution specifies both the circumstances in which a norm of international law forms part of the legal system, and the status of such a norm relative to other norms within the system.
The section after next examines the reverse effect, the influence of municipal law over international law. When interpreting rights, or restricting the freedom of states to interfere with those rights, some international organs are influenced by the extent to which a state’s behaviour violates the shared norms of constitutional law in the area of the parties to the treaty. The result may be to inhibit the development of important norms of international human rights law, to the detriment of citizens in the states. This raises questions about the legitimacy of approaches to interpreting human rights treaties which form the second focus of the essay.
International human rights law influences constitutional law in several different ways, depending on the historical and legal setting in which the constitution operates. For example, the first ten amendments to the United States of America Constitution, forming its Bill of Rights, were written many years before the twentieth-century rise of human rights. The Bill of Rights has influenced constitutions in numerous other states and has contributed to the attitudes that informed the development of a human rights ethos in international law. Admittedly there has been relatively little feedback from international law into United States constitutional law. Yet the Constitution gives a significant place to treaties as a source of law, under strict conditions.
The President of the United States of America has power to make treaties, by and with the advice and consent of the Senate (with the concurrence of two-thirds of the Senators present). Article VI of the Constitution provides that treaties made ‘under the Authority of the United States’ form, together with the Constitution itself and federal legislation, ‘the supreme Law of the Land; and the Judges in every state shall be bound thereby, any Thing in the Constitution or Laws of any state to the Contrary notwithstanding’. There is a hierarchy within this supreme Law of the Land: treaties are subject to the substantive provisions of the Constitution, and treaties can neither extend the powers nor limit the duties of any branch of Government under the Constitution. What is more, only self-executing treaties (that is, those which require no intervention by the legislature in order to determine how the obligations under the treaties are to be carried into effect) have the force of law without Congressional legislation to make them part of municipal law.
On the other hand, international treaties may have the effect of overriding the laws or policies of the states of the United States, and so affect the balance between federal and state authorities notwithstanding the Tenth Amendment. Thus in order to discharge its obligations under the Migratory Bird Treaty the federal authorities were able to acquire powers by Act of Congress in relation to a matter which would otherwise have fallen exclusively within the jurisdiction of the states, notwithstanding Article X of the Constitution.
Even an executive agreement with another sovereign state which does not comply with the requirements of Article II, § 2 of the Constitution, and which has not been given effect in municipal law by an Act of Congress, may generate executive powers that the courts will recognise. For example, an exchange of correspondence, approved by Congress, embodying the recognition of a foreign state, and accepting obligations towards it, may effectively allow the federal authorities to enforce those obligations within the states of the United States despite the Tenth Amendment.
The advantage of this, from the perspective of municipal law, is that it prevents the federal executive or legislature from using the treaty-making power to extend their constitutional powers, avoiding the normal controls on constitutional amendment under Article V of the Constitution, but allows the federal authorities to conduct international relations. Every sovereign state must strike a compromise between the objectives of maintaining the constitutional division of powers and democratic accountability within the state and allowing the executive to function reasonably efficiently on the international stage. The form that the compromise takes will be reflected in the law governing the relationship between constitutional and international law in the state. The approach of the United States courts to Article II, § 2 of the Constitution and the Tenth Amendment is such a compromise.
The Commonwealth of Australia offers a contrasting illustration of the working of the compromise. Australia has a dualist constitution, like that of the United Kingdom. As in the United Kingdom, customary law can form part of municipal law, generating some human rights, particularly in relation to the development of native title to land. The Commonwealth Constitution, unlike the United States Constitution, makes no express provision for treaties to form part of municipal law. However, the Commonwealth Parliament’s limited powers include a power to legislate in the field of external affairs. The High Court of Australia has held that this allows the Parliament to legislate to give effect to international treaties to which Australia is party. As any Australian state legislation which is incompatible with an Act of the Commonwealth Parliament made within its powers is ineffective to the extent of the incompatibility, the power to legislate for international relations sometimes extends the powers of the Commonwealth into areas which would otherwise fall within the exclusive power of the states.
Extending the powers of federal authorities by reference to treaties is controversial. The decisions in these cases alter the balance between the states and the federal authorities. To some extent this can be justified by reference to the role of the Australian Senate in passing the necessary legislation. In theory, the Senate (like the United States Senate) represents the interests of the states. Where the Senate has approved a Bill, the resulting Act can be seen as consistent with the constitutional principle of respect for the separate interests and dignity of states. Nevertheless, to some extent this is a fiction. Senate elections are conducted on the same political party lines as elections to the Australian House of Representatives, and the party in government often commands a majority in the Senate as well as in the House of Representatives. Where, in addition, the federal executive is entitled to conclude treaties without reference to the Australian Parliament or the states, the reality is that the federal government controls the entire process from treaty-making to legislation, and could (subject to the possibility of judicial review if the courts concluded that a government had entered into a treaty purely for the purpose of evading constitutional controls on federal legislative competence) extend its legislative authority by international treaty.
These examples are drawn from federal states, but even in unitary states a dualist approach may be useful to ensure that the executive cannot change municipal law without the checks and balances normally provided by the legislature. As well as maintaining democratic accountability for law-making, the relationship between municipal and international law affects the capacity of the state to maintain the rule of law, according to which laws must comply with the requirements of the Constitution, both procedural and substantive. Compliance with the rule of law is a major component of the constitutional legitimacy of any legal rule. However, constitutional legitimacy is not the only type of legitimacy for which legal arrangements may be tested. A rule, decision or act may be legitimate or illegitimate not only:
(a) constitutionally, in the sense of being accepted by the ultimate arbiters of constitutionality as consistent (or not inconsistent) with relevant constitutional norms and values;
but also by reference to other value systems:
(b) politically, in the sense of being accepted by some or all participants in political discourse as justified (or at any rate not condemnable) by reference to principles concerning the proper allocation or use of state power to which those participants are prepared to subscribe (often including, but not limited to, a model of democracy); or
(c) morally, in the sense of being accepted by those who hold particular moral standards as justified (or not condemned) under those standards; or
(d) consequentially, in the sense of being calculated to promote some objective which is regarded as being specially valuable (one possible object, although by no means the only or a necessary one, being to maintain the stability of the Constitution); or
(e) procedurally, in the sense of being entitled to special respect by virtue of having been arrived at by way of a specially approved decision-making procedure (which may or may not be connected with a conception of democracy).
