A human rights act

The courts & the constitution
Michael McHugh

The terms of reference of the National Human Rights Consultation Committee make it clear that, for the foreseeable future, there is no prospect of the Parliament of the Commonwealth introducing legislation for the purpose of amending the Constitution to insert a Bill of Rights. It seems inevitable therefore that any protection of human rights at the federal level will be the product of legislation passed by the Parliament and not a constitutional amendment. The rationale for a legislative, as opposed to a constitutional, Bill of Rights is that it maintains parliamentary sovereignty. Given that a constitutional bill of rights is not presently feasible, the issue is, what form of federal legislation will best serve the cause of protecting human rights against the inroads of federal legislation while at the same time preserving parliamentary sovereignty?

The two legislatures which have introduced human rights legislation in Australia - the ACT and Victoria - have both opted for the so-called dialogue model, which involves the courts drawing the attention of the legislature to legislation which impermissibly affects human rights and provoking a response from the legislature. The dialogue focuses on the consistency of laws with the rights included in the human rights instrument. It requires courts to interpret legislation so far as is reasonably possible with the object of preserving the rights referred to in the human rights legislation. But where it is not possible to interpret legislation to give effect to the human rights, the dialogue model requires the courts to make a declaration of incompatibility and transmit it to the Attorney-General. The legislation then requires the Attorney-General to prepare a response and bring the declaration and the response before the legislature within a fixed period. This machinery creates a mechanism for a dialogue between the judiciary, the executive and the legislature in which the courts inform the executive government of human rights deficiencies in the jurisdiction's legislation and enables first the executive and then the legislature to consider what if, any justifications, support the legislative intrusion into the human rights in question and whether the legislation should be amended to overcome these deficiencies. Hence, one of the purposes of this model of human rights legislation is to promote a 'human rights culture'.

The Australian Capital Territory was the first Australian jurisdiction to enact a dialogue model when it enacted the Human Rights Act 2004 (the ACT Human Rights Act). Two years later, Victoria enacted the Charter of Human Rights and Responsibilities Act 2006 (the Victorian Charter). Subsequently, after public consultation concerning the need for, and the form of, a Charter of Rights, New Matilda, a private organization, has drafted a Human Rights Bill (the New Matilda Bill) which is now in the public domain. Unlike the ACT Human Rights Act and the Victorian Charter, the New Matilda Bill is concerned with federal laws. For that reason, my discussion concerning some of the potential constitutional difficulties arising from the dialogue model will concentrate on that Bill.

Section 4 of the New Matilda Bill enacts: 'The human rights in this Act are exercisable by everyone within Australia's jurisdiction.' Part 3 (ss.10-43) specifies the human rights and freedoms that are exercisable in accordance with the Act. The International Covenant on Civil and Political Rights which Australia ratified in 1980 and the International Covenant on Economic, Social and Cultural Rights which it ratified in 1975 are the source of the human rights which the New Matilda Bill seeks to have enacted into Australian federal law. Sections 11- 36 of the Bill describe the civil and political rights that are exercisable. They include:

• the right to life so that no one may be subject to capital punishment
• protection from torture and cruel, inhuman or degrading treatment
• freedom from slavery
• the right to liberty and security of the person so that no one may be arbitrarily arrested or detained or deprived of liberty except in accordance with procedures established by law
• humane treatment when deprived of liberty
• the separation of accused children from detained persons who are over the age of 18 years
• the right to a fair trial and equal treatment before courts and tribunals
• the presumption of innocence until proved guilty according to law
• compensation for wrongful conviction
• the right not to be tried or punished more than once for the same offence
• the prohibition of retrospective criminal laws
• the equal protection of the law without discrimination
• the right not to have one's privacy interfered with unlawfully or arbitrarily and to be protected against unlawful attacks on honour and reputation
• the right to marry
• the protection of the family
• various rights of children
• the right to freedom of thought, conscience, religion and belief
• the right to peaceful assembly and freedom of association
• the right to freedom of expression and freedom of movement
• the right and the opportunity to take part of the conduct of public affairs
• the right of those who belong to cultural or religious or linguistic minorities to enjoy their culture, practise their religion and to use their language
• the right to asylum
• the right not to be deprived of property arbitrarily
• various rights of indigenous people.

These are fundamental human rights, and s.4 of the New Matilda Bill declares that they are exercisable by everyone in Australia's jurisdiction. Sections 37-41 of the Bill describe the economic and social rights that are exercisable in Australia. They include the right to education, the right to work, the right to an adequate standard of living, the right to the highest attainable standard of physical and mental health and the right to Social Security. However, s.42 limits these economic and social rights by declaring that they "are subject to progressive realisation and that their realisation may be limited by the financial resources available government." As a result, in any proceeding that raises the application of operation of economic and social rights, a court must consider all of the relevant circumstances of the particular case before determining that the provisions of any law or that the acts or conduct of a public authority are incompatible with the Act. Among the circumstances that the court must take into account are the nature of the benefit or detriment likely to be suffered by the person concerned and the financial circumstances and estimated amount of expenditure required to be made by a public authority to act in a manner compatible with human rights.

However, what s.4 describes as human rights are not absolute. Section 10 of the Bill declares:

(1) This Act guarantees the rights and freedoms set down in it subject only such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.
(2)In determining whether a right or freedom may reasonably be limited in this way, all relevant factors should be taken into account including -
(a) the nature of the right; and
(b) the importance of the purpose of the limitation; and
(c) the nature and extent of the limitation; and
(d) the relationship between the limitation and its purpose; and
(e) any less restrictive means reasonably available to achieve the purpose that the limitation seeks to achieve.

There is a considerable body of decisions on the Canadian equivalent of s.10 (1) which is found in Section 1 of the Canadian Charter of Rights and Freedoms. The most recent and authoritative decision on this limitation on a Charter right is Canada (Attorney-General) v Hislop where six members of the Supreme Court of Canada said: 'Under s. 1, the government has the burden to demonstrate that a discriminatory provision is a reasonable limit on a s. 15(1) Charter right. If it meets this burden, the law will be saved as being a demonstrably justified reasonable limit on that right.

The framework for a s. 1 analysis is the well-known Oakes test (see R. v. Oakes). The Oakes test may be formulated as two main tests with subtests under the second branch, but it may be easier to think of it in terms of four independent tests. If the legislation fails under any one test, it cannot be justified. The four tests ask the following questions: (1) Is the objective of the legislation pressing and substantial? (2) Is there a rational connection between the government's legislation and its objective? (3) Does the government's legislation minimally impair the Charter right or freedom at stake? (4) Is the deleterious effect of the Charter breach outweighed by the salutary effect of the legislation?'

Thus, the effect of s.10 is that the rights to which s 4 refer must give way to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society. Section 10 introduces a proportionality test which balances the rational needs of society against the human rights of individuals. The four prong test re-formulated in Canada (Attorney-General) v Hislop is objective and similar to tests that courts in this country, particularly the High Court, already use in the constitutional area. As Pamela Tate SC, the Solicitor-General for Victoria has said of the proportionality test in the human right area: 'This approach exposes the Benthamite flaw of considering human rights as absolute and inalienable and substitutes in its place a reasoned and logical approach to the justification of interferences with human rights.' Whatever model of human rights is adopted in this country, adoption of the s.10 limitation is essential. Together with the objective four pronged test laid down by the Supreme Court of Canada in Hislop, it provides a compelling answer to those opponents of human rights legislation who believe that giving courts the power to determine whether legislation interferes with human rights will make an unelected judiciary the governors of Australia.

However, although s.4 says that the human rights in this Act are exercisable by everyone, it does not expressly say that it is creating rights, and, when the Bill is read as a whole, the better view is that, apart from section 54 which gives a right of action against a public authority, the New Matilda Bill does not create rights or causes of action. Instead, with the exception of federal laws that are incompatible with the rights specified in the Bill, it provides immunity from the operation of laws including the common law that are inconsistent with those rights . If this is the correct construction of s.4, read in the context of the Bill, it would invalidate the effect of any State or Territory law that was inconsistent with the rights exercisable under s.4. Whether that was intended may be doubted. Section 47 declares that Part 5 - which concerns the interpretation of legislation and the declarations of incompatibility - "applies to all Commonwealth laws." This rather suggests that the New Matilda Bill was intended to operate only in the federal sphere. But, as presently drafted, the language of s.4 would not permit State or Territory laws to operate inconsistently with the human rights referred to in Part 3 of the Bill. However, the immunity given in respect of federal laws is more limited. In the context of federal legislation, effect will only be given to the section 4 rights in so far as it is possible to interpret federal legislation consistently with those rights.

Section 49 of the Bill declares:

(1) So far as it is possible to do so, primary and subordinate legislation must be read and given effect in a way which is compatible with human rights.
(2) This section -
(a) applies to primary and subordinate legislation whenever enacted;
(b) does not affect the validity, continuing operation or enforcement of any incompatible primary legislation; and
(c) does not affect the validity, continuing operation or enforcement of any incompatible subordinate legislation if primary legislation prevents the removal of that incompatibility.

