Sydney Law Review
This article could well have been entitled, ‘To recognise or not to recognise same-sex marriages in the United States, Canada, Australia and New Zealand’. The article was prompted by developments in the United States and, to a much lesser extent, Canada, and also Australia in view of federal legislation passed in 2004 that was designed to preclude the recognition of such marriages in this country. It seemed useful at the same time also to take account of older developments in New Zealand on the same subject. The comparative survey provides a fascinating interplay of constitutional and statutory interpretation, federalism, the role of the judiciary and also the constitutional aspects of private international law. In addition it calls attention to the perennial issue of how far the courts can act contrary to public opinion.
The origin of the problem canvassed in this article can be summarised in the following way. It began by calls for the recognition of same-sex marriage in the United States that were made primarily through the courts rather than by seeking a change in the law by legislation. These calls led to State judicial decisions which decided that the explicit failure of the common law and statutory definitions of marriage to include such marriages violated the Equal Protection clauses of certain State Constitutions. This resulted in considerable public confusion when city officials in other States began licensing same-sex marriages despite State legislation which defined ‘marriage’ as the voluntary union of a man and a woman. These developments stirred up much anger on the opposing sides of the debate in the United States which, in turn, led to calls for constitutional amendments to prevent the recognition of such marriages. The developments in the United States can be contrasted with a more muted response to decisions in Canada which ultimately led to the recognition and greater public acceptance of same-sex marriages in that country.
As I will later demonstrate, although the problem in Australia has yet to reach the same depths of political intensity, it did call attention to a debate on the capacity of the Commonwealth Parliament to legislate for such marriages in the exercise of its legislative power over ‘marriage’. This was an issue which also arose in Canada because of the power of the Dominion Parliament to legislate with respect to marriage.
The accepted dictionary meaning of the term ‘marriage’ is:
‘[M]arriage’ 1. the legal union of a man and a woman in order to live together and often to have children. 2. an act or ceremony establishing this union …
The American dictionary meaning is not noticeably different, except for reference to ‘the institution whereby men and women are joined in a special kind of social and legal dependence for the purpose of founding and maintaining a family’.
Doubtless those meanings reflect the historical, cultural and religious understanding of the concept of marriage which confines the relationship to persons of the opposite sex and its associated concern for the procreation of children. This is so even though the capacity to bear children or the fulfilment of that capacity has not been made a condition of the creation or continuance of a valid marriage. The latter aspect is underlined by the reference in the Australian definition quoted above to the purpose of marriage as being: ‘often to have children’ (emphasis added).
Nevertheless, the traditional meaning has to be counterbalanced with an acknowledgment of the capacity of the relationship to be affected by significant legal and social change which shows that the meaning is not immutable. This can be demonstrated without necessarily accepting suggestions that same-sex marriages were accepted and celebrated ‘in ancient Greece, Mesopotamia, Rome, and even Christian states’. Perhaps the outstanding illustration is provided by the fact that marriage is no longer treated as a relationship for life. The concept has also travelled a long way since the primitive practice of endogamy, which was a restriction based on the practice of marrying someone within one’s own tribe or group — apparently one of the oldest social regulations of marriage. A variation of this kind of restriction was the law in ancient Greece which prohibited Athenians from marrying foreigners. In some countries there were also prohibitions on the marriage of persons of different races (miscegenation laws). It was not until modern times that marriage became a matter of free choice.
For these purposes it is also instructive to contemplate the concept of ‘open marriage’ as practised by some in the community which makes nonsense of another supposedly essential feature of marriage, namely, that marriage is a ‘voluntary union … to the exclusion of all others’ as is currently emphasised in the statutory definitions of that term. Although the open marriage concept is obviously not presently reflected in the law or gaining acceptance as a feature of marriage, it needs to be remembered that, however belatedly, the law has a tendency to catch up with social practices as they evolve.