These dimensions of legitimacy overlap, providing alternative or cumulative tests for the legitimacy of any legal arrangement. None of them is intrinsically more important or valuable than the others. A judgment as to their relative importance will be made in each case, taking account of the context as a whole, including the constitutional rules applicable at the time and the political theory on which the state is constructed. It is therefore possible to say that particular rules establishing, say, a monist relationship between public international law of human rights and municipal law are:
(a) constitutionally legitimate, in that they are compatible with fundamental principles of the constitution;
(b) politically illegitimate, in that they unnecessarily restrict the scope for national self-determination and democratically responsive law-making by state institutions in ways which cannot be reconciled with locally prevailing notions of national sovereignty and democracy;
(c) morally legitimate, in that international law provides for morally desirable standards of protection for human rights which are morally preferable to those in force under municipal law;
(d) consequentially legitimate, in that the rules foster acceptance of the state and recognition of its constitution and government in the international community, as well as producing (perhaps) desirable outcomes in particular cases; and
(e) procedurally illegitimate, in that the rules permit municipal law to be directly affected by international law-making processes outside the control of the national legislature.
Of course, all this will depend on the precise content of the rules themselves. For example, political and procedural legitimacy may depend on the constitutional provision (if any) for review and approval of treaties by the legislature before they become binding, such as Article II § 2 of the United States Constitution: similarly section 231(2) and (3) of the Constitution of the Republic of South Africa, 1996, provides that 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 of a technical, administrative or executive nature, or one which does not require ratification or accession. Political and procedural legitimacy may also depend on an acceptable procedure being followed before a treaty binding in international law can become part of municipal law. For example, section 231(4) of South Africa’s 1996 Constitution provides:
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.
One must keep all the dimensions of legitimacy in mind. Making a judgment about the overall legitimacy of a rule (as opposed to its constitutional, political, moral, consequential or procedural legitimacy) involves assessing the significance of each dimension relative to the others, in the context in which the rule is to operate.
Looking solely at constitutional legitimacy, the making of municipal law through international treaties is constitutionally acceptable only if the process by which the state may bind itself in international law is itself governed by rules of the constitution. However, where (as in the United Kingdom) the constitution lays down no procedural or substantive legal rules to govern the state in entering into international obligations, both the process and the substance of the obligations are outside constitutional control. Law made in that way would not be constitutionally legitimate, even if it was supported by a majority in the legislature or the state and so (at least arguably) was politically legitimate. It would be derived from no constitutional standards, and would not meet formal rule of law requirements in a constitution committed to the rule of law.
The great advocate of monism, Hans Kelsen, recognised this. He made the controversial claim that only a monist approach to international law was logically possible. However, he did not argue that norms of international law must be regarded as hierarchically superior to norms of municipal law within a legal system which incorporated them all. Norms of international law might, he thought, be subject to some norms of municipal law, particularly constitutional law. Whether and to what extent this was so in any system would depend on the terms of the constitution. Monism requires only that legal norms be part of a single system of law. It says nothing about the hierarchical ordering of the norms within that unified system. Thus a monist constitution, such as that of France’s Fifth Republic, may provide that the international legal norms forming part of French law shall have a status superior to that of ordinary French law but inferior to that of the Constitution itself.[27 ]
Within such a system, the constitution remains the arbiter of which norms are legal, and of whether a particular norm takes precedence over another that appears to be inconsistent with it.
Kelsen’s approach has twin advantages for monists who are concerned about the legitimacy of applying international law in municipal fora. First, his model of monism accommodates the rule of law by ensuring that the Constitution will contain what Hart called ‘rules of recognition’ identifying those norms of international law which form part of municipal law, and the relative status of norms derived from different sources. Second, Kelsen’s model makes it possible to maintain democratic legitimacy by drafting the constitutional rules of recognition so that approval by a democratically accountable law-making organ of the state is a precondition to a norm of international law taking effect within the system. Constitutions will not necessarily contain such provisions, but where they exist they effectively answer the democratic critique of monism.
This is particularly important where the norms of international law derive from human rights treaties, which purport to confer rights on, or recognise the rights of, individuals rather than states, or are norms of international humanitarian law. Protection for human rights often depends on making remedies for violations available in municipal tribunals, especially where the treaty in question does not permit an aggrieved individual to petition an international tribunal. The United Kingdom illustrates how dualism can pose a major hurdle to the effective enforcement of individual rights under international law. The strongest form of dualism operated in Scotland, where judges refused to consider international treaties for any purpose until the Court of Session in T, Petitioner brought Scots law into line with the approach in England and Wales and Northern Ireland, where limited use of treaties was permitted, as described above, prior to the coming into force of the Human Rights Act 1998.
The legislature had more freedom than the courts to take account of human rights treaties when considering legislative proposals. However, Parliament had no systematic procedure for scrutinising provisions for compatibility with human rights. Its success in preventing violations by legislation depended, first, on the pre-legislative scrutiny which took place in the sponsoring Department, over which Parliament had no control; second, on briefings provided to parliamentarians by non-governmental organisations such as JUSTICE and Liberty; and, third, on the expertise of the small number of human rights lawyers in Parliament who might contribute to debates.
The new scheme introduced by the Human Rights Act 1998 alters this to some extent. It makes 'Convention rights' part of municipal law in the United Kingdom. 'Convention rights' are those rights arising under the European Convention on Human Rights that are identified in section 1 of the Act and set out in Schedule 1, subject to various other provisions.
Under section 2, the rights are to be interpreted taking into account the case-law of the Convention organs. By section 3, all legislation is to be interpreted so far as possible in a way that is compatible with Convention rights. Under section 6, a public authority (which includes courts and tribunals but excludes Parliament) acts unlawfully to the extent that it does anything which is incompatible with a Convention right, unless its action is required by primary legislation which cannot be interpreted so as to be compatible with Convention rights. Where primary legislation is incompatible with a Convention right, it is not thereby made invalid, but a superior court may make a declaration of incompatibility, which should lead to an amendment of the law. To assist parliamentary scrutiny, section 19 requires a Minister introducing a government Bill to Parliament to make a statement in writing (which is printed on the front cover of the Bill) that in his or her opinion the Bill either does or does not comply with Convention rights. Parliament is to establish a Joint Select Committee of its two Houses to monitor and report on matters relating to human rights in the United Kingdom and report to each House.
This structure preserves the doctrine of parliamentary sovereignty while providing a system for remedying violations of Convention rights in municipal courts and tribunals, whereas previously it would have been necessary for victims of violations to go to Strasbourg for a remedy (after exhausting any domestic remedy which might have been available in municipal law). The 1998 Act, not the Convention, is the source of the authority for municipal tribunals to apply the rights. Even the meaning of the rights in municipal law is not necessarily the same as in international law, as the interpretation of the European Court of Human Rights provisions by Strasbourg organs is persuasive but not binding on municipal tribunals.
However, in practice the separation between municipal and international law is weakened. The Act recognises that the meaning of the Convention rights is dynamic rather than fixed, and acknowledges that the Strasbourg organs have a part to play in driving forward the interpretation of the rights. The Act also makes the Strasbourg case law dispositive of some questions of municipal law, particularly standing to assert Convention rights in proceedings under the Act. Furthermore, the obligation on courts and tribunals under section 2 of the Act to have regard to relevant case-law of the Strasbourg organs will inevitably bring other treaties into play, since the European Court of Human Rights regularly refers to instruments such as the International Covenant on Civil and Political Rights and the Convention on the Rights of the Child, as well as some ‘soft law’, norms or standards which are intended to guide decisions or actions but which are not intended to be legally enforceable, when interpreting provisions in the European Court of Human Rights.