Section 51 of the Bill provides:

(1) If a Court is satisfied that a provision of primary legislation is incompatible with a right or freedom set down in this Act, it may make a declaration of incompatibility.
(2) If a Court is satisfied that a provision of subordinate legislation, made in the exercise of a power conferred by primary legislation, is incompatible with a right or freedom set down in this Act, and that the primary legislation concerned prevents removal of the incompatibility, it may make a declaration of that incompatibility.
(3) If a Court is satisfied that a provision of subordinate legislation, made in the exercise of a power conferred by primary legislation, is incompatible with a right or freedom set down in this Act, and that the primary legislation concerned does not prevent removal of the incompatibility, it may invalidate that provision.
(4) A declaration under sub-sections (1) and (2) (a declaration of incompatibility)
(a) does not affect the validity, continuing operation or enforcement of the provision in respect of which it is given; and
(b) is not binding on the parties to the proceedings in which it is made.
(5) A Court must transmit a copy of any declaration of incompatibility to the Attorney-General.
(6) This section applies when a court is exercising jurisdiction in any cause or matter pending before it.

Section 52 then declares:

(1) This section applies if the Attorney-General receives a copy of a declaration of incompatibility.
(2) The Attorney-General must present a copy of the declaration of incompatibility to the House of Representatives within 15 sitting days after the day the Attorney-General receives the copy.
(3) The Attorney-General must prepare a written response to the declaration of incompatibility indicating what action if any is proposed in relation to it and the reasons for that action or non-action, and present it to the House of Representatives not later than 6 months after the day the copy of the declaration is presented to the House of Representatives.

The result of these sections is that, in so far as federal legislation applying to a controversy can be interpreted consistently with a human right specified in Part 3 of the New Matilda Bill, the rights of the parties to that controversy must be determined, in whole or in part, by reference to that human right, if it is relevant to the controversy. However, as I have explained, with the exception of s.54, the New Matilda Bill creates immunities, not rights or causes of action. That means that, with the exception of the right of action given against public authorities by s.54, the plaintiff's cause of action must exist at common law or arise from a statute independently of the Bill. Many persons including myself think that is a great weakness in the New Matilda Bill, as it is in the ACT Human Rights Act and the Victorian Charter. However, the effect of s.4 of the Bill is to prevent any legislation, apart from incompatible federal legislation, modifying the human rights referred to in Part 3 of the Bill in so far as they are relevant to a cause of action or defence. And that is a step forward from the present legal situation.

The ACT Human Rights Act and the Victorian Charter contain similar provisions to the incompatibility provisions of the New Matilda Bill. So does similar legislation in the United Kingdom, although the interpretative provisions of the United Kingdom legislation are more radical than those in the legislation of the Australian Capital Territory and Victoria. New Zealand has an interpretative provision similar to those in the Australian Capital Territory and Victoria and since 2001 has had a declaration of incompatibility mechanism. English Law Lords have spoken approvingly of the interpretative and dialogue provisions in the UK legislation. Lord Steyn has described the Human Rights Act as 'carefully and subtly drafted'; Lord Rodger of Earlsferry has described it as 'beautifully drafted'.

But what may be praised as 'carefully and subtly drafted' in the United Kingdom may sow the seeds for constitutional destruction of similar legislation in Australia. Under the external affairs power conferred by s.51(xx) of the Constitution, the federal Parliament has power to enact legislation that gives effect to human rights conventions and treaties such as the International Covenant on Civil and Political Rights and the International Covenant on Economic, Social and Cultural Rights. As I have already mentioned, these Covenants are the source of the human rights which the New Matilda Bill seeks to have enacted into Australian federal law. The International Covenant on Civil and Political Rights is also the source of the ACT Human Rights Act and the Victorian Charter.

However, unlike the United Kingdom, Australia has a written constitutional separation of legislative, executive and judicial power to which all federal legislation must conform. Section 1 of the Constitution vests the legislative power of the Commonwealth in the Federal Parliament. Section 61 vests the executive power of the Commonwealth in the Queen and declares that it is "exercisable by the Governor-General as the Queen's representative" although in constitutional practice the executive power is exercisable by the government of the day. Section 71 vests the judicial power of the Commonwealth in a Federal Supreme Court to be called the High Court of Australia and such other federal courts as the Parliament creates, and in such other courts as it invests with federal jurisdiction. These three sections are the source of the long accepted view that our federal Constitution incorporates the political doctrine of the separation of powers. As a result 'the Parliament is restrained both from reposing any power essentially judicial in any other organ or body, and from reposing any other than that judicial power in such tribunals' . So the question arises whether the provisions in sections 51 or 52 of the New Matilda Bill, creating a dialogue between the courts and Parliament, are invalid in that they invest a court exercising federal jurisdiction with non-judicial power. Closely allied to this question is whether the issues which would arise under sections 51 and 52 of the New Matilda Bill would involve "matters" within the meaning of sections 75, 76 and 77 of the Constitution. Those three sections empower the Parliament to give jurisdiction to the federal courts and to State courts invested with federal jurisdiction. Federal judicial power can be exercised only in respect of 'matters'. Unless the claim involved in a legal proceeding constitutes a 'matter', as that term has been defined by the High Court, the Parliament cannot give federal jurisdiction to any court.

Any human rights legislation that is inconsistent with the separation of powers doctrine embodied in the Commonwealth of Australia Constitution Act 1900 (UK) will be invalid to the extent of that inconsistency. Although there can hardly be any doubt of the power of the federal Parliament to give legislative effect to the rights enshrined in the International Covenant on Civil and Political Rights and the International Covenant on Economic, Social and Cultural Rights, powerful arguments support the conclusion that the federal Parliament cannot constitutionally give effect to those rights by enacting the dialogue provisions of human rights legislation that is embodied in the New Matilda Bill. Even those who believe that the federal Parliament can constitutionally enact legislation in the form of the dialogue provision acknowledge the force of these contrary arguments. I am afraid I am more pessimistic about the prospects of a dialogue model being upheld by the High Court than those who believe that it would be constitutional if enacted at the federal level. It would not surprise me if the High Court upheld legislation in the form of the dialogue model. There are some persuasive arguments in favour of the constitutionality of such a model. Conversely, it would not surprise me if the High Court held that so much of the dialogue model as required a federal court to make declarations of incompatibility and communicate them to the Attorney-General was invalid. Although it will be a close run thing, I think that the better view is that the High Court will hold that that the incompatibility provisions of the legislation are invalid unless the High Court can be persuaded to adopt a more radical and functional approach to what is judicial power and what is a "matter" for the purpose of Chapter 3 of the Constitution than it has done in the past.

The arguments concerning the constitutional validity of ss 51 and 52 that set up the dialogue procedures of the New Matilda Bill are lengthy, technical, complex and difficult, and their exposition would be out of place in a discussion such as this. I have set out the arguments for and against the constitutional validity of the dialogue model of human rights protection in an Appendix to this paper, which can be read when this paper is published.

I can briefly summarise one of the most significant problems concerning the declaration of incompatibility and a possible answer to it. The New Matilda Bill states that a declaration of incompatibility 'is not binding on the parties to the proceedings in which it is made'. At first sight, this stipulation strongly points against the making of a declaration of incompatibility being an exercise of judicial power. As Deane, Dawson and Gaudron JJ and I pointed out in Brandy v Human Rights and Equal Opportunities Commission, in describing what is meant by judicial power, '[i]t is traditional to start with the definition advanced by Griffiths CJ in Huddart, Parker & Co Pty Ltd v Moorehead' where his Honour said:

I am of opinion that the words 'judicial power' as used in sec 71 of the Constitution mean the power which every sovereign authority must of necessity have to decide controversies between its subjects, or between itself and its subjects, whether the rights relate to life, liberty or property. The exercise of this power does not begin until some tribunal which has power to give a binding and authoritative decision (whether subject to appeal or not) is called upon to take action.

As the judgment of Griffith CJ in Huddart, Parker & Co Pty Ltd v Moorehead shows, the vesting in the tribunal of a "power to give a binding and an authoritative decision" is a strong indicator of judicial power. Conversely, apart from some special cases, its absence is a clear indicator of a power or procedure being non-judicial. Its absence is perhaps an even stronger indicator of the non-existence of a 'matter' within the meaning of Chapter III of the Constitution. Yet the making of a declaration of incompatibility is not binding on the parties to a dispute concerning a human right. However, commentators who argue in favour of the constitutionality of the dialogue model of human rights legislation rely on a number of arguments to counter what at first sight seems a difficult, if not an insuperable, hurdle to a holding that the making of a declaration of incompatibility is an exercise of judicial power.

One of the consequences that flow from the issue of a declaration is that the Bill imposes certain duties on the Attorney-General. But those duties arise from the terms of section 52 of the New Matilda Bill. I can see no reason why a party to a declaration of incompatibility could not enforce the Attorney-General's duty to present a copy of the declaration to the House of Representatives, prepare a written response to it and present the response to that House. Although the Attorney-General's duty is connected to parliamentary proceedings, an order of the court would not interfere with the parliamentary process or interfere with the workings of the Parliament. In Egan v Willis I referred to the traditional view that courts cannot investigate or take action in respect of conduct that "relates only to the internal procedure of a House of Parliament." But any failure by the Attorney-General to comply with the obligations under section 52 would be a failure that occurs prior to the engagement of the Parliamentary procedure. For that reason, it seems likely that a court would not be precluded from issuing an order in the nature of mandamus to the Attorney-General for breach of the obligations that flow from the making of the declaration of incompatibility. Thus, a party in the proceedings that was the subject of a declaration would be entitled to a process in the same sense that the applicants in Minister for Immigration and Ethnic Affairs v Teoh, Abebe v Commonwealth and Croome v Tasmania were entitled to a process.