It is convenient to begin by quoting the prescient observations of Professor Cass Sunstein in the course of accepting the arguments in favour of recognising that the exclusion of same-sex marriage violates the Equal Protection Clause of the United States Constitution. Despite that acceptance, he wrote both prophetically and somewhat paradoxically:
An immediate judicial vindication of the principle could well jeopardise important interests. It could galvanise opposition. It could weaken the antidiscrimination movement itself. It could provoke more hostility and even violence against gays and lesbians. It could jeopardise the authority of the judiciary. It could well produce calls for a constitutional amendment to overturn the [United States] Supreme Court’s decision. At a minimum, courts should generally use their discretion over their dockets in order to limit the nature and the timing of relevant intrusions into the political process. Courts should also be reluctant to vindicate even good principles when the vindication would clearly compromise other important principles, including ultimately the principles themselves.
Earlier he wrote in the same book:
I believe that at the national level and in the short term, the [United States] Supreme Court should be extremely reluctant to require states to recognise same-sex marriages. It is far better for these developments to occur at the state level, usually through legislatures but sometimes through courts …
These remarks will of course call for amplification.
The modern judicial recognition of same-sex marriages in the United States begins with their recognition in a number of cases decided by State courts on State constitutional grounds. They appear to involve the interpretation of State constitutional guarantees of equality — as distinct from the guarantees of due process, as was the case with the invalidation of criminal laws which prohibited homosexual conduct between consenting adult males in Lawrence v Texas. No attempt is made in this article to deal with the additional reliance that was placed on that ground as well; or to dwell in great detail on the degrees of judicial scrutiny which were held to apply in determining whether, to the extent that laws excluding the recognition of same-sex marriages were required to be justified, they could be seen to further countervailing legitimate public interests.
The decisions provoked, on the one hand, community anger and criticism directed at ‘activist judges’, and on the other, moves to have the same recognition accorded under similar guarantees in other States. It seems difficult for those in countries which have yet to adopt a judicially protected Bill of Rights to understand the depth of that public anger in a country which cherishes the judicial protection of rights. As the late Justice William Brennan in the United States observed in 1985 in response to the asserted need to leave substantive value choices to the ordinary democratic processes, the very purpose of a Bill of Rights is to declare certain values beyond the reach of temporary political majorities. The derogation from the right of the majority to decide is thought to be justified because the majority cannot be expected to rectify the claims of the minority that arise as a response to the outcomes of the majoritarian processes.
It is striking to note how the Supreme Judicial Court of Massachusetts not only invalidated the laws which excluded same-sex marriages in 2003 but also subsequently decided in 2004, in an advisory opinion, that a law which would have made provision for civil unions between persons of the same sex would also have violated the Equal Protection guarantee, even though the partners to such a union would have enjoyed the same rights and duties as partners to a traditional marriage. The essential flaw in such a law was thought to be the failure of the law to label the civil unions as ‘marriages’.
It is also worth mentioning in this connection the first and leading Canadian decision which upheld the recognition of same-sex marriages in that country by reference to the guarantee of equality contained in s 15(1) in the Canadian Charter of Rights and Freedoms. This was decided by the Ontario Court of Appeal in Halpern v Canada (Attorney-General) in 2003.
The laws that were held invalid for not recognising the efficacy of same-sex marriages discriminated between those marriages and marriages between persons of the opposite sex despite the argument that the laws prohibited both men and women from doing the same thing, namely, marrying persons of the same sex. Although it is true that the failure to recognise same-sex marriages does prohibit both men and women from doing the same thing, this argument ignores the discriminatory effect or impact of such a prohibition on the sexual orientation of homosexual persons. The argument was rejected essentially because it perpetuates a view that same-sex couples are less capable or worthy of recognition or value as human beings, to use the language used by the Ontario Court of Appeal. It was the same kind of argument which was also raised and rejected by the United States Supreme Court when it invalidated laws prohibiting persons of different races marrying each other in Loving v Virginia. This was essentially because miscegenation laws were designed to keep people of different races apart and were founded on views of racial superiority.