These steps towards a more outward-looking approach to the development of municipal law in the United Kingdom, tentative though they may seem, represent a fundamental shift. Having been taken by Act of Parliament, the steps are procedurally legitimate, and the objective of protecting human rights should ensure that they are also morally legitimate. On the other hand, the change was not necessitated by any major constitutional upheaval. Nor was it the result of widespread public demands for improved protection for human rights. The Human Rights Act 1998 came about through long-term pressure from an influential minority of lawyers, parliamentarians, political scientists and journalists, building on concern about the United Kingdom’s reputation abroad in the light of repeatedly having been held in violation of the European Court of Human Rights in cases reaching the European Court of Human Rights. Had there been a sense that a Bill of Rights was historically, socially or constitutionally necessary (rather than just desirable), it would probably have been a more far-reaching instrument than the 1998 Act, giving greater weight to the international instruments and imposing stronger checks on parliamentary sovereignty. There would also probably have been a more intensive and far-reaching campaign of public consultation to ensure that the rights and methods of enforcement commanded maximum support among the population. Public support is important to make constitutional change in a democracy politically legitimate, and would help to maintain its legitimacy in the face of the possible unpopularity of some consequences of applying the rights and of some of the causes espoused by those who may assert them. Instead, the debate was mainly conducted by an informed élite within a charmed circle. The Labour Government of Mr Blair could claim a democratic mandate on the basis that the idea of a Bill of Rights based on the European Convention on Human Rights formed part of the party's manifesto for the 1997 general election campaign, but the lack of detailed public consultation means that the Human Rights Act 1998 can be said to be for, but not of, the people. Furthermore, the dependence on principles drawn from outside the municipal legal system to make sense of the Convention rights in municipal law presents problems for the rule of law, as the status and proper weight of those principles is not defined with any precision by the Act. It is arguable, therefore, that the scheme of the Act lacks a degree of constitutional legitimacy, due to the absence of effective rules of recognition. Nevertheless, for many commentators the overall balance of legitimacy favours the Act, because of the benefits which it may be expected to confer on people in their dealings with public authorities in the United Kingdom.
In terms of the political, procedural and consequentialist legitimacy of constitutional changes affecting the relationship between international and municipal law, the genesis of South Africa’s Interim Constitution of 1993 and its final Constitution of 1996 provides an instructive contrast. The influence of history on the shape of protection for rights can be seen in the South African Constitutions of 1993 and 1996. Replacing the apartheid era with a new system based on the ideas of equality and majority rule required a major constitutional resettlement, and a sensitive and inclusive approach was needed from political parties and factions which had historically been deeply in conflict with each other. The leaders of the two main parties, Nelson Mandela of the African National Congress and President F W De Klerk of the National Party, provided vital moral as well as political leadership here, building a negotiated consensus which made it possible to marginalise the non-compliant groups, particularly Inkatha and the Afrikaner Resistance Movement (AWB).
When planning the process for building a new constitutional settlement, a two-stage procedure was adopted. First, an Interim Constitution was established to provide a framework for transition to majority rule guided by agreed values. Second, a Constitutional Assembly prepared a final draft, which was enacted by the legislature in accordance with procedures laid down in the Interim Constitution: a special majority was needed but, in the event of it not being obtained, a tie-break procedure, including a referendum on the proposed draft, was laid down.
A constructive and reconciliatory attitude was perhaps an essential condition for achieving fairness and establishing the legitimacy of the Constitution in the eyes of the population as a whole, in the light of the history of South Africa. Sir Sydney Kentridge QC, who became a judge of the Constitutional Court of South Africa, paid tribute in print to the spirit in which the leaders of the country approached the task of constitution-making.[36 ]The process of framing the 1993 Constitution involved applying the expertise of lawyers, political scientists and politicians to the needs and desires of the people. The drafters harnessed modern communications technology: the press, the World Wide Web, electronic mail, the mail, and telephone, to inform the population about the process of negotiation and drafting as it proceeded, and to allow everyone to comment on and contribute to it. A newspaper was published to explain the proposals in detail to the polyphonic population, and made extensive use of imaginative strip cartoons to get the message across to those whose literacy in any language was limited. This approach continued after 1993, as the Constitutional Convention set about producing the final version of the Constitution, which was eventually adopted in 1996.
The Interim Constitution was to be a beacon for the ‘rainbow nation’. It had symbolic importance. The government was trying to establish the conditions for peaceful co-existence of different racial communities in a significantly revised framework of values. The debate over whether or not the Constitution should include a Bill of Rights provides a good example of the tensions that had to be overcome. The minority white community was worried about reprisals being exacted from them, and regarded a constitutional Bill of Rights as an important sign of the new regime’s commitment to equality and security for them. On the other hand, the African National Congress (ANC) was opposed to a justiciable Bill of Rights, fearing that it would limit the capacity of a democratically elected ANC government to introduce radical and important social and economic reforms.
In the end, both the Interim and final Constitutions contained entrenched and justiciable constitutional Bills of Rights.[38 ]The Bill, comprising Chapter 2 of the 1996 Constitution, was drafted in the light of the needs and conditions of South Africa, and motivated by the fact that ‘those who had suffered most under the old regime, and those who were now about to come into power, not least Mr Nelson Mandela, had a deep and genuine desire to ensure that no future government, even their own, should be able to act against any individual or group as the National Party had acted during its 46 years of power.’
The shape of the Bill reflects both South Africa’s internal tensions and its determination to operate as a full member of an international community of nations, complying with all the international obligations and adopting the highest values of a civilised nation. Social as well as individual rights are included, although the drafting clearly recognised the programmatic quality of the former. The selection and framing of the rights were influenced by international human rights instruments and by the constitutions of other states, notably Canada, Ireland and India. This outward-looking approach to fundamental rights extends from the stage of constitutional drafting to that of adjudication and enforcement. The 1993 Constitution made it possible in some circumstances for treaties to take effect in municipal law without the need for legislation, although the 1996 Constitution resiled from this, while still allowing customary international law to form part of municipal law so far as it is not inconsistent with the Constitution or Acts of Parliament. In relation to human rights, the 1993 and 1996 Constitutions both require courts, when adjudicating on matters raising issues of constitutional rights, to have regard to relevant international treaties and decisions as an aid to interpretation. Section 39(1) of the 1996 Constitution, in some respects broader than section 35(1) of the Interim Constitution, provides:
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.
The enthusiasm and efficiency with which the authority to consider international and foreign law has been taken up have made the judgments of the Constitutional Court excellent points of reference for judges in other jurisdictions who want a reliable but concise introduction to the comparative law of human rights, both international and municipal, for guidance in developing the interpretation of rights in their own jurisdictions. The usefulness of the case law of the Court is enhanced by the very high quality of the analysis and reasoning in the judgments themselves.