An argument is certainly open, therefore, that, although the making of a declaration does not affect the rights of the parties between themselves, it leads to enforceable obligations against the Attorney-General and is sufficiently connected to the determination of the proceedings between the parties to be constituted as an incident of or ancillary to proceedings in which rights and obligations are determined and made binding. If this view is adopted, the High Court may be persuaded to find that the making of a declaration is an incident in the exercise of judicial power and a "matter" for the purpose of Chapter III of the Constitution. As cases concerning judicial advice to trustees, liquidators and others show, there may be an exercise of judicial power although there is 'no lis inter partes or adjudication of rights' . Although cases of this nature are 'exceptional', they probably also constitute 'a matter' for the purpose of Chapter III of the Constitution. Thus the fact that the Attorney-General may not be a party to the proceedings or may not even appear in opposition to the making of a declaration of incompatibility is not necessarily decisive on the questions of judicial power and "matter".

On balance, however, I fear that this argument will fail to persuade the High Court.

Constitutional imperatives arising from the doctrine of the separation of powers also limit the scope of courts, exercising federal jurisdiction under the dialogue model, to interpret legislation in a way that gives the fullest scope to human rights. I will discuss some of these imperatives and their consequences for the dialogue model in discussing the direction in the New Matilda Bill that courts should interpret legislation so far as it is possible to do so in a way that is compatible with human rights. I will conclude this discussion about the dialogue model by contending that that is fraught with constitutional difficulties in the federal sphere and that the campaign to enact it should be abandoned. Instead, the Parliament should follow the 1960 Canadian model of a legislative Bill of Rights and use it to give legal effect to the International Covenant on Civil and Political Rights and, if thought necessary, the International Covenant on Economic, Social and Cultural Rights by legislation that empowers courts invested with federal jurisdiction to hold that State and Territory legislation that is inconsistent with the human rights legislation is invalid and that federal legislation is to be read subject to the Bill of Rights Act that gives effect to those international covenants.

The provisions of the New Matilda Bill or similar dialogue models that will have the greatest effect on the work of the courts are sections 3, 10, 48, 49 and 50 of that Bill.

Section 49 (1) declares that "so far as it is possible to do so, primary and subordinate legislation must be read and given effect in a way which is compatible with human rights." Section 50 declares that when "interpreting this Act a court must have regard to the objects of the Act as set out in s. 3 of this Act." The principal object of the Act in terms of the daily work of the courts is found in paragraph (a) of section 3 which declares that one of the objects of the Act is "to respect, protect and promote human rights in Australia." Section 48 will also be important in the day to day workings of the courts because it declares that: "International law, and the judgments of foreign and international courts tribunal is relevant to a human right, may be considered in interpreting that human right." Thus, the great body of decisions of the Human Rights Committee of the United Nations and the European Court of Human Rights and the Supreme Court of Canada will be relevant material for the consideration of the Australian courts.

Obviously, a key provision in this proposed legislation is section 49 of the Bill with its direction that "so far as it is possible to do so, primary and subordinate legislation must be read and given effect in a way which is compatible with human rights". This section is identical in substance with s.3 (1) of the United Kingdom Act. It differs however from s.32 (1) of the Victorian Charter and s.30 of the ACT Human Rights Act, both of which require statutory provisions to be interpreted in a way that is compatible with human rights but also consistently with their purpose. As a result, the New Matilda Bill, like its United Kingdom counterpart, gives the courts a far more radical power of interpretation that is found in the Victorian and ACT legislation. As I will later show, this is a matter of considerable constitutional significance.

It is a settled principle of the interpretation of statutes in Australia that statutes are to be interpreted and applied, so far as their language permits, so as not to derogate from fundamental rights and freedoms. In Coco v The Queen, four members of the High Court including myself said:

The insistence on express authorization of an abrogation or curtailment of the fundamental rights, freedom or immunities must be understood as a requirement for some manifestation or indication that the legislature has not only directed its attention to the question of the abrogation or curtailment of such basic rights, freedoms or immunities but has also determined upon abrogation or curtailment of them. The courts should not impute to the legislature an intention to interfere with fundamental rights. Such intention must be clearly manifested by unmistakable and unambiguous language. General words will rarely be sufficient for that purpose if they do not specifically deal with the question because, in the context in which they appear, they will often be of ambiguous on the aspect of interference with fundamental rights.

It is also a settled principle of the interpretation of statutes in Australia that, so far as their language permits, statutes are to be interpreted in conformity with international law. However, the direction as to interpretation in section 49 of the New Matilda Bill goes beyond these common law principles of statutory interpretation. It will apply to a number of rights that under the common law would not be regarded as fundamental for the purpose of those principles of statutory interpretation, particularly the economic and social rights set out in ss. 37-41. But more importantly it declares that, so far as it is possible to do so, legislation must be read in a way that is compatible with human rights. As cases in the United Kingdom show, the courts have read this direction as empowering them to re-draft legislation so as to give it an interpretation that is contrary to the intention of Parliament. Lord Steyn has said that '[t]his is the intention of Parliament, expressed in s 3, and the courts must give effect to this intention. This far reaching principle of statutory interpretation has resulted from the word 'possible' in s.3 of the UK Act, when it is read in the context of the section.

In Ghaidan v Godin-Mendoza, in a long but significant passage concerning the interpretation of that section, Lord Nicholls said:

30 ... [T]he interpretive obligation decreed by s 3 is of an unusual and far reaching character. Section 3 may require a court to depart from the unambiguous meaning the legislation would otherwise bear. In the ordinary course the interpretation of legislation involves seeking the intention reasonably to be attributed to Parliament in using the language in question. Section 3 may require the court to depart from this legislative intention, that is, depart from the intention of Parliament which enacted legislation. The question of difficulty is how far, and in what circumstances, s 3 requires a court to depart from the intention of the enacting Parliament. The answer to this question depends upon the intention to be reasonably attributed to Parliament in enacting s 3...

32. ... the conclusion which seems inescapable is that the mere fact the language under consideration is inconsistent with a Convention-compliant meaning does not of itself make a Convention-compliant interpretation under section 3 impossible. Section 3 enables language to be interpreted restrictively or expansively. But section 3 goes further than this. It is also apt to require a court to read in words which change the meaning of the enacted legislation, so as to make it Convention-compliant. In other words, the intention of Parliament in enacting section 3 was that, to an extent bounded only by what is 'possible', a court can modify the meaning, and hence the effect, of primary and secondary legislation.

33. Parliament, however, cannot have intended that in the discharge of this extended interpretative function the courts should adopt a meaning inconsistent with a fundamental feature of legislation. That would be to cross the constitutional boundary section 3 seeks to demarcate and preserve. Parliament has retained the right to enact legislation in terms which are not Convention-compliant. The meaning imported by application of section 3 must be compatible with the underlying thrust of the legislation being construed. Words implied must, in the phrase of my noble and learned friend Lord Rodger of Earlsferry, 'go with the grain of the legislation'. Nor can Parliament have intended that section 3 should require courts to make decisions for which they are not equipped. There may be several ways of making a provision Convention-compliant, and the choice may involve issues calling for legislative deliberation.

Other Law Lords in Ghaidan v Godin-Mendoza placed similar qualifications on the right of a court to depart from the meaning of the statute. Those qualifications and the reasons for a radical interpretation of s.3 were summarized by Lord Bingham in Sheldrake v Director of Public Prosecutions:

First, the interpretive obligation under s 3 is a very strong and far reaching one, and may require the court to depart from the legislative intention of Parliament. Secondly, a convention-compliant interpretation under s 3 is the primary remedial measure and a declaration of incompatibility under s 4 an exceptional course. Thirdly, it is to be noted that during the passage of the Bill through Parliament the promoters of the Bill told both Houses that it was envisaged that the need for a declaration of incompatibility would rarely arise. Fourthly, there is a limit beyond which a convention-compliant interpretation is not possibleÂ… In explaining why a Convention-compliant interpretation may not be possible, membersÂ… used differing expressions: such an interpretation would be incompatible with the underlying thrust of the legislation, or would not go with the grain of it, or would call for legislative deliberation, or would change the substance of a provision completely, or would remove its pith and substance, or would violate a cardinal principle of the legislationÂ…All of these expressions, as I respectfully think, yield valuable insights, but none of them should be allowed to supplant the simple test enacted in the Act: 'So far as it is possible to do soÂ…'. While the House declined to try to formulate precise rules (at [50]), it was thought that cases in which s 3 could not be used would in practice be fairly easy to identify.

The effect of these principles of interpretation can be seen in the decision in Ghaidan v Godin-Mendoza itself. The Rent Act 1977 gave "a person who is living with the original tenant as her wife or husband" certain rights. The plain meaning of the relevant provision was contrary to the human rights of a homosexual person in that it permitted the survivor of a heterosexual, but not a homosexual, couple to become a statutory tenant by succession. Despite the plain language of the statutory provision, the House of Lords, applied s 3 of the Human Rights Act and interpreted it to give the same rights to the surviving partner of a same-sex couple.