If discrimination includes the different treatment of like persons or things without a rational reason, was there a rational reason based on legitimate interests of society to justify the differential treatment of both kinds of marriage? The courts have either implicitly or explicitly taken the view expressed by the Ontario Court of Appeal when it emphasised that it is not enough to show that historically and according to religious beliefs marriage was inherently limited to opposite-sex relationships; nor was it enough to assert that marriage is ‘heterosexual because “it just is”’ because this was thought to amount to circular reasoning. In the same vein, the Massachusetts Supreme Judicial Court stated that ‘it is circular reasoning, not analysis, to maintain that marriage must remain a heterosexual institution because that is what it historically has been’. This view seems to be sound, subject to whether the morality of the majority in a community is sufficient by itself to constitute a ‘legitimate and compelling state interest’ when it is based on strongly held historical and religious beliefs.
Leaving that consideration aside, it is necessary to describe briefly the most important social and public interests that were advanced by the State to justify the non-recognition of same-sex marriage. In the first place it was argued that such laws provided a favourable setting for procreation. The obvious reply to that argument was that partners to a valid marriage were not required to show a capacity to procreate before or after the marriage is solemnized. One could, presumably, also rely on modern technological developments regarding the procreation of children as well as changes in adoption laws which increasingly allow persons of the same sex to adopt children.
Secondly, it was argued that laws that do not recognise the efficacy of same-sex marriages ensure an optimal setting for child rearing. This was rejected because the argument ignores changes in the diverse composition of modern American families and changes in laws relating to adoption and legitimacy. This consideration no doubt presupposes that persons of the same sex have been accorded the same or similar rights as those granted to persons of opposite sex under those laws. The State had failed to establish that the failure to recognise will increase the number of couples who will choose to enter into opposite-sex relationships in order to raise children. Such laws could in fact make the setting worse if same-sex relationships are not recognised, given that the children raised by persons of the same sex would be punished because of the stigma attached to the relationships entered into by their parents. Without dealing with them in any detail, the courts also seemed to have little difficulty in rejecting a number of other arguments: the importance of ensuring uniformity and avoiding conflict with other States given that most States did not recognise same-sex marriages; the assertion that it would de-stabilize the institution of marriage; and that it could be assumed that same-sex couples are more financially independent than married couples and thus less needy of public marital benefits, such as tax advantages, or private marital benefits, such as employer-financed health plans that include spouses in their coverage.
The cases discussed above give rise to an acute problem regarding the availability of appropriate judicial relief for dealing with the effect of invalidity which results from any form of discrimination. In this case the failure to provide for same-sex marriage could have invalidated the whole of the legislation dealing with marriage when the clear intent of any guarantee of equality in this context would almost certainly have been to preserve the legal facility of marriage and extend its availability to the form of marriage that was not recognised in the legislation. This called for a degree of judicial creativity, which has been encountered before in relation to the effect of discrimination in other contexts. The problem was resolved in Goodridge by the court ‘refining’ the definition of marriage to mean the ‘voluntary union of two persons as spouses to the exclusion of all others’ but suspending the effect of its judicial declaration for 180 days to permit the legislature to take such action as it deemed appropriate in light of the opinion of the Court. The court in Halpern followed the same course, except that it did not see the need to suspend the effect of its declaration under which the definition of marriage was declared invalid to the extent that it referred to a union between ‘one man and one woman’. The definition was reformulated so as to read ‘a voluntary union for life of two persons to the exclusion of all others’. That remedy was best thought to achieve the equality required by s 15(1) of the Canadian Charter of Rights and Freedoms at the same time as ensuring that the legal status of marriage was not left in a state of uncertainty.
Reference was made earlier to the considerable public confusion generated when city officials in some States began licensing same-sex marriages despite State legislation which defined marriage as the voluntary union of men and women. This provoked at least one successful legal challenge to their authority to license those marriages. The Supreme Court of California decided, in effect, that city officials could not take it upon themselves to assume the invalidity of a duly enacted statute in anticipation of a judicial declaration of invalidity of that legislation. The decision was not, however, to be taken as indicating one way or the other what view the court would take about the validity of those statutes. The case highlights the difficult situation that can arise when public officials are faced with administering a law which they have reason to believe may be constitutionally invalid. Despite the dilemma such officials faced, it was thought that the preferable course in this instance was to apply the law as it stood and allow affected persons to challenge the application to themselves of the allegedly invalid law.