This express constitutional authority to use both international law and comparative law as guides to the interpretation of the Bill of Rights is unparalleled. It reflects an openness to outside influences which is symbolically as well as practically significant, as it emphasises a decisive break with the political and legal tradition of the previous apartheid regime which had persistently disregarded norms of international law and had developed a siege mentality under external pressure. With all its problems, the new South Africa intended to be a full and responsible member of the international legal community from its inception, and its constitutional arrangements reflected that.
One effect of these constitutional arrangements is to blur the distinction between international and municipal law. As in the United Kingdom under the Human Rights Act 1998, it is impossible to say what municipal law is in relation to human rights issues without reference to international law, and, in the case of South African courts, the case law of other jurisdictions, even where international law does not form part of municipal law. This presents a difficulty for dualist positivists. It is hard to accept that municipal law can be a closed system of rules when the system itself makes norms from other systems, over which it exercises no control, decisive of, or persuasive in, the determination of issues. Although these outside influences are likely to be limited in their application by the Constitution, they are not controlled by it. A mixed system like that now operating in the United Kingdom, and a fortiori that which has been adopted in South Africa, draws its operating rules from a range of sources, not all of which can be said to be legally validated by reference to criteria contained in the national constitution. This closely resembles Kelsen’s model of a monist system in which the application of international law is subject to constitutional constraints.
As noted earlier, this raises questions of constitutional legitimacy, which are emphasised where a Bill of Rights is to be interpreted dynamically. The meaning of rights may develop under the influence of changing international norms and judicial decisions from elsewhere which were not dreamed of at the time when the Bill of Rights was framed. Sometimes the state has little or no control over changes in international law. For example, the European Court of Human Rights adopts a dynamic approach to interpreting the Convention rights, directly affecting the interpretation and application of the rights in municipal systems (such as the United Kingdom under section 2 of the Human Rights Act 1998). Can these problems be addressed in a principled way, or are they fatal to the legitimacy of an open, outward-looking approach to interpretation?
I suggest that the problem of legitimacy, although real, is not insuperable if one bears in mind the different dimensions of legitimacy in constitutional, political, and moral discourse, outlined earlier. In South Africa, the process by which the Bill of Rights was drafted and enacted, together with the rigorous conditions for allowing international law to have effect in municipal law, lay to rest any concerns about political or procedural legitimacy, and do much to offset the consequentialist challenges based on unpopular outcomes in particular cases. In view of the fact that international law and foreign law must be considered by, but are not binding on, municipal courts under section 39 of the Constitution, the constitutional rules of recognition seem sufficiently clear to set at rest concerns based on the rule of law. The substance of the international law rules that are incorporated does not, in the case of human rights law, detract from their moral legitimacy or undermine their constitutional or political legitimacy. It simply underpins the commitments to democratic values, social justice, fundamental human rights, the rule of law, equality before the law, and acceptance of a united and democratic South Africa by the international community which formed the cornerstones of the social, political and constitutional settlement of which the 1993 and 1996 Constitutions form part, as the Preamble to and section 1 of the 1996 Constitution make clear.
On the basis of the South African example, it is safe to conclude that, with appropriate safeguards, any state which is sufficiently committed to the values of the rule of law in municipal and international law, and to democracy and human rights, can legitimately make arrangements to treat international law as part of municipal law. Although the discussion here has concentrated on human rights, similar arguments apply in respect of other areas of law, such as environmental and humanitarian law. But whether or not a state has achieved legitimacy will depend on the constitutional law of each state, and what is true of South Africa may not be true of, for example, the United Kingdom.
If, as suggested above, it is legitimate under certain conditions for international human rights law to affect municipal law directly, is it equally legitimate for national law to affect the content of international human rights law? There can be no doubt that it does so, in a number of ways. First, one would expect the representatives of states who negotiate human rights treaties to be reluctant to agree to binding obligations unless they expect their states to be able to comply with them without major changes to their laws. For example, the United Kingdom would probably have been far more reluctant to sign and ratify the European Convention on Human Rights had the government not been confident that the law in the United Kingdom complied with the Convention rights. Of course, states can be surprised by subsequent developments, particularly when a treaty establishes an independent tribunal with power to interpret the treaty authoritatively and to impose obligations on states to comply with its judgments. The United Kingdom has been caught out frequently by the dynamic interpretation of the Convention by the Strasbourg organs. Nevertheless, the view which states take of their own law is likely to affect the kinds of obligations that they are willing to shoulder in international law.
Second, a state may place reservations when entering into treaties if its own law would make any of the treaty obligations impossible or unduly difficult to comply with, although questions may arise as to the acceptability of a reservation.[45 ]For example, the United States entered a reservation to Article 20 of the International Covenant on Civil and Political Rights, which demands that states prohibit propaganda for war and certain kinds of hate speech, in order to guard against an obligation which would probably have been incompatible with the First Amendment to the Constitution. On ratifying the Covenant, the United Kingdom entered a reservation to the same Article, to the effect that ‘having legislated in matters of practical concern in the interests of public order (ordre public) [the Government] reserve the right not to introduce any further legislation’. In such cases, if the reservation is accepted, the current state of municipal law limits the obligations which the state will be deemed to have under international law, although of course one state’s reservation does not affect the way in which international law affects other states.
Third, there are cases in which an international legal tribunal must ascertain to the best of its ability the municipal law of a state, then apply it to the instant case.
Fourth, customary international law is built on the practices of states, which are to some extent dictated, or at least evidenced by, municipal courts’ decisions. However, this may be limited. For example, torture is a criminal act in international law, and its prohibition
has evolved into a peremptory norm or jus cogens, that is, a norm that enjoys a higher rank in the international hierarchy than treaty law and even ‘ordinary’ customary rules. The most conspicuous consequence of this higher rank is that the principle at issue cannot be derogated from by states through international treaties or special customs or even general customary rules not endowed with the same normative force.
It follows that there are strict limits to the capacity of municipal law to affect the content of customary international criminal law.
The interpretation of human rights treaties, and the European Convention on Human Rights in particular, offers perhaps the most significant field in which national law affects the interpretation of international law. International tribunals increasingly take account of decisions of municipal courts for guidance both as to the content of international custom and as to the interpretation of international treaties. The European Court of Human Rights developed a theory of the ‘margin of appreciation’, designed to take account of the variety of moral opinions among states and to allow the municipal law of states to reflect the moral and social beliefs current in the state. In theory, it helps to offset the potential challenge to the legitimacy of applying international standards to states, based on principles of self-determination and democracy, as discussed in the previous section of this paper. When deciding whether an interference with a right such as the right to freedom of expression under Article 10(1) of the Convention is ‘necessary in a democratic society’ for a purpose permitted under Article 10(2), the Court insists that the national authorities should show that the interference was a proportionate response to a pressing social need. Ultimately the Court has to decide that question, but it gives a degree of respect to the assessment by national authorities of the existence of a pressing social need and the appropriateness of particular responses to it. The leeway allowed to states is widest where there is no established or emerging consensus among European states about the proper way of dealing with the issue in hand. As the Court wrote in its opinion in the Handyside Case, in relation to matters of sexual morality it had proved to be impossible:
to find in the domestic law of the various Contracting states a uniform European conception of morals. The view taken by their respective laws of the requirements of morals varies from time to time and from place to place, especially in our era which is characterised by a rapid and far-reaching evolution of opinions on the subject. By reason of their direct and continuous contact with the vital forces of their countries, state authorities are in principle in a better position than the international judge to give an opinion on the exact content of these requirements as well as on the ‘necessity’ of a ‘restriction’ or ‘penalty’ intended to meet them.