Three other decisions show the far reaching effect of the House of Lords interpretation of section 3 of the United Kingdom Human Rights Act 1988 which is identical in substance with s.49(1) of the New Matilda Bill. In R v A (No 2), the House of Lords had to interpret a rape shield provision that limited the circumstances in which a complainant could be cross-examined about his or her sexual past. However, the House read down the provision to ensure a fair trial as required by article 6 of the European Convention on Human Rights. The House held that a court could not reject questioning or evidence as required by the rape shield provision if that questioning or evidence concerned logically relevant sexual experiences between the complainant and the accused. Thus, the interpretive provision in section 3 of the Human Rights Act 1998 had the effect of amending the rape shield provision that Parliament had enacted. Moreover, in that case the rape shield provision had been enacted after the enactment of the Human Rights Act. In Sheldrake, the House of Lords found that a provision which, if 'conventionally interpreted' would have imposed a legal burden on an accused was inconsistent with the presumption of innocence guaranteed by Article 6(2) of the Convention. So the House interpreted the legal burden to mean an evidentiary burden despite an express acknowledgement that this was not Parliament's intention when it enacted the legislation. In Secretary of State for the Home Department v MB, the House of Lords had to interpret a provision in the Prevention of Terrorism Act 2005. The provision required a court to refuse a person subject to a control order access to material if it was contrary to the public interest to disclose it. Baroness Hale said that the provision should be read and given effect unless to do so would be incompatible with the right of that person to a fair trial.

These decisions and the reasoning in them support a number of propositions in respect of an interpretative provision expressed in the terms of s. 3 of the UK Act, which, as I have pointed out, is not materially different from s.49 of the New Matilda Bill. They include: (1) Courts have power to modify legislation to give effect to human rights. (2) Courts can modify legislation whose meaning is unambiguous.(3) Courts can modify legislation even though the modification is contrary to the intention of Parliament. (4) Courts can modify legislation that is enacted after the enactment of the Human Rights Act 1998. In addition, it can be said that, despite the qualifications on the power of modification expressed by the Law Lords in Ghaidan v Godin-Mendoza, the actual decisions of the House of Lords in subsequent cases are not easy to reconcile with those qualifications.

Given the manner in which the House of Lords has interpreted section 3 of the Human Rights Act 1998, it is not surprising that, according to a talk given by Mr Murray Hunt to the Australian Human Rights Commission in February this year, the 10 years operation of that Act has produced 26 declarations of incompatibility and seven of them were overturned on appeal. That is to say, on only 17 occasions have the courts been unable to interpret legislation "in a way which is compatible with the Convention rights." This is a remarkable statistic. It indicates that either the great bulk of English legislation enacted in the past or since 1998 has been compatible with human rights or the United Kingdom courts have exercised substantial legislative power to bring United Kingdom legislation into line with the Convention for the Protection of Human Rights and Fundamental Freedoms (popularly known as the European Convention on Human Rights). Of these alternatives, the latter seems the most likely one. That is, this achievement of only 17 declarations of incompatibility in 10 years has been achieved by judicial amendment of parliamentary legislation. Remarkably on all 17 occasions, the Parliament has later amended the legislation in question to make it compatible with the Convention. Proponents of the dialogue model naturally point to these results to prove that the dialogue model promotes a human rights culture and has not diminished the relationship between the judiciary and the legislature and executive branches of government.

Although s.49 (1) of the New Matilda Bill reproduces the substantial effect of s.3 of Human Rights Act 1998, I think that there is little, if any, chance that the High Court of Australia would hold that section 49 (1), as interpreted by the House of Lords, was a valid Act of the federal Parliament. As I have indicated, our federal Constitution gives effect to the political doctrine of the separation of powers. Not only does the Constitution prohibit the legislature and the executive being given judicial power but the Constitution also prohibits the judiciary being given legislative power. In the words of Marshall CJ in Wayman v Southard which were cited by Isaacs J. in New South Wales v Commonwealth (Wheat Case): "The difference between the departments undoubtedly is, that the legislature makes, the executive executes, and the judiciary construes the law." Isaacs J. went on to say that it would require "very explicit and unmistakable words to undo the effect of the dominant principle of demarcation."

In interpreting what are known as severance clauses - clauses that declare that, where an enactment is construed as being in excess of power, it shall nevertheless be a valid enactment to the extent to which it is not in excess of that power - the High Court has frequently said that, if the Court holds that parts of legislation are invalid, a severance clause cannot constitutionally require the Court to re-draft or save the valid parts of the legislation if that would involve the court in a legislative exercise. The Parliament cannot use the Court as a delegate to produce a new law from the valid parts of the old if to do so would give the law "a new scope and object" or "a new meaning producing a new operationÂ… of a kind uncontemplated by the legislature" . In Strickland v Rocla Concrete Pipes Ltd, Menzies J. said: "It is for Parliament to make laws and the courts' only authority is to interpret them and apply them. Parliament cannot direct courts to reconstruct out of the ruins of one invalid law of general application a number of valid laws of particular application."

In Western Australia v The Commonwealth (the Native Title Act Case), speaking of a federal law, the High Court said: "If the 'common law' in section 12 is understood to be the body of law which the courts create and define, s 12 attempts to confer legislative power upon the judicial branch of government. That attempt must fail either because the Parliament cannot exercise the powers of the Courts or because the Courts cannot exercise the powers of the Parliament." The High Court went on to say that "[u]nder the Constitution, the Parliament cannot delegate to the Courts the power to make laws involving, as that power does, a discretion or, at least, a choice as to what that law should be.' It follows, if s 49 (1) were interpreted in the same manner that United Kingdom courts have interpreted section 3 (1) of the Human Rights Act 1998, it would be invalid. I think it is highly unlikely that the High Court would countenance an interpretation of s.49 of the New Matilda Bill that would authorise courts exercising federal jurisdiction to amend federal Acts of Parliament. Accordingly, the overwhelming probability is that the High Court would give s.49 a more modest operation than its United Kingdom counterpart has achieved. It seems probable that, to keep s.49 consistent with the doctrine of the separation of powers, the High Court would give s. 49 (1) a meaning that was little different from the wording of s.32 of the Victorian Charter and s.30 of the ACT Human Rights Act. That is, it would hold that, on its proper construction, s. 49 required legislation to be interpreted in a way that is compatible with human rights only when such an interpretation was consistent with the purpose of the legislation. The Court could do this by interpreting the words "[s]o far as it is possible to do so" as being limited to an interpretation that was not inconsistent with the purpose of the legislation.

To depart from the extended construction that the House of Lords gave to section 3 does not mean that section 49 (1) of the New Matilda Bill would do little more than express in statutory form the common law principle of interpretation that statutes should not be read as intending to diminish fundamental rights. Legislative provisions are frequently open to various interpretations, and an interpretation that is compatible with human rights will often be consistent with the purpose of the legislation. A New Zealand decision serves as an illustration. Under the New Zealand Bill of Rights Act, a person had a right to freedom of expression. The accused was charged under a statute that made it an offence to damage 'the New Zealand flag in any manner with the intention of dishonouring it.' As a symbol of his opposition to the policies of the government, the accused held a New Zealand flag upside down and burnt it. He was convicted at first instance after the District Court interpreted the word "dishonouring" as meaning "disrespecting". On appeal, the Court sought "to identify the meaning which constituted the least possible limitation on the right or freedom in question". It accepted the appellant's submission that, when read consistently with the right to freedom of expression, the proper meaning of "dishonour" was "vilify" or "defile' the flag, neither of which the appellant had the intention of doing. He was therefore acquitted.

However, when the purpose of Parliament is clear and the purpose of a statutory provision seems incompatible with the human rights in question, the court will have no option under the dialogue model except to make a declaration of incompatibility, assuming such declarations pass the constitutional test of validity. There must be doubt, for example, whether Mr Al Kateb would have fared better in the High Court if the New Matilda Bill had been in force as a federal enactment. His application for refugee status had been denied by the Commonwealth. Under ss. 196 and 198 of the Migration Act 1958, he had to be apprehended and then detained in custody until either he obtained a visa or was deported. Unfortunately no country would agree to receive him if he was deported from Australia. By a 4 - 3 majority, the High Court held that, despite the fact that he could not be deported and the Minister would not grant him a visa, the terms of the Migration Act required his continued detention. The minority Justices held that he could not be detained for as long as he could not be deported.

Gleeson CJ, one of the minority Justices, said that the period of detention of Al Kateb was defined by reference to the fulfilment of the purpose of removal under s 198. If that purpose could not be fulfilled, the choice lay between treating the detention as suspended, or as indefinite. In making that choice, he was influenced by the general principle that Courts do not impute to the legislature an intention to abrogate or curtail certain human rights or freedoms (of which personal liberty is the most basic) unless such an intention is clearly manifested by unambiguous language, which indicates that the legislature has directed its attention to the rights or freedoms in question, and has consciously decided upon abrogation or curtailment. The primary purpose of Al Kateb's detention was in suspense, but it had not been made permanently unattainable. Gleeson CJ said that the Act made no express provision for suspension, and possible revival, of the obligation imposed by s 196 of the Act to detain Al Khateb, according to the practicability of effecting removal under s 198. Similarly, it made no express provision for indefinite, or permanent, detention in a case where the assumption underlying s 198 (the reasonable practicability of removal) was false. In resolving questions raised by the legislative silence, his Honour said resort could, and should, be had to the fundamental principle of interpretation to which I have referred.