Perhaps because of the adverse public opinion generated by the decisions discussed above, courts in other States subsequently declined to accord recognition to same-sex marriages despite the attempts that were made to rely on similar constitutional grounds available under the constitutions of those States. In fact, the number of States where this has occurred, when coupled with the number of States that adopted amendments to their own State Constitution which purport to prevent such recognition, now far outnumbers those States where some form of recognition has been accorded. The former jurisdictions even include States such as New York and California which have been regarded as traditionally liberal in outlook in the past.
The differences with the cases already discussed are striking. First, there was a partial or full acceptance of the view that there was no discrimination because both sexes were treated in the same way. The case of Loving v Virginia was distinguished essentially on the ground that it dealt with laws based on racial superiority and race discrimination.
Secondly, the State legislatures were entitled to believe that the challenged laws were rationally related to the furtherance of the public interests that were unsuccessfully relied on to sustain such laws in Goodridge, namely, the procreation and raising of children. Although it was acknowledged that same-sex couples could ‘become parents by adoption, or by artificial insemination or other technological marvels’ it was stated that ‘they do not become parents as a result of accident or impulse’ — although the precise relevance of accidental or impulsive procreation was not made clear. In one case it was emphasised that procreation was essential to the survival of the human race, while in another it was asserted that same-sex relationships were not as stable as opposite-sex relationships. Surveys establishing that children of same-sex relationships did not suffer from being raised in such families were rejected on the ground that there had not been enough time to study the long-term results of child rearing by same-sex couples. These considerations prevailed despite the failure to show why persons of the opposite sex could legally marry even if they did not have the capacity to bear children by normal means, so as to create an over-inclusive class of persons allowed to marry by reference to the public interest asserted; and also creating an under-inclusive class of persons who were not allowed to marry because they were not of the opposite sex if the sole purpose of marriage was not child-procreation or child-raising. There was also a failure to address why the relevance of the considerations relied on was not confined to the rights of same-sex persons to adopt and raise children, as distinct from their right to marry each other. In effect, the considerations relied on showed why allowing persons of the opposite sex to marry did not further the interests concerning procreation of children. But they failed to explain how not allowing the marriage of (same-sex) persons who could not have children, at least in the same way furthered that interest (or, for that matter, interfered with its furtherance) since it was not shown that non–recognition would encourage persons of the same sex to change their ways.
Underlying all these considerations was the strong reliance placed on history and tradition — the very considerations which were rejected in Goodridge and Halpern as being insufficient by themselves to justify the refusal to recognise same-sex marriages. Strong reliance was also placed on the application in these cases of the highly deferential rational basis of review. Under this standard, the court may assume the existence of ‘any conceivable state of facts that could provide a rational basis for classification’. This standard may be contrasted with the strict or heightened form of judicial scrutiny in order to justify the validity of certain laws challenged under both the Equal Protection and Due Process clauses. Persons of the same sex were not seen as being members of a ‘suspect class of persons’ who were entitled to heightened judicial scrutiny and neither had it been shown that previous Supreme Court cases had recognised as a ‘fundamental right’ the right to marry persons of the same sex. The failure to recognise such a right also has the effect of preventing the federal and State Due Process clauses from being construed as requiring the recognition of same-sex marriages as part of the right to privacy.