This makes it clear that the Court looks to the municipal law of states for a consensus on the appropriate way of dealing with a particular problem. If this exercise in comparative law discloses no consensus, the Court will defer (within certain limits) to the national authority’s assessment of local needs. Finding consensus is particularly difficult in matters of sexual morality. For example, the United Kingdom has been granted a wide margin of appreciation in deciding how, in the public interest, to respond to what its authorities saw as a pressing social need to protect public health and public morality against sado-masochism and blasphemous videos. On the other hand, the ‘margin of appreciation’ seems to be less extensive where particularly fundamental rights or intimate interests are involved. In Z v Finland, the majority of the Court (Judge De Meyer dissenting on this point) was prepared to accept that the national authorities had a margin of appreciation:
in striking a fair balance between the interest of publicity of court proceedings, on the one hand, and the interests of a party or a third person in maintaining the confidentiality of such data, on the other hand. The scope of this margin will depend on such factors as the nature and seriousness of the interests at stake and the gravity of the interference …
In practice, the use of the ‘margin of appreciation’ is less than satisfactory, not least because the Court does not always consider how carefully the national authority has considered the matters which need to be balanced before arriving at a conclusion. Even when being properly operated, the ‘margin of appreciation’ is a controversial doctrine, because it has the effect of allowing variable levels of protection for rights as between different states. In his partly dissenting opinion in Z v Finland, Judge De Meyer attacked the use of the margin of appreciation doctrine in a case where human rights (as opposed to, for example, the length of a sentence for a crime) are concerned. He considered that the Court:
has already delayed too long in abandoning this hackneyed phrase and recanting the relativism it implies … But where human rights are concerned, there is no room for a margin of appreciation which would enable the states to decide what is acceptable and what is not. On that subject the boundary not to be overstepped must be as clear and precise as possible. It is for the Court, not each state individually, to decide that issue, and the Court's views must apply to everyone within the jurisdiction of each state.
It is noteworthy that the Human Rights Committee, when adjudicating on a communication from an individual under the First Optional Protocol to the International Covenant on Civil and Political Rights, has set its face against allowing national moral viewpoints to dictate the scope of the right to privacy under Article 17, as to do so would effectively undermine protection for privacy which is a right particularly likely to fall foul of local moral codes. The same could be said of other rights in respect of which the European Court of Human Rights has allowed a wide ‘margin of appreciation’ where morality is in issue, such as the right to freedom of expression under Article 10 of the European Court of Human Rights.
For the moment, however, the ‘margin of appreciation’ remains significant in the case law of the Court, particularly where the legal systems of states do not disclose a common approach to the problem in hand. Naturally, it would be wrong to over-estimate the significance of municipal law for the rights under the Convention. An emerging consensus can be found in the reports of supra-national or regional bodies such as the Council of Europe or the European Parliament. This can be seen in relation to legal recognition of the status of post-operative transsexuals, in relation to which a growing consensus, fuelled by reports from international and supra-national bodies, is gradually eroding support on the Court for a ‘margin of appreciation’. This process gradually limits the scope of a state’s freedom to interfere with the relevant rights, regardless of the law of any one or more states. Nevertheless, the law of a state can affect the extent to which the state will be regarded as being justified in interfering with a person’s Convention rights.
In addition, there are other ways in which the law of a state can affect the content of a person’s rights against the state in international law. The European Court of Human Rights has decided that generally states are free to apply their own rules of evidence, as long as they do not deprive people of a fair trial. Until recently, the Court has been willing to discharge its responsibility to decide, ultimately, whether a person has received a fair trial for the purpose of Article 6(1) of the Convention. However, in a very recent case the Court seems to have abrogated its responsibility. Khan v UK concerned a defendant who had been convicted of a drug-trafficking offence solely on the basis of evidence derived from an audio bugging device which the police had attached to the wall of a house in violation (as the Court held) of his right to respect for private life under Article 8 of the Convention. The question arose whether he had received a fair trial, as required by Article 6(1). Instead of deciding the question for itself, the majority of the Court, by six votes to one, simply deferred to the judgment of the English courts on the question of fairness. The English courts had considered the matter in the context of section 78 of the Police and Criminal Evidence Act 1984, which gives a court a discretion to exclude evidence if it appears ‘that, having regard to all the circumstances, including the circumstances in which the evidence was obtained, the admission of the evidence would have such an adverse effect on the fairness of the proceedings that the court ought not to admit it’. As the English courts had fully considered the application of section 78, concluding that the trial judge had been entitled to decide that the admission of the evidence would not make the trial so unfair that it ought to be excluded, the Court was prepared to accept that the defendant’s right to a fair trial under Article 6(1) had not been violated. Yet the test applied by the English courts is manifestly different from the one that the European Court of Human Rights should have been applying under Article 6(1), namely whether the defendant had been deprived of a fair trial within the meaning of the Strasbourg case-law. By allowing what is, in effect, a renvoi from international law to English law, the Court deprived the notion of a fair trial under Article 6(1) of any substantial content independent of English law. The majority failed to have regard to the Convention case-law which treats ‘fair trial’ as an autonomous concept, with a meaning independent of that which is given to it in any state.
The majority was thereby prevented (or perhaps saved) from tackling the really thorny issue which had never previously arisen before the Court: in view of the obligation of states under Article 1 ‘to secure to everyone within their jurisdiction the rights and freedoms defined in Section 1 of this Convention’, could it be said that a state had given a fair trial to someone convicted entirely on evidence obtained in violation of his right under Article 8 (which forms part of Section 1 of the Convention)? In a dissenting opinion, Judge Loucaides answered this substantial question in the affirmative. He went on to observe:
I cannot see the relevance to our question of the approach of the domestic courts as regards the question of fairness of admitting the evidence in issue, given that under English law the concept of ‘fairness’ as regards the relevant test of admissibility of evidence, was never incompatible with illegality … What is more, according to the English law, there is nothing unlawful about a breach of privacy like the one which occurred in the present case.
Indeed, he might have added, the majority had acknowledged this point when holding that there had been a violation of the defendant’s right to an effective remedy, under Article 13, for the violation of his right under Article 8.