The majority Justices - of which I was one - took the view that the plain purpose of the legislation was that an unlawful entrant into Australia - which Mr Al Kateb was - had to be detained in immigration detention until that person was either given a visa or deported. At least so far as I was concerned, to introduce issues of suspension of detention was not an application of the fundamental principle of construction to which the Chief Justice referred but an amendment of the imperative terms of ss.196 and 198 of the Migration Act. In applying the fundamental principle of construction, one usually reads the legislation as not intended to impair a fundamental right. But in reality that was not what the minority Justices did. Logically, they had to accept, as I think they did, that Mr Al Kateb's initial apprehension and detention was lawful. However, in effect they said that, once it appeared that no country would accept him, he had to be released and could not be apprehended while that state of affairs existed. I am afraid that, much as I deplored the result, I thought that constituted an amendment of the Migration Act, not a giving effect to the fundamental principle of statutory construction. I could not accept that the Migration Act could be construed as also providing for the release of Mr Al Kateb, subject to conditions or otherwise, while no other country would receive him when the legislation expressly said that he had to be detained until he was given a visa or removed from Australia.

The circumstances of Mr Al Kateb's case were tragic, and I lamented that Australia did not have a Bill of Rights that could have overcome the mandatory terms of ss.196 and 198 of the legislation. But would he have done better before the High Court under a Bill of Rights, framed as the New Matilda Bill is? I think that there must be serious doubt about the answer to that question. The majority Justices may well have held that to give effect to his right to liberty and freedom from arbitrary detention under the New Matilda Bill (s.14) was inconsistent with the purpose of the Migration Act which was to detain a person such as Mr Al Khateb in custody until he was removed from the country or given a visa. On the other hand, I have little doubt that, if he had been detained under similar circumstances in the United Kingdom, a court would have held that he was entitled to be released from detention until some country gave him asylum. The case of Al Kateb shows a weakness in the dialogue model of human rights protection. If a person's human rights under that model cannot prevail if they would conflict with the purpose of legislation, they must go unprotected, and that may possibly occur in many situations. On any view, the interpretive provisions of the dialogue model are suboptimal.

I think the better course is to reject the dialogue model as a vehicle for implementing the protection of human rights in the federal sphere, and indeed at the State level. It has a number of significant weaknesses in terms of protecting human rights. First, at the very least, there is grave doubt as to the constitutionality of its declaration of incompatibility provisions which are at the heart of the dialogue. Second, even if those provisions are upheld, by reason of the Parliament's inability to confer legislative power on courts exercising federal jurisdiction, an interpretive provision such as s.49(1) may often fail to protect the human rights that are exercisable under the Act. Third, if s.4 means what it says and the "human rights in this Act are exercisable by everyone within Australia's jurisdiction", the New Matilda Bill will affect the interpretation of State legislation because, by reason of section 109 of the Constitution, the State laws would be inoperative to the extent that they were inconsistent with section 4. Yet there is no dialogue between the Federal courts and the State legislatures, nor constitutionally could there be. Fourth, if the human rights referred to in section 4 only engage with federal laws, there will be a very large gap in the protection of human rights in Australia unless and until all the States and Territories adopt similar legislation. Fifth, apart from the right of action against public authorities, the dialogue model creates no rights or causes of action. Sixth, as a result, those whose human rights have been infringed have no remedies for infringements of those rights. They cannot obtain damages or injunctions to restrain the conduct that infringes their rights. This is a breach of Article 2 (3) of the International Covenant on Civil and Political Rights which provides that, if a person's rights under that Covenant have been violated, that person has a right to an effective remedy. Under that Article, each State Party - and Australia is one - undertakes to ensure that any person whose rights or freedom is violated shall have an effective remedy and that the person claiming such a remedy shall have his or her right determined by a competent judicial, administrative or legislative authorities, or by another competent authority provided by the legal system of the State. Seventh, even if the Parliament amends its legislation after receiving a declaration of incompatibility, considerable time will usually elapse before anything is done. Even when the amendment has been made, it may be of no benefit to the person whose rights have been infringed. Eighth, because courts, exercising federal jurisdiction, cannot be given legislative type powers that the courts of the United Kingdom have been using, it seems a near certainty that declarations of incompatibility will be made far more frequently in Australia than in the United Kingdom. This may have serious consequences for the work load of the Parliament. Although the Attorney-General's response must be presented to the House of Representatives not later than six months after the day a copy of the declaration is presented to that House, there is no time limit imposed for the Parliament to take action in respect of that response.

What may have been overlooked by those who champion the enactment of a dialogue model of human rights for Australia is that what may work effectively in a jurisdiction with an unwritten constitution and a single legislature, as in the United Kingdom and New Zealand, may not work as effectively in a federal jurisdiction with a written constitution that incorporates the political doctrine of the separation of powers.

Instead of the dialogue model, the Parliament should give effect to the International Covenant on Civil and Political Rights and, if thought necessary, the International Covenant on Economic, Social and Cultural Rights by legislation that empowers courts invested with federal jurisdiction to hold that legislation that is inconsistent with the human rights legislation is invalid in the case of State and Territory legislation and that, in the absence of an express statement to the contrary, all federal legislation is to be read subject to the human rights legislation of the Parliament. The result would be that private citizens would have judicially enforceable human rights that were not affected by State, Territory or federal legislation inconsistent with those rights and would have immediate judicial remedies for breaches of those rights. In the absence of a 'notwithstanding' clause, their rights would not be dependent upon whether their right was consistent with the purpose of the legislation, but would be judged and applied on its merits and by reference to the federal equivalent of s.10 of the New Matilda Bill.

A human rights legislative model on these lines would have only a minimal effect on parliamentary sovereignty. Under my preferred model, it would be open to the Parliament of the Commonwealth to insert in any federal legislation a "notwithstanding' clause which required the courts to give effect to that particular legislation notwithstanding the enactment of the human rights legislation. And, of course, it would be open to the Parliament after any decision with which it disagreed to insert a 'notwithstanding' clause in the legislation which the court had said should be ignored in determining rights and obligations. Finally, such a model would be well within federal constitutional power and would not be open to the constitutional attacks that undoubtedly await the dialogue model.

I am conscious that my criticisms of the dialogue model will provide ammunition for those who are opposed to the enactment of any form of a Bill of Rights, statutory or constitutional. I regret that this is so. But it would be a tragedy for the human rights movement in Australia if the dialogue model was enacted and the declaration of incompatibility provisions was struck down as unconstitutional. In that event, there would be no dialogue between the judiciary and the legislature and the executive, and, apart from the admittedly important right of action against federal public authorities, the human rights of the people of the Commonwealth would have only marginally more protection than they presently do under the common law principles of statutory interpretation. Moreover as I have pointed out, the dialogue model has other deficiencies, not the least of which is the lack of judicially enforceable remedies, except as against public authorities of the Commonwealth.

 


The Hon. Michael McHugh AC QC was a justice of the High Court of Australia from 1989 to 2005. This paper was presented to the Australian Human Rights Commission on 5 March 2009.


Appendix

Section 1 of the Constitution vests the legislative power of the Commonwealth in the Federal Parliament. Section 61 vests the executive power of the Commonwealth in the Queen and declares that it is "exercisable by the Governor-General as the Queen's representative". Section 71 vests the judicial power of the Commonwealth in a Federal Supreme Court to be called the High Court of Australia and such other federal courts as the Parliament creates, and in such other courts as it invests with federal jurisdiction. These three sections are the source of the long accepted view that our federal Constitution incorporates the political doctrine of the separation of powers. As a result 'the Parliament is restrained both from reposing any power essentially judicial in any other organ or body, and from reposing any other than that judicial power in such tribunals' . So the question arises whether the provisions in sections 51 or 52 of the New Matilda Bill are invalid in that they invest a court exercising federal jurisdiction with non-judicial power. Closely allied to this question is whether the issues which would arise under sections 51 and 52 of the New Matilda Bill would involve "matters" within the meaning of sections 75, 76 and 77 of the Constitution. Those three sections empower the Parliament to give jurisdiction to the federal courts and to State courts invested with federal jurisdiction. Federal judicial power can be exercised only in respect of 'matters'. Unless the claim involved in a legal proceeding constitutes a 'matter', the Parliament cannot give federal jurisdiction to any court.

As Deane, Dawson and Gaudron JJ and I pointed out in Brandy v Human Rights and Equal Opportunities Commission , in describing what is meant by judicial power, '[i]t is traditional to start with the definition advanced by Griffiths CJ in Huddart, Parker & Co Pty Ltd v Moorehead' where his Honour said:

I am of opinion that the words 'judicial power' as used in sec 71 of the Constitution mean the power which every sovereign authority must of necessity have to decide controversies between its subjects, or between itself and its subjects, whether the rights relate to life, liberty or property. The exercise of this power does not begin until some tribunal which has power to give a binding and authoritative decision (whether subject to appeal or not) is called upon to take action.

Another widely used description of judicial power is that given by Kitto J in R v Trade Practices Tribunal; Ex parte Tasmanian Breweries Pty Ltd:

Thus a judicial power involves, as a general rule, a decision settling for the future, as between defined persons or classes of persons, a question as to the existence of a right or obligation, so that an exercise of the power creates a new charter by reference to which that question is in future to be decided as between those persons or classes of persons. In other words, the process to be followed must generally be an inquiry concerning the law as it is and the facts as they are, followed by an application of the law as determined to the facts as determined; and the end to be reached must be an act which, so long as it stands, entitles and obliges the persons between whom it intervenes, to observance of the rights and obligations that the application of law to facts has shown to exist.