The final appellate court in New Jersey also strongly relied on history and tradition as a reason for refusing to, in effect, re-define marriage to include same-sex marriages — at least by judicial decision-making. But that Court recognised all the force of the considerations relied on in Goodridge to uphold a constitutional obligation on the State to provide a parallel system of law to recognise same-sex relationships and confer upon the members of such relationships the same rights and benefits as those conferred on persons who were married in the traditional sense with one significant difference. That difference was that such a relationship need not be labelled as a ‘marriage’. Significantly, the State in that case did not seek to place any reliance on the interests of child-procreation and child-raising in order to justify the exclusion of same-sex marriages. The Court ignored the growing trend away from the recognition of same-sex relationships by recognising the ability of States in a federal country to serve as social and economic ‘laboratories’; and that equality of treatment was a dominant theme of the laws of New Jersey and the central guarantee accorded to it by that State’s Constitution. This was seen as fitting for a State with so diverse a population. The result was to put New Jersey in the same class as Vermont which has created a system of civil unions. The result represents a compromise between the growing recognition of the unfairness of discriminating against homosexual persons on the one hand and, on the other, the difficulties in the way of courts, as distinct from the other branches of government, treating the term ‘marriage’ as encompassing same-sex relationships when interpreting constitutional guarantees of equality. Those difficulties were not, however, thought to be sufficient to prevent courts from recognising the need to provide a parallel system of rights and obligations for same-sex partners because of the same guarantee.
Enough has been said above to understand why Professor Sunstein thought that there was a real chance that the Supreme Court of the United States may one day accept that laws that fail to recognise same-sex marriages breach the Equal Protection Clause of the United States Constitution. But it is important to recall how he relied on prudential considerations to postpone testing that proposition before the nation’s highest court in favour of first deferring to State legislative or judicial decisions. Apart from fully vindicating his predictions regarding the likely public reaction to the judicial recognition of same-sex marriages, the advice he gave appears to have been heeded since some gay lobby groups have decided not to challenge the constitutional validity of State constitutional amendments which have sought to prevent similar developments in many other States. Those States illustrate one disadvantage of relying on State action to recognise same-sex marriages. I will deal later with another major disadvantage of that course of action. That disadvantage relates to the vulnerable and uncertain status of such marriages in States that do not recognise them and the ability of same-sex partners to obtain divorces in relation to such marriages in those States. At this point it is only necessary to emphasise that under the United States federal system of government the power to make laws to with respect to marriage rests, for the most part, with the State legislatures and not Congress. It is open to Congress to define what it means when it uses the term ‘marriage’, subject to compliance with any rights guaranteed by the Constitution.
It is convenient at this stage to turn to the position in Australia. As is well known, the absence of a constitutional Bill of Rights forms a major source of difference between the Australian and United States constitutions despite the recent developments which have recognised some implied constitutional protections. Notwithstanding some judicial suggestions to the contrary, an implied right to equality does not appear to be one of those protections. Accordingly, Australia presently lacks a national constitutional guarantee of equality. Essentially it remains the case that the Australian Constitution is generally based on trust rather than mistrust.
So the matter in Australia must be decided by legislation. This is so in relation to same-sex marriages celebrated in Australia (domestic marriages). The recognition of the same marriages celebrated overseas will depend on the common law principles of private international law as modified by any relevant legislation (foreign marriages). The critical issue becomes legislation passed by which Parliament: federal or State?
Dealing first with domestic marriages, the issue is whether the national Parliament can rely on its power to make laws with respect to ‘marriage’ to make a uniform Australian law which would recognise same-sex marriages as a marriage within the meaning of that term in s 51 (xxi) of the Constitution — in other words as part of the subject matter of that power. This involves a question concerning the progressive principles of constitutional interpretation and how the courts interpret the meaning of a constitutional term. Orthodox principles insist on concentrating on the essential meaning which constitutional terms had as at the date when the Constitution was enacted in 1900, although there is now at least one member of the High Court who does not subscribe to that orthodoxy; namely, Kirby J who prefers to rely on the meaning which constitutional terms have now. This inevitably gives rise to a familiar debate concerning original intent and originalism.
But even the orthodox approach is tempered by two major considerations. The first is that even that approach concentrates on the essential rather than non-essential meaning of terms. Secondly, it has long been acknowledged that there is a need to interpret constitutional powers broadly, given the difficulty of amending the Constitution and the need to ensure that it adapts to new developments not foreseen by the framers. To take a hypothetical example, if the Commonwealth Parliament had been given the power to legislate with respect to ‘transportation’, new forms of transportation not contemplated at the time the power was first conferred, whether in the Constitution as originally enacted or as subsequently amended, would still be treated as coming within that power. Actual examples can be drawn from the power to make laws with respect to ‘postal, telegraphic, telephonic, and other like services’ in s 51(v) in relation to radio and television broadcasting and now almost certainly the internet as well.