This points in the direction of an answer to the question raised at the start of this section: if it is legitimate under certain conditions for international human rights law to affect municipal law directly, is it equally legitimate for national law to affect the content of international human rights law? The answer, it seems, has to be ‘no’, at least as regards human rights standards. Just as the notion of an international crime depends on peremptory norms rather than common practice, the protection of human rights should not depend on purely local opinion or practice. The 'margin of appreciation' in the law of the European Convention on Human Rights has some value in helping to overcome the strains of commitment which a state may feel to a Convention mechanism which significantly restricts the state's freedom to legislate for its own people, thus fostering loyalty to the Convention system as a whole. However, the ‘margin of appreciation’ has its own dangers. When applied in assessing states’ purported justifications for interfering with rights, it undermines the uniformity of protection for the rights within the Council of Europe. When deference to local opinion is extended to the content of the rights themselves, as happened in relation to the right to a fair trial in Khan v UK, it threatens to deprive the right of any substance whatever.
This indicates that, for human rights protection to be effective at the international level, no one state’s legal system can be allowed to determine the interpretation of the rights or of the permissible grounds for interfering with them. It is of the essence of the structure of international human rights law that rights at the international level should influence the content of municipal law, and that the influence should flow in that direction only. Once the text of a human rights treaty has been agreed, then (subject to any permitted reservation or derogation) the state must be subject to the obligations arising under it, which should not be amenable to alteration by reference to municipal law. This is even true of programmatic social and economic rights. The content of the programs must be set uniformly, even if the speed at which (and the methods by which) the programs are pursued may necessarily vary according to social structures and levels of economic development. It is a fortiori true of civil and political rights.
The first part of this paper suggested that the arguments against the legitimacy of allowing international human rights law to influence municipal law directly can be overcome if the matter is appropriately addressed in the constitutional rules governing the procedure by which international law may become part of municipal law, and particularly if a sufficient basis for the use of international law is provided in constitutional rules of recognition. The second part of the paper argued that, in order to maintain the effectiveness of international human rights law, doctrines which tend to allow municipal law to dictate the impact of international law, such as the ‘margin of appreciation’ or any other principle of deference on the part of international tribunals to municipal law, are to be deprecated, and, on the rare occasions when they are legitimate, their operation must be carefully scrutinised and restricted. They must never be used to allow rights in international law to be deprived of their substantial content or to permit an international tribunal to abrogate its duty to determine whether state action is lawful as a matter of international law, as occurred in Khan v UK.
The limits of these arguments must be stressed. It will not always be either constitutionally or morally legitimate to apply international human rights law directly in municipal courts. That will depend on the content of the human rights law, and on the constitution of the state. Nor is it ever improper for a state to adopt a more liberal interpretation of a right than is enforced in international law. International human rights form a base below which states are not to fall. To that extent, it is proper for an international tribunal to consider what can properly be regarded as the floor below which protection for rights ought not to be allowed to drop in the states within its system. However, it is never permissible for a state to be permitted to drop below that floor merely because the organs of the state have considered where the floor should be laid, if (in the opinion of the international tribunal) the state got it wrong. Such latitude would undermine the coherence of human rights, the principle of pacta sunt servanda as it operates between states which enter into treaties in order to achieve a consistent standard of protection for all their inhabitants, and ultimately the legitimacy of international human rights law itself.
A v United Kingdom  ECHR 85; (1998) 27 EHRR 611 16
Airlines of New South Wales Pty Ltd v New South Wales (No 2)  HCA 3; (1965) 113 CLR 54 6
B v France (25 March 1992) (ser A) 232–C 20
Bugdaycay v Secretary of State for the Home Department  UKHL 3;  AC 514 2
Commonwealth v Tasmania (Tasmanian Dam Case) (1983) 158 CLR 1 2, 6
Cossey v The United Kingdom  ECHR 21; (1990) 13 EHRR 622 20
Foster and Elam v Neilson  USSC 16; 27 US 253 (1829) 4
Handyside v United Kingdom 24 ECHR Ser A (1976) 18, 19
John v MGN Ltd  QB 586 2
Kaur v Lord Advocate 1981 SLT 322 10
Khan v UK (12 May 2000) Eur Ct HR App No 35394/97 20, 21, 22
Koowarta v Bjelke-Petersen  HCA 27; (1982) 153 CLR 168 2, 5, 6
Laskey, Jaggard and Brown v UK  ECHR 4; (1997) 24 EHRR 39 18
Mabo v Queensland (No 2) (1992) 175 CLR 1 5
McCann v United Kingdom  ECHR 31; (1996) 21 EHRR 97 16
Minister for Immigration and Ethnic Affairs v Teoh  HCA 20; (1995) 183 CLR 273 2
Missouri v Holland  USSC 87; 252 US 416 (1920) 5
Moore v Secretary of State for Scotland (1985) SLT 38 10
National Council for Gay and Lesbian Equality v Minister for Justice 1999 (1) SA 642 14
Osman v United Kingdom  1 FLR 193 16
Polyukhovich v Commonwealth  HCA 32; (1991) 172 CLR 501 6
Prosecutor v Furundzija ICTY (2000) document IT, 95-17/1 17
Queensland v Commonwealth  HCA 36; (1989) 167 CLR 232 3
R v Bow Street Metropolitan Stipendiary Magistrate; Ex parte Pinochet Ugarte (No 3)  UKHL 17;  2 WLR 827 17
R v Burgess; Ex parte Henry  HCA 52; (1936) 55 CLR 608 6
R v Director of Public Prosecutions; Ex parte Kebilene  3 WLR 175 2
R v Khan (Sultan)  UKHL 14;  AC 558 20
R v Ministry of Defence; Ex parte Smith  EWCA Civ 22;  QB 517 2
R v Secretary of State for the Home Department; Ex parte Ahmed and Patel  INLR 570 2
R v Secretary of State for the Home Department; Ex parte Brind  UKHL 4;  AC 696 2
Rees v UK  ECHR 11; 9 EHRR 56 20
Richardson v Forestry Commission  HCA 10; (1988) 164 CLR 261 3, 6
Saloman v Commissioners of Customs and Excise  2 QB 116 2
Serbian and Brazilian Loans Cases  PCIJ (ser A) 20 17
Sheffield and Horsham v UK  ECHR 69; 27 EHRR 163 20
State v Makwanyane 1995 (3) SA 391 14
State v Williams 1995 (3) SA 632 14
T, Petitioner 1997 SLT 724 10
Toonen v Australia 1 IHRR (1994) 97 6, 19
Trendtex Trading Corporation v Central Bank of Nigeria  QB 529;  2 WLR 356 1
United States v Belmont  USSC 98; 301 US 324 (1937) 5
Western Sahara Advisory Opinion  ICJ Rep 12 5
Wingrove v UK  ECHR 60; (1996) 24 EHRR 1 18
X, Y and Z v UK  ECHR 20; (1997) 24 EHRR 143 20
Z v Finland  ECHR 10; 25 EHRR 371 19
Human Rights (Sexual Conduct) Act 1994 (Cth) 7
Human Rights Act 1998 (UK) 2, 11, 13, 16, 17
Police and Criminal Evidence Act 1984 23
War Crimes Act 1945 (Cth) 7
Convention on the Elimination of All Forms of Racial Discrimination 1965 9
Convention on the Rights of the Child 1989 11
European Convention for the Protection of Human Rights and Fundamental Freedoms 1950 10, 12, 16, 17, 21
International Covenant on Civil and Political Rights 1966 9, 11, 17
Optional Protocol to the International Covenant on Civil and Political Rights 1966 19
UNESCO Convention Concerning the Protection of the World Cultural and Natural Heritage 1972 6
[*] Professor of Jurisprudence in the University of Birmingham, England, and Legal Adviser to the Joint Select Committee on Human Rights in the United Kingdom
 See Trendtex Trading Corporation v Central Bank of Nigeria  QB 529.