However, these descriptions, or definitions if you like, are neither conclusive nor exhaustive of what constitutes judicial power. In Precision Data Holdings Ltd v Wills, Mason CJ, Brennan, Deane, Dawson, Toohey, Gaudron and I said that 'framing a definition of judicial power that is at once exclusive and exhaustive' is difficult if not impossible. Earlier Windeyer J had said that 'the concept seems Â… to defy, perhaps it were better to say transcend, purely abstract conceptual analysis'. As Deane, Dawson and Gaudron JJ and I also pointed out in Brandy v Human Rights and Equal Opportunities Commission:

Difficulty arises in attempting to formulate a comprehensive definition of judicial power not so much because it consists of a number of factors as because the combination is not always the same. It is hard to point to any essential or constant characteristic. Moreover, there are functions which, when performed by a court, constitute the exercise of judicial power but, when performed by some other body, do not.

These difficulties caused Aickin J to say in R v Quinn; Ex parte Consolidated Foods Corporation that 'in substance all that the courts have been able to say towards a definition has been the formulation of negative propositions by which it has been said that no one of a list of factors is itself conclusive and perhaps the presence of all is not conclusive'.

As a result, whether or not the vesting of a power in attributable is judicial or non-judicial depends upon a value judgment made after weighing the factors that point to the vesting of judicial power against the factors in the point against the vesting of judicial power in the tribunal. Inevitably, this often makes it difficult to predict with confidence the constitutional validity of a legislative scheme investing a tribunal with novel powers and functions.

As the judgment of Griffith CJ in Huddart, Parker & Co Pty Ltd v Moorehead shows, one of the strong indicators of judicial power is the vesting in the tribunal of a "power to give a binding and an authoritative decision". However, the New Matilda Bill states that a declaration of incompatibility 'is not binding on the parties to the proceedings in which it is made'. At first sight, this stipulation strongly points against the making of a declaration of incompatibility being an exercise of judicial power. However, commentators who argue in favour of the constitutionality of the dialogue model of human rights legislation rely on a number of arguments to counter what at first sight may seem a difficult, if not an insuperable, hurdle to a holding that the making of a declaration of incompatibility is an exercise of judicial power.

Thus, Dominique Dalla-Pozza and Professor George Williams argue that the "binding and authoritative decision" indicator of judicial power can be found in the declaration mechanisms. They contend that it is important to recognise that declarations generally 'are statutory rather than equitable remedies' and that this 'suggests that when a parliament bestows the power to make declarations on a court such declarations are intended to be statements of law.' That is undoubtedly so, but, with respect it gives no assistance in a statutory context which expressly says that a declaration of incompatibility is not binding upon the parties to the controversy.

They also point out that the mechanisms under the ACT Human Rights Act, the Victorian Charter and the New Matilda Bill make clear that binding obligations do flow once courts have declared that an incompatibility or inconsistency exists. Thus, if the New Matilda Bill were to be enacted by the federal Parliament, s 51(5) would require 'a Court' to provide a copy of the declaration to the Attorney-General. Section 52 then imposes obligations on the Federal Attorney-General who is required to present a copy of the declaration within 15 sitting days of receiving it while the Attorney-General's mandatory written response is to be presented to the House of Representatives 'not later than 6 months' after the declaration is given to the House.

Ms Dalla-Pozza and Professor Williams contend that important consequences follow from these consequences of the making of a declaration. The relevant sections impose a duty upon the first law officer in each jurisdiction to respond in the legislature to the making of a declaration. They say that this must be seen in light of the responsibility of ministers to Parliament. Thus, the Parliament could dismiss the executive by way of a no-confidence motion, or could pass a no-confidence motion in the Attorney-General. The Parliament could also amend a law that has been held to be incompatible by a court. But as these commentators acknowledge, the efficacy of doctrines of responsible government and no-confidence motions in an era when Parliament is dominated by party discipline and strong executive power is problematical. Nevertheless, they see the important point as being that the New Matilda Bill and the ACT Human Rights Act and the Victorian Charter stipulate that a declaration triggers binding obligations and that the duties imposed by the courts arise in the context of the 'dialogue' which those enactments or the proposed Bill seek to implement. They argue that human rights are intended to be interpreted and applied not only by the judiciary but as part of a dialogue between the judiciary, executive and legislature, and that, seen as part of this process, the duties imposed upon the Attorney-General are an important and substantive consequence of a litigant gaining a declaration of incompatibility. They refer to a statement of the ACT Bill of Rights Consultative Committee that the mechanism is a 'sufficiently strong and appropriate enforcement mechanism to underpin the dialogue approach of the ACT Human Rights Act'. They also refer to the second reading speech of the Victorian Attorney-General who indicated that the purpose of giving the power to make declarations to the courts 'is to allow the Parliament to reconsider the provision in light of the declaration of inconsistent interpretation.' They rely on these statements as a legislative recognition that the declaration mechanisms are an (albeit novel for Australia) form of legal remedy because they produce foreseeable and practical consequences in response to a judicial finding that a law is inconsistent with a protected human right.

Ms Dalla-Pozza and Professor Williams acknowledge that implicit in discussions of judicial power, such as that of Kitto J in Tasmanian Breweries, is the proposition that judicial power is applied to bind the parties to the proceedings. They also acknowledge that '[w]hile we have found that obligations bestowed by the declaration mechanisms are binding, it is the Attorney-General and not the parties who are bound by these obligations.'

It seems clear from this acknowledgement that the learned authors contend that, in making a declaration of compatibility, the court is exercising judicial power in respect of a controversy between the parties, even though the parties are not bound by the obligations arising from the declaration. They treat the fact that a declaration is not binding on the parities as nothing more than a negative factor in determining whether the power to make a declaration of incompatibility in the proceedings is a judicial power. But the High Court may regard this as downplaying the significance of a factor that more often than not is decisive of the presence or absence of judicial power.

Despite the eminence of these two commentators and my respect for their views, I do not find the reasons for their argument concerning the binding and authoritative nature of a declaration of incompatibility persuasive as an indicator of an exercise of judicial power. I do not think that the "binding and authoritative decision" indicator of judicial power can be found in the declaration mechanisms itself. Rather, I think that, if the declaration involves and exercise of judicial power, it is because it leads to separate obligations on the part of the Attorney-General which can be enforced by the parties as a consequence of their rights which can be the subject of a binding and authoritative decision. On that view, the High Court might conclude that making the declaration is so closely connected to the authoritative decision that it is incident thereto. When Griffith CJ spoke of the "power to give a binding and authoritative decision" he surely meant, as he has always been understood to mean, a binding and authoritative decision which determined a controversy between subjects or between the sovereign and a subject. A declaration of incompatibility does not determine any controversy between subjects or any controversy between the parties to the controversy that gave rise to the declaration. Nor does it bind the parties. The New Matilda Bill, like its State and Territory counterparts, expressly states that a declaration of incompatibility 'is not binding on the parties to the proceedings in which it is made'. Moreover, although a declaration imposes obligations on the Attorney-General, it is imposed by the Bill, not by the court in deciding the controversy between the parties. The Attorney's obligation is a consequence of the making of the declaration in the same way that a sheriff's obligation to enforce a judgment is the consequence of a court making a judgment.

Accordingly, I find unpersuasive the reasons that Ms Dalla-Pozza and Professor Williams give for their conclusion that the "binding and authoritative decision" indicator of judicial power can be found in the declaration mechanisms. However, it may be that these learned authors are saying no more than that the making of a declaration of incompatibility is an exercise of judicial power because it is a consequence of a binding and authoritative decision of a dispute between the parties. If this is so, I agree that this argument supports the validity of the declaration mechanism. As I have indicated, it may be that the High Court would find persuasive the fact that a party to the proceedings can enforce obligations imposed on the Attorney-General as being sufficiently incidental to the main proceedings as to envelop the declaration with the mantle of an exercise of judicial power. Nevertheless, to do so is taking a large leap.

Ms Dalla-Pozza and Professor Williams rely on two other factors as indicating that the making of the declaration is an exercise of judicial power. First, that the court must use legal standards to resolve the controversy. Second, the declaration is enforceable. If a tribunal is to determine a dispute by the application of non-legal standards and its determinations are not enforceable, it is almost certain that it is not exercising judicial power. But it does not follow that, if it must apply legal standards and that its decisions are enforceable, it is exercising judicial power. Many administrative tribunals, which no one would suggest are exercising judicial power, do both.

As Ms Dalla-Pozza and Professor Williams point out one indication of judicial power is that it is a power that requires the use of legal standards, as opposed to policy criteria, in determining a dispute and that the New Matilda Bill has been drafted to ensure that this is the case. In the New Matilda Bill, 'Human Rights' are defined in s 6 as 'the civil, political, economic, social and cultural rights set down in part 3'. Hence, they argue that this legislative proposal sets up specific legal criteria by which compatibility with human rights can be judged. Furthermore, s. 51(6) of the Bill declares that a court has the power to make a declaration of incompatibility 'when a court is exercising jurisdiction in any cause or matter pending before it' with the result that a declaration could only be made in determining an existing controversy between parties. But at its highest that statement means no more than the declaration would be made in the course of or as a consequence of determining a controversy between parties. And it may be, as the decision in Solomons v District Court of New South Wales suggests, that the federal jurisdiction of a court would be spent before it makes the declaration. But, assuming that the declaration is made in the exercise of jurisdiction, the making of the declaration has a temporal character, not a determinative character that defines the rights of the parties. That is because, by force of the New Matilda Bill, the declaration 'is not binding on the parties to the proceedings in which it is made'. Whatever may be the controversy between the parties, the making of the declaration is related to it only in a temporal sense, for the declaration does not determine the controversy between the parties.