What is different about the changes that may have occurred in relation to same-sex marriages is that those changes relate to cultural and social values in contrast to changes which involved scientific developments and inventions. It has been said that the power of the Commonwealth Parliament to legislate with respect to marriage ‘is predicated upon the existence of marriage as a recognizable (although not immutable) institution’. At the time of federation the meaning of the term ‘marriage’ most commonly acknowledged was that contained in the cases which refused to recognise foreign polygamous marriage because such unions did not satisfy the traditional meaning of marriage now explicitly embodied in the Marriage Act 1961 (Cth). Not surprisingly this will make it difficult for the Court to accept that same-sex marriages now come within the meaning of the term ‘marriage’ in s 51(xxi) of the Commonwealth Constitution — a view that has already attracted some judicial support.
Although difficult and probably unlikely at the moment, despite the progressive nature of the principles of constitutional interpretation mentioned above, it is however by no means impossible, given the inherent flexibility of the relevant principles of constitutional interpretation. The possibility that the term ‘marriage’ may be interpreted as being wide enough to include same-sex marriage is now contemplated by some judges and scholars, some of whom subscribe to the orthodox principles of constitutional interpretation. Judicial developments in Canada provide support for that view. The Ontario Court of Appeal in Halpern’s case accepted that marriage could now encompass same-sex marriage within the meaning of the same term in the Constitution Act 1867 (UK) s 91(26). That Court observed:
In our view, ‘marriage’ does not have a constitutionally fixed meaning. Rather like the term ‘banking’ in s 91(15) and the phrase ‘criminal law’ in s 91(27), the term ‘marriage’ has the constitutional flexibility necessary to meet changing realities of Canadian society without the need for recourse to constitutional amendment procedures.
The latter view was emphatically endorsed by the Canadian Supreme Court in its advisory opinion regarding the power of the Dominion Parliament to legislate with respect to same-sex marriages.
However, Canadian authority cannot necessarily control the outcome in Australia, given certain differences regarding the distribution of legislative powers between the Australian and Canadian Constitutions and the effect of those differences on the ways courts in both countries characterise laws. One of those concerns the exclusive nature of the legislative powers provided under the Canadian Constitution and the division of the powers in relation to the solemnisation of marriage and other aspects of marriage. Associated with this consideration was the significance of the Provincial power to make laws on solemnisation of marriage in determining whether the residue of authority in relation to marriage should be vested in either the Dominion or the Provincial Parliaments. It was significant because of ‘the principle of exhaustiveness’ which was seen as an ‘essential characteristic of the federal distribution of powers’ in Canada. That principle ‘ensures that the whole of legislative power, whether exercised or merely potential, is distributed as between’ those Parliaments subject only to the guarantee of the freedoms and rights contained in the Charter of Rights and Freedoms. Presumably the exclusion of same-sex marriage from the respective Dominion and Provincial powers over the relevant aspects of ‘marriage’ in both ss 91(26) and 92(12) of the Canadian Constitution Act 1867 would have meant that neither of those Parliaments would have had the power to deal with such marriages, assuming that the same power did not also come within the residual Dominion legislative power in s 91(13).
Further support for the power of the Commonwealth Parliament to legislate for the recognition of same-sex marriage under the marriage power may be provided by the purposes of the power: to ensure uniformity and mobility. Although by no means compelling, the adoption of a functional approach to characterisation may also assist in the same direction, given the need for Parliament to control areas that are at least closely related to traditional marriages. That said, I would not be surprised if a majority of justices of the High Court felt unable to ignore the strong influence of history and traditional understanding in interpreting the essential meaning of the term ‘marriage’ if the issue was tested now. On the other hand, perhaps the longer the issue is postponed for decision in the future, the greater will be the chances of its eventual acceptance.