 See the list given by Lord Bingham CJ in his maiden speech as a member of the House of Lords during a debate on constitutional reform: House of Lords Official Report, 3 July 1996, vol 574, col 1466.
 Saloman v Commissioners of Customs and Excise  2 QB 116; R v Secretary of State for the Home Department; Ex parte Brind  UKHL 4;  AC 696; M Hunt, Using Human Rights Law in English Courts (1996) esp ch 1.
 John v MGN Ltd  QB 586.
 R v Ministry of Defence; Ex parte Smith  EWCA Civ 22;  QB 517.
 Bugdaycay v Secretary of State for the Home Department  UKHL 3;  AC 514.
  HCA 20; (1995) 183 CLR 273.
 R v Secretary of State for the Home Department; Ex parte Ahmed and Patel  INLR 570.
 R v Director of Public Prosecutions; Ex parte Kebilene
 3 WLR 175 affirmed on this point  UKHL 43;  3 WLR 972.
 See L Betten and N Grief, EU Law and Human Rights (1998) esp ch 4; P Alston and J H H Weiler, ‘An “Ever Closer Union” in Need of a Human Rights Policy: The European Union and Human Rights’ in P Alston, M Bustelo and J Heenan (eds), The EU and Human Rights (1999) 3.
 See the discussion on these matters in the High Court of Australia in Koowarta v Bjelke-Petersen  HCA 27; (1982) 153 CLR 168, Commonwealth v Tasmania (Tasmanian Dam Case) (1983) 158 CLR 1, Richardson v Forestry Commission  HCA 10; (1988) 164 CLR 261 and Queensland v Commonwealth  HCA 36; (1989) 167 CLR 232; and see D Feldman, ‘Human Rights and International Law and Institutions’ in American Bar Association (ed), Common Law, Common Values, Common Rights (2000) 86.
 I take constitutionalism to mean a commitment on the part of officials of the system to use constitutional values and arguments as a threshold test for the acceptability of decisions, actions or rules.
 For discussion, see J Raz, ‘The Rule of Law and its Virtue’ in J Raz, The Authority of Law: Essays on Law and Morality (1979) 210; J Jowell, ‘The Rule of Law Today in J Jowell and D Oliver (eds), The Changing Constitution (3rd ed, 1994) 57.
 See R Jennings and A Watts (eds), Oppenheim’s International Law (9th ed, 1992) 7–8; L Henkin, ‘Human Rights and State “Sovereignty”’ (1995/96) Georgia Journal of International and Comparative Law 31, 37; L Henkin, R C Pugh, O Schachter and H Smit, International Law Cases and Materials (3rd ed, 1993) 92–94.
 US Constitution, art II § 2. No state may enter into a treaty: art I § 10.
 Reid v Covert  USSC 73; 354 US 1, 16–17 (1957) (Black J). See generally on this subject L Henkin, Foreign Affairs and the Constitution (2nd ed, 1972) passim.
 Foster and Elam v Neilson  USSC 16; 27 US 253 (1829).
 Article X of the Bill of Rights reads: ‘The powers not delegated to the United States by the Constitution, or prohibited by it to the States, are reserved to the States respectively, or to the people.’
 Missouri v Holland  USSC 87; 252 US 416 (1920) (enforcement of the Migratory Bird Treaty Act (3 July 1918)).
 United States v Belmont  USSC 98; 301 US 324 (1937) (recognition of the USSR).
 Mabo v Queensland (No 2) (1992) 175 CLR 1. The High Court’s use of international law in this case has been criticised. See S Scott, ‘The High Court’s use of the Western Sahara Case in Mabo’ (1996) 45 International and Comparative Law Quarterly 923; cf S Pritchard, ‘Native Title from the Perspective of International Standards’  AUYrBkIntLaw 4; (1997) 18 Aust YBIL 127.
 Constitution, s 51(xxix).
 See ss 107, 108 and 109 of the Constitution; Koowarta v Bjelke-Petersen  HCA 27; (1982) 153 CLR 168 on the validity of the Racial Discrimination Act 1975 (Cth) enacted to give effect to the International Convention for the Elimination of All Forms of Racial Discrimination; Commonwealth v Tasmania (Tasmanian Dam Case) (1983) 158 CLR 1 on World Heritage Properties Conservation Act 1983 (Cth) and Richardson v Forestry Commission  HCA 10; (1988) 164 CLR 261 on Lemonthyme and Southern Forests (Commission of Inquiry) Act 1987 (Cth), both enacted to give effect to the UNESCO Convention Concerning the Protection of the World Cultural and Natural Heritage, UNESCO General Conference, 17th Session, 16 November 1972, Paris. See also Human Rights (Sexual Conduct) Act 1994 (Cth), responding to Toonen v Australia, Comm No 415/1990, Human Rights Committee Views, 26 March 1992, UN Doc A/47/40, supp 40, 325. Note that the external affairs power does not depend on there being relevant treaty obligations: see, eg, Polyukhovich v Commonwealth  HCA 32; (1991) 172 CLR 501 on the War Crimes Act 1945 (Cth) as amended by the War Crimes Amendment Act 1988 (Cth).
 See R v Burgess; Ex parte Henry  HCA 52; (1936) 55 CLR 608. Cf Airlines of New South Wales Pty Ltd v New South Wales (No 2)  HCA 3; (1965) 113 CLR 54, 85 (Barwick CJ); Koowarta above n 23, 200 (Gibbs CJ), 217 (Stephen J), 231 (Mason J), 260 (Brennan J).
 R Barker, Political Legitimacy and the State (1990) makes the point that it is worthwhile to distinguish between political legitimacy and other forms of legitimacy, not least because a study of political legitimacy helps to answer some fundamental questions in political theory and political science about the nature of the sense of obligation to obey state commands. As explained below, I have some reservations about the usefulness of Barker’s approach outside political science. In particular, I have distinguished in the past between political and constitutional legitimacy: D Feldman, ‘The Left, Judicial Review and Theories of the Constitution’ in W Watts Miller (ed), Socialism and the Law, 49 Archiv für Rechts- und Sozialphilosophie (1992) 71, 74. For the purpose of discussing the legitimacy of constitutional change, other factors may be relevant.