Ms Dalla-Pozza and Professor Williams do not appear to accept that the declaration does not determine the controversy between the parties. They contend that '[i]t is also possible to argue that issuing a declaration resolves the controversy between the parties.' But they advance no specific reasons in support of this argument. Instead, they say that the making of the declaration can be distinguished from the situation in Solomons v District Court of New South Wales. Solomons had been acquitted of a federal offence after a trial in the District Court. Under the Costs in Criminal Cases Act 1967 (NSW), a person who had been acquitted of an offence could apply to the court for a certificate. If the certificate was granted, the acquitted person could apply to a State official for reimbursement of his or her costs. Gleeson CJ, Gaudron, Gummow, Hayne and Callinan JJ found that, if the District Court, a court that had been exercising federal jurisdiction, granted a certificate, it 'would involve it in the exercise of a power not provided in Chapter III of the Constitution. It would be productive of futility, not the resolution of any claim or controversy'. That was because the controversy between the Crown and Solomons, arising out of the federal charge, concerned the application of federal law and the granting of a certificate as a step in making an application under New South Law was not part of that federal controversy, there being no similar costs scheme at the federal level.

Ms Dalla-Pozza and Professor Williams contend that, in contrast to the position in Solomons, a declaration under the New Matilda Bill would place an obligation on the federal Attorney-General in the same jurisdiction as gave rise to the initial controversy. That can be accepted, but, with respect, it does not bear on the question whether 'issuing a declaration resolves the controversy between the parties.' A declaration of incompatibility does not resolve the dispute that gives rise to the federal court's jurisdiction over the parties. Nor can it do so in the future. The rights and obligations of the parties will be determined by whatever order the federal court makes in the proceedings. The declaration says nothing about those rights or obligations or the claim that gave rise to the controversy. If a person sued the Commonwealth for false imprisonment for being wrongly detained under terrorist legislation and the court thought that that legislation was incompatible with a right in Part 3 of the New Matilda Bill, the claim or controversy would be determined by whether the legislation authorized the detention, not by the declaration which is not binding on the parties.

Solomons is not necessarily decisive as to whether making a declaration of incompatibility is an exercise of non-judicial power. But it does show that the High Court is likely to hold that a court is not exercising federal judicial power merely because making a declaration or grant in respect of an issue arising out of the federal proceedings may have beneficial consequences for a party to those proceedings. Indeed, any beneficial consequences for a party arising under the dialogue model, if they exist at all, are further removed from the controversy between the parties than they were in Solomons' case.

In Brandy, a majority of the High Court identified the 'enforceability of decisions' as a critical indicator of judicial power in cases where the characterisation of a function 'is otherwise equivocal'. It follows that, if declarations of incompatibility are unenforceable in the relevant sense, they are unlikely to be recognised as an expression of judicial power.

Under the New Matilda Bill, courts have no power to enforce the rights which are the subject of a declaration of incompatibility. Section 51 of the Act excludes such rights from the court's jurisdiction. However, Ms Dalla-Pozza and Professor Williams meet this hurdle by adopting a suggestion by Wendy Lacey and David Wright that the enforceability requirement can be met if that requirement is given an expansive understanding. Those authors suggest that a declaration may fall within the definition of an exercise of judicial power 'provided it is conclusive of the controversy regarding inconsistency' They recognise that that could occur only where the factual circumstances giving rise to the incompatibility can be 'merged in the judgement'.

This is a persuasive argument that may succeed before the High Court. If it did, it would be a strong indicator that the making of a declaration of incompatibility is an exercise of judicial power. But the argument is not without its difficulty. The contrary view is that any legal controversy regarding inconsistency is divorced from the making of the declaration of incompatibility. In the terrorist example that I gave, the plaintiff would argue that, upon the proper interpretation of the legislation, it did not interfere with his human right of freedom from arbitrary arrest and detention referred to in s.14 of the New Matilda Bill. That is, the plaintiff would argue that the terrorist legislation can be interpreted consistently with the s.14 human right. The defendant would argue that the legislation was incompatible with that right. So there would exist a controversy between the parties in respect of the issue of incompatibility. If the court accepted the defendant's argument, it would determine the plaintiff's claim by reference to the terms of the terrorist legislation. But any declaration of incompatibility that it made would not declare, let alone, determine the rights of the parties. That is because s.51(4) (b) declares that the declaration is 'is not binding on the parties in the proceedings in which it is made.'

A statement made by Ms Stephanie Wilkins in respect of the ACT Human Rights Act seems equally applicable to the New Matilda Bill. She has written:

Conversely, however, the description of the declaration process is clearly separated in the Act from the interpretive process, and an interpretation may affect the rights and obligations of the parties while a declaration will not. Further, there is a legislative intention that the declaration initiates 'dialogue' between the legislature and the judiciary, a role not envisaged in relation to the interpretive process. The more likely position is therefore that the declaration process will be regarded as separate from interpretation.

The argument suggested by Wendy Lacey and David Wright would have greater force if the declaration was binding on the parties. It might then be seen as ancillary or incidental to or inherent in determining the rights of the parties. The scope for making declarations in curial proceedings is very wide. Courts will not grant a declaration where the issue is hypothetical or where it 'will produce no foreseeable consequences for the parties', but it has long been established that they will do so if the applicant has no other cause of action. Statutory provisions or Rules of Court in various Australian jurisdictions also provide that a proceeding is not open to objection on the ground that only a declaratory order is sought by the proceeding. They also provide that the court may make binding declarations of right in an action whether any consequential relief is or could be claimed in that action or not. If the declaration was binding on the parties, it would declare as between them that, on its proper interpretation, the relevant legislation was incompatible with the human right in question. The court's obligation to transmit a copy of the declaration to the Attorney-General might then be seen as an incidental, ancillary or necessary act in the exercise of federal judicial power. But given the presence of s.51 (4) (b), the making of the declaration does not make any determination of the parties' rights. On that view, the making of a declaration does not cause any change in or recognition of the rights of the parties. It does not involve determining a right or duty. Nor, of itself, is it an attempt to administer the law. Hence, on the basis that it is not binding on the parties, it does not involve an exercise of federal judicial power, as between them.

However, if, as is probably the case, the declaration is divorced from the determination of the parties' rights as between themselves, it does not follow that the making of a declaration means that there is no exercise of judicial power. Stephanie Wilkins has argued, in discussing the ACT position, that, although no substantive rights are involved in or arise from a declaration of incompatibility, a party who obtains a declaration is 'entitled to a process: a response from the Attorney General (albeit one directed to the Legislative Assembly), and the possibility of a reconsideration (albeit of a law rather than of an administrative decision).' She refers to Minister for Immigration and Ethnic Affairs v Teoh, Abebe v Commonwealth and Croome v Tasmania in support of the view that federal judicial power may be exercised even though the only practical entitlement of the applicant is the fulfillment of a formality by the executive government and the possibility of a reconsideration of an original decision. Thus, in Teoh, the applicant was not entitled to the fulfillment of his legitimate expectation that Australia's international obligations would be fulfilled, only a right to have the original decision reconsidered by taking into account those obligations. In Abebe, Gummow and Hayne JJ referred to the applicant's right to have the Refugee Review Tribunal consider her case. In Croome, the Court held that although the applicant was not at risk of the law being enforced against him, he had the right to have the validity of the State law considered because his conduct fell within its purview.

Ms Wilkins relies on these cases for the proposition that federal judicial power can be exercised although no substantive rights are involved or arise in the proceeding. It is enough that the applicant is 'entitled to a process'. But, as she recognizes, the issue of entitlement to a process is bound up with the right of a party to enforce the obligations that the Bill and the ACT Human Rights Act and the Victorian Charter impose upon the Attorney-General in each jurisdiction. Her argument that the making of a declaration of incompatibility involves the exercise of judicial power is therefore different from that of Ms Dalla-Pozza and Professor Williams. While they evaluate the issue of judicial power in the context of the controversy between the parties, Ms Wilkins sees the issue of judicial power as involving the relationship of a party and the Attorney-General. It is convenient to delay discussing the question of enforcing the Attorney-General's obligations until I have discussed the judicial expositions of the term "matter" and whether a declaration of incompatibility is an advisory opinion or an impermissible intrusion by courts exercising federal jurisdiction into the realm of legislative or executive power.

Closely allied to the question of whether a declaration of incompatibility constitutes an exercise of judicial power is the requirement that the making of the declaration involves a "matter" for the purposes of section 75, 76 and 77 of the Constitution. In In re Judiciary and Navigation Acts, Knox CJ, Gavan Duffy, Powers, Rich and Starke JJ said:

We do not think that the word 'matter' in sec 76 means a legal proceeding, but rather the subject matter for determination in a legal proceeding. In our opinion there can be no matter within the meaning of the section unless there is some immediate right, duty or liability to be established by the determination of the Court. If the matter exists, the Legislature may no doubt prescribe the means by which the determination of the Court is to be obtained, and for that purpose may, we think, adopt any existing method of legal procedure or invent a new one.

In re Judiciary and Navigation Acts concerned an attempt by the Parliament to confer jurisdiction on the High Court to hear and determine any question of law as the validity of an enactment of the Parliament whenever the Governor-General referred such a matter to the Court for determination. The High Court held that the Parliament had no power to vest such a jurisdiction in a federal court. The reason given for that conclusion in In re Judiciary and Navigation Acts was that it did not involve the exercise of federal judicial power because it did not give rise to a "matter" within the meaning of Chapter III. Determining such a question constituted an advisory opinion, and advisory opinions are not "matters" for the purpose of Chapter III of the Constitution. Hence, any legislation that requires a federal court to give an advisory opinion is inconsistent with Chapter III of the Constitution. If the making of a declaration of incompatibility constituted an advisory opinion, a court exercising federal jurisdiction could not be given the power to make it.