For the sake of completeness it only remains to mention in this connection that the case of Russell v Russell suggests that the marriage power extends to dealing with the rights and duties which arise out of a marriage. If the focus is placed on those rights and duties, the fact that two persons were not themselves married, or even capable of being married, would not prevent the conferral of the rights and duties of marriage on those persons. However, the High Court may have assumed that the power of Parliament to deal with such rights and duties still presupposes the existence of a marriage — in which case the ability of the Commonwealth Parliament to provide for same-sex marriage would continue to depend on the meaning of the term ‘marriage’ and what is incidental to the furtherance of the power to make laws with respect to ‘marriage’.
It is likely that ‘marriage’ was probably already confined to unions between persons of the opposite sex with that term being defined as a ‘[u]nion of a man and a woman to the exclusion of all others’ under the federal Marriage Act, even before it was amended in 2004. The amending legislation was designed to put this beyond any doubt. But it is worth mentioning that the Full Family Court of Australia has interpreted such a definition to include a person registered as a female at birth who subsequently became a male transsexual by a medical operation so as to be capable of marrying as a male.
Even if the Commonwealth Parliament can legislate to recognise same-sex marriages, its failure to do so did not in my view invalidate the Marriage Act as amended. There is no legal obligation on Parliament to exercise the totality of its legislative powers and the nature of the subject matter of the power with respect to marriage is not such that the failure to legislate for all kinds of marriage would prevent the legislation being characterised as one with respect to ‘marriage’. Neither is there a general prohibition contained in the Commonwealth Constitution on the enactment of discriminatory legislation.
So far as same-sex marriages celebrated overseas are concerned, Parliament could probably rely on the external affairs power to provide for recognition of foreign same-sex marriages regardless of the doubts under the marriage power. The legislation passed in 2004 for the opposite purpose of precluding such recognition is probably valid under both the marriage and external affairs powers. The legislation in question was designed to eliminate suggested doubt regarding whether under the relevant provisions of the Marriage Act 1961 (Cth) Part VA same-sex marriages celebrated in countries that recognise them — such as Canada and Denmark — were required to be recognised in Australia. The doubt was alleged to exist despite the well known case of Hyde v Hyde (1866).
It only remains to advert to the effect it would have on State legislative power to deal with same-sex marriages, if the term ‘marriage’ in s 51(xxi) of the Constitution did not include such marriages. Because State Parliaments enjoy general residual powers of legislation, it would seem that State legislative power must be available to recognise domestic same-sex marriages, as has occurred with legislation to deal with rights and incidents of de facto relationships. This, however, is subject to the possible ability of the Federal Parliament to ban the description of same-sex unions as ‘marriages’ in the exercise of the incidental power — even if it could not provide for the creation and regulation of those marriages. In my view, it would be open to the Commonwealth Parliament to pass laws that prevent the term ‘marriage’ being confused with or mistaken about a relationship which was not described as a ‘marriage’ for the purposes of comprehensive federal legislation on that topic.
Although not accepted by all, I believe it is fairly arguable that the provisions of the Marriage Amendment Act 2004 (Cth) achieved that objective, even though they did not make it an offence for private individuals wrongly to use that term to describe a same-sex union. I suggest that much depends on the highly symbolic significance attached to the use of the term ‘marriage’. It is strongly arguable that the amending legislation has attempted to define exhaustively which relationships may be described as ‘marriages’ so as to confine the use of that description to the kind of traditional marriage referred to in the definition of ‘marriage’ in s 5(1) of the Marriage Act. In the words used in the Minister’s second reading speech, the Marriage Amending Act was designed ‘to provide certainty to all Australians about the meaning of marriage in the future’. It is true that that legislation contains explicit provisions which require foreign unions between persons of the same sex not to be recognised as a marriage in Australia but fail to make similar provision in relation to domestic unions of that kind. However, there are dangers in applying the principle of construction embodied in the maxims expressio unius est exclusio alterius and expressum facit cessare tacitum, and it is possible that the same kind of provisions for those unions may have been thought unnecessary from a technical drafting point of view since such marriages would not be governed by any foreign system of law in any sense. In the final analysis, it would seem highly odd that the Marriage Amendment Act would treat both kinds of same-sex unions in a different way.