 H Kelsen, The Pure Theory of Law (1967) 328–347.
 Article 55 of the 1958 Constitution of the Fifth French Republic provides that treaties duly ratified or approved are to prevail over legislation, as long as they are reciprocally applied by the other party. By art 54, a treaty may be referred the Conseil Constitutionnel for a decision about its compatibility with the Constitution.
 H L A Hart, The Concept of Law (2nd ed, 1994) 94–96, 100–110.
 It is significant that many important human rights treaties, including the International Covenant on Civil and Political Rights, art 2(3), the Convention on the Elimination of All Forms of Racial Discrimination, art 6, and the Convention on the Elimination of All Forms of Discrimination against Women, art 2(c), provide for a right to a remedy in national law despite offering no, or no compulsory, procedure for awarding remedies in international law. See D Shelton, Remedies in International Human Rights Law (1999) 15–37, 138–182.
 Kaur v Lord Advocate 1981 SLT 322; Moore v Secretary of State for Scotland 1985 SLT 38.
 1997 SLT 724.
 See C Gearty, ‘The United Kingdom’ in C Gearty (ed), European Civil Liberties and the European Convention on Human Rights: A Comparative Study (1997) 65–71, 77–78; P Silk and R Walters, How Parliament Works (4th ed, 1998) 56–60, 64–66; D Kinley, The European Convention on Human Rights: Compliance without Incorporation (1993) passim.
 Section 10 of, and sch 2 to the Act provides a special procedure for making amendments to incompatible primary legislation by way of subordinate legislation in certain situations.
 D Feldman, ‘The Human Rights Act 1998 and Constitutional Principles’ (1999) 19 Legal Studies 165, fn 193.
 See H Corder, ‘Towards a South African Constitution’ (1994) 47 Melbourne University Law Review 491–533; H Corder, ‘South Africa’s Transitional Constitution: Its Design and Implementation’  Public Law 291–308.
[36 ] S Kentridge, ‘Bills of Right: The South African Experiment’ (1996) 112 Law Quarterly Review 237, 261: ‘The present constitution is infused with a generosity of spirit which is in part no doubt derived from the characters of the remarkable political leaders who negotiated it.’
 For a full account of the negotiations, see R Spitz and M Chaskalson, Politics of Transition: The Hidden History of South Africa’s Negotiated Settlement (2000).
[38 ] A Cockrell, ‘The South African Bill of Rights and the “Duck/Rabbit”’ (1997) 60 Melbourne University Law Review 513.
 Kentridge, above n 36, 242.
 Compare s 231 of the 1996 Constitution with s 231 of the 1993 Constitution: see D J Devine, ‘The Relationship Between International Law and Municipal Law in the Light of the Interim South African Constitution 1993’ (1995) 44 International and Comparative Law Quarterly 1.
 See, eg, State v Makwanyane 1995 (3) SA 391 (death penalty); State v Williams 1995 (3) SA 632 (whipping); National Council for Gay and Lesbian Equality v Minister for Justice 1999 (1) SA 642 (sexual orientation discrimination).
 See s 231 of the 1996 Constitution. Cf the less clear, and so constitutionally less legitimate, provisions in s 231 of the 1993 Interim Constitution, and the comments of J Dugard, ‘International Human Rights’ in D Van Wyk, J Dugard, B de Villiers and D Davis (eds), Rights and Constitutionalism: The New South African Legal Order (1996) 171, 193 and J Dugard, ‘International Law and the South African Constitution’ (1997) 1 European Journal of International Law 77.
 See E Wicks, ‘The UK Government’s Perceptions of the European Convention on Human Rights at the Time of Entry’  Public Law 438–455.
 See, eg, the extension of rights under ECHR art 2 to cover the system for command and control of anti-terrorist operations in McCann v United Kingdom, Eur Ct HR (ser A) 324  ECHR 31; (1996) 21 EHRR 97; of rights under art 3 to cover parental punishment of children in A v United Kingdom Eur Ct HR  ECHR 85; (1998) 27 EHRR 611; and of rights under art 6 to cover public-policy immunities for the police to liability for negligence in Osman v United Kingdom Eur Ct HR  1 FLR 193.
[45 ] See M Shaw, International Law (4th ed, 1997) 641; D W Greig, ‘Reservations: Equity as a Balancing Factor?’  AUYrBkIntLaw 2; (1995) 16 Aust YBIL 21.
 On ratification by the USA and the reservations and declarations entered, see H J Steiner and P Alston, International Human Rights in Context: Law, Politics, Morals; Text and Materials (2nd ed, 2000) 1029–1047.
 Serbian and Brazilian Loans Cases  PCIJ (ser A) 20, 21; L Henkin, R Crawford Pugh, O Schachter and H Smit, International Law Cases and Materials (3rd ed, 1993) 149–153.
 Prosecutor v Furundzija, Tribunal for Former Yugoslavia, Case No 17–95–17/1-T. See also R v Bow Street Metropolitan Stipendiary Magistrate; Ex parte Pinochet Ugarte (No 3)  UKHL 17;  2 WLR 827.
 R Y Jennings, ‘The Judiciary, International and National, and the Development of International Law’ (1996) 45 International and Comparative Law Quarterly 1, 3.
 See D J Harris, M O’Boyle and C Warbrick, Law of the European Convention on Human Rights (1995) 14 ff.
 (7 December 1976) 24 Eur Ct HR (ser A);  ECHR 5; 1 EHRR 737 .
 See, eg, Laskey, Jaggard and Brown v UK (19 February  ECHR 4; 1997) 24 EHRR 39 on sado-masochism, and Wingrove v UK (25 November  ECHR 60; 1996) 24 EHRR 1 on a blasphemous video refused a certificate by the British Board of Film Classification.
 (25 February 1997) RJD 1997–I, No 31;  ECHR 10; 25 EHRR 371  of the Judgment.
 Toonen v Australia, Comm No 415/1990, Human Rights Committee Views, 26 March 1992, UN Doc A/47/40, supp 40, 325.
 See, eg, Handyside v United Kingdom, above n 51.
 See Rees v UK(17 October  ECHR 11; 1986) 9 EHRR 56; Cossey v UK, Judgment of 27 September 1990 (ser A) 184; X, Y and Z v UK  ECHR 20; (1997) 24 EHRR 143; Sheffield and Horsham v UK (30 July  ECHR 69; 1998) 27 EHRR 163; B v France (25 March 1992) (ser A) 232–C.
 (12 May 2000) Eur Ct HR App No 35394/97.
 R v Khan (Sultan)  UKHL 14;  AC 558, HL, affirming  QB 27.
 See the judgment of the Court –.
 I have explored this issue at greater length in D Feldman, ‘Human Rights Treaties, Nation States, and Conflicting Moralities’ (1995) 1 Contemporary Issues in Law 61.