In Mellifont v Attorney-General (Queensland) , Mason CJ, Deane, Dawson and Gaudron JJ and I said that the opinion of the Court that was sought in In re Judiciary and Navigation Acts 'was academic, in response to an abstract question, and hypothetical in the sense that it was unrelated to any actual controversy between parties'. Earlier in Mellifont, we had said, after citing a passage from the majority judgment in In re Judiciary that it contained 'two critical concepts. One is the notion of an abstract question of law not involving the right or duty of any body or person; the second is the making of a declaration of law divorced or dissociated from any attempt to administer it.' It follows that any determination which does not involve a right or duty in any person or is divorced from any attempt to administer the law will not be a "matter" for the purpose of Chapter III.

Although In re Judiciary and Navigation Acts has never been overruled, is frequently cited, and is regarded as the seminal decision on the term "matter", subsequent decisions of the High Court are not easily reconciled with all reasoning in that case. In particular, it is not easy to reconcile some decisions with the statement in that case 'there can be no matter within the meaning of the section unless there is some immediate right, duty or liability to be established by the determination of the Court (my emphasis).' Despite the decision in In re Judiciary and Navigation Acts, the High Court has determined questions where the connection between the declaratory relief sought and an immediate right or duty has been tenuous or in reality involved an abstract question of law.

In O'Toole v Charles David Pty Ltd, for example, the High Court held 'that answers given by the full court of a court to questions reserved for its consideration in the course of proceedings in a 'matter' pending in that court did not constitute an advisory opinion or abstract declaration of law of the kind dealt with in In re Judiciary and Navigation Acts.' The Court reached that conclusion even though it was possible that the controversy might have been decided on grounds unconnected with the questions reserved. In Mellifont v Attorney-General (Queensland), the High Court held that no advisory opinion was involved in a procedure which empowered the Attorney-General to refer a point of law to a Court of Criminal Appeal in cases where an accused person had been acquitted of a charge. The Court said:

True it is that the purpose of seeking and obtaining a review of the trial judge's reasoning was to secure a correct statement of the law so that it would be applied correctly in future cases. However, in our view, in the context of the criminal law, that does not stamp the procedure ... as something which is academic or hypothetical so as to deny that it is an exercise of judicial power.

The Court said that the review procedure arose out of the proceedings on the indictment and 'the reference was made with respect to a 'matter' which was the subject of the legal proceeding at first instance and was not divorced from the ordinary administration of the law.'

Croome v Tasmania involved different considerations. The plaintiffs were homosexuals. The Tasmanian Criminal Code made certain homosexual practices unlawful. However, a federal law held that no offence was committed in respect of "sexual conduct involving only consenting adults acting in private". The plaintiffs argued that the Tasmanian law was invalid under section 109 of the Constitution because it was inconsistent with federal law. Tasmania met the claim by contending that that no prosecution under the Code was pending or threatened and that, accordingly, the plaintiffs' claim did not constitute a "matter". Brennan CJ, Dawson and Toohey JJ said:

It is a misconception of the principle in In re Judiciary and Navigation Acts to suggest that, in proceedings for a declaration of invalidity of an impugned law, no law is administered unless the executive government has acted to enforce the impugned law. The law that is being administered in such proceedings is not the impugned law but the constitutional or administrative law which determines the validity or invalidity of the impugned law.

Applying for a declaration of incompatibility can also be seen to constitute a 'matter' because such a request is not divorced from an attempt to administer the law. The above statement in Croome v Tasmania shows that, where the constitutionality of a particular law is in question, the law that is being administered is 'the law governing the controversy about the impugned law'. The substance of a declaration of incompatibility under the New Matilda Bill is that a federal enactment is inconsistent with a section 4 right. So, it is the federal enactment that is 'the law governing the controversy' concerning incompatibility and it is that law that is being administered. There is therefore a close analogy between asking a court to make a declaration of incompatibility and asking a court to declare that particular State legislation is inconsistent with a federal law that grants a right or privilege, as in Croome. Moreover, the New Matilda Bill provides that declarations can only be issued in the context of a pre-existing dispute which supports the argument that the law is being administered and that the law that is being administered is the federal law that is alleged to be incompatible with the s.4 protected human right. This supports the conclusion that the making of a declaration is not a declaration concerning an abstract question of law and is not divorced from the administration of the law. It is therefore not an advisory opinion.

However, it is one thing to say that a declaration is not an advisory opinion and another thing to say that it is a 'matter' or an exercise of federal judicial power. In Kable v Director of Public Prosecutions (NSW) Gummow J said "that the term 'matter' in Ch. III does not exhaust, and is narrower than, what otherwise might be within ordinary concepts of judicial power." His Honour went on to say that In re Judiciary and Navigation Acts "was decided on the footing that judicial power in a general sense was not coextensive with the narrower limits of judicial power with which Ch. III of the Constitution is concerned."

One of the consequences that flow from the issue of a declaration is that the Bill imposes certain duties on the Attorney-General although those duties arise from the terms of section 52 of the New Matilda Bill and not the declaration itself. I can see no reason why a party to a declaration of incompatibility could not enforce the Attorney-General's duty to present a copy of the declaration to the House of Representatives and to prepare a written response and at present it to that House. Although the Attorney-General's duty is connected to parliamentary proceedings, an order of the court would not interfere with the parliamentary process or interfere with the workings of the Parliament. In Egan v Willis, I referred to the traditional view that courts cannot investigate or take action in respect of conduct that "relates only to the internal procedure of a House of Parliament." But any failure by the Attorney General to comply with the obligations under section 52 would be a failure that occurs prior to the engagement of the Parliamentary procedure. For that reason, it seems likely that a court would not be precluded from issuing an order in the nature of mandamus to the Attorney-General for breach of the obligations that flow from the making of the declaration of incompatibility. Thus, a party in the proceedings who was the subject of a declaration would be entitled to a process in the same sense that the applicants in Minister for Immigration and Ethnic Affairs v Teoh, Abebe v Commonwealth and Croome v Tasmania were entitled to a process.

An argument is certainly open, therefore, that, although the making of a declaration does not affect the rights of the parties between themselves, it leads to enforceable obligations against the Attorney-General and is sufficiently connected to the determination of the proceedings between the parties to be constituted as an incident of law or ancillary to proceedings in which rights and obligations are determined and made binding. If this view is adopted, the High Court may be persuaded to find that the making of a declaration is incidental or ancillary to the exercise of judicial power and a "matter" for the purpose of Chapter III of the Constitution. As cases concerning judicial advice to trustees, liquidators and others show, there may be an exercise of judicial power although there is 'no lis inter partes or adjudication of rights' . Although cases of this nature are 'exceptional', they probably also constitute 'a matter' for the purpose of Chapter III of the Constitution. Thus the fact that the Attorney-General may not be a party to the proceedings between the contending parties or may not even appear in opposition to the making of a declaration of incompatibility is not necessarily decisive on the question of judicial power and "matter".

Another argument that is open is that the making of a declaration resolves the issue between the Attorney General and the party who claims that the legislation, properly interpreted, is rights compliant. The difficulty with that argument is that that it is only when the Court is considering making a declaration of incompatibility and the Commonwealth is not a party to the proceedings that the Court must give notice of the issue to the Attorney General. Moreover, the Bill imposes no obligation on the Attorney-General to appear in the proceedings; nor does it expressly make the Attorney General a party to the proceedings if he or she does appear or intervenes in those proceedings. Thus, a declaration of incompatibility may be made even though the Attorney General does not get involved in whether or not a declaration of incompatibility should be made.

As Ms Dalla-Pozza and Professor Williams have written :

It remains difficult to predict how this issue would be resolved should the question of the constitutional validity of declarations of incompatibility come to be adjudicated. This is because the jurisprudence surrounding the 'matter' concept is complex and is characterised by 'a deep division of judicial opinion'. Mantziaris and McDonald have identified two competing interpretations of the term emerging from contemporary High Court decisions. They have labelled these the 'broad' and 'narrow' view of what constitutes a 'matter'. They note that under the 'broad' reading of the concept a 'matter' might exist even though there is 'no lis inter partes or adjudication of rights'. Although the judges who expound that view acknowledge that such cases are 'exceptional', they would still be able to be classified as 'a matter'. Conversely, the 'narrow view' requires that a matter is only established in a proceeding if the 'subject matter for determination', the 'right duty or liability to be established' and the 'controversy' that needs to be quelled can each be specifically identified. In Griffiths v Tang, a High Court majority adopted the narrow view.

Although there are persuasive arguments for holding that the making of a declaration of incompatibility is an exercise of judicial power and also a "matter" for the purpose of Chapter III of the Constitution, I think the more probable view is that the High Court will adopt the narrow view of "matter" and invalidate the declaration of incompatibility mechanism.

[The footnotes for this paper ae available on request to the Evatt Foundation.]

Also on the Evatt site:

Suggested citation
McHugh, Michael, 'A human rights act', Evatt Journal, Vol. 9, No. 1, March 2009.<http://evatt.org.au/papers/human-rights-act.html>