The foregoing discussion deals with the federal legislative power to prohibit the creation under State law of same-sex unions when those unions are described as marriages. Whatever may be the position as regards such power, it is much more doubtful whether the Commonwealth Parliament could go further and ban the creation of civil unions to which were attached the same rights and duties as those which arise out of a marriage relationship under the argument adverted to above. According to that argument, the power deals with the rights and duties of marriages which arise independently of the existence of a marriage. The Marriage Act as it stands cannot be said to cover the field in relation to the law which governs the rights and duties of the partners to a same-sex union, leaving the way open for such matters to be governed by the States — at least, on the argument advanced above, if the union is not described as a ‘marriage’. This was conceded by the Government when the Marriage Amendment legislation was debated in Parliament.
But absent any valid federal legislation to the contrary, and as in the United States, the key question would then arise is what effect would such (State) civil unions — whether or not they are described as marriages — have in other States or parts of Australia and that brings us to the issue of full faith and credit.
The federal constitutions of both the United States and Australia have similar provisions which require full faith and credit to be accorded to judgments and public Acts of sister States and also provide federal legislative power to give effect to this guarantee. Thus art IV, § 1 of the United States Constitution states:
Full faith and credit shall be given in each State to the public acts, records, and judicial proceedings of every other State. And the Congress may by general laws prescribe the manner in which such acts, records, and proceedings shall be proved, and the effect thereof.
Section 118 of the Commonwealth Constitution states:
Full faith and credit shall be given throughout the Commonwealth to the laws, the public Acts and records, and the judicial proceedings of every State.
Section 51(xxv) empowers the Commonwealth Parliament to make laws with respect to ‘the recognition throughout the Commonwealth of the laws, the public Acts and records, and the judicial proceedings of every State’.
Quick and Garran suggested that the guarantee of full faith and credit in the Commonwealth Constitution:
contains a constitutional declaration in favour of inter-state official and judicial reciprocity, which the Federal Parliament and the States may assist to effectuate, but which they cannot render nugatory.
The same learned authors treated the Constitution as converting ‘rules of international and inter-state comity, as well as the common law’ (ie, the rules of private international law) which were capable of being altered or abolished by the States into a ‘rule of law in order to promote uniformity of regulation’. This was, however, seen as providing a remedy against States which sought to deprive residents of other States from asserting their rights and privileges in the judicial proceedings of the former States in times of antagonism and contention between States. Whatever may have been the intended purpose of the guarantee, it is fair to surmise that this kind of constitutional guarantee has the effect of facilitating national unity to enable citizens to travel freely within a federation without being denied their legal rights and also avoiding the inefficient re-litigation of legal rights.
It might, therefore, be thought that the result of the constitutional obligation to accord full faith and credit would be to ensure that if same-sex unions were recognised as marriages in a State such as Massachusetts, and civil unions in Vermont, those relationships would be recognised in another State such as Virginia, even if it opposes those relationships. A similar conclusion might be drawn in Australia if one or more States did provide for the creation of those unions in the absence of any inconsistent federal legislation when other States failed to provide for such unions. The recognition of the union in those States that failed to provide for same-sex unions could prove critically important to the partners of those unions once one or both of them reside in those States in relation to such matters as the dissolution of same-sex marriages or civil unions and also the custody and maintenance of children of those relationships. Leaving aside constitutional questions of full faith and credit, these kinds of issues would be governed by the application of principles of private international law, and evoke in Australia the kinds of issues that used to arise in relation to similar issues before the advent of uniform federal marriage and divorce legislation in the middle of the last century. No attempt is made here to essay the application of the common law principles of private international law since the focus of attention is directed to the constitutional guarantee to accord full faith and credit.
As surprising as it may seem, there are still murky questions about whether the United States guarantee would still allow a State to refuse to recognise an interstate marriage if it regarded such marriages as contrary to its own public policy. That uncertainty persists despite the relatively long history of the United States Constitution. Although it has been suggested that such a qualification should have little room to operate in a federal country, that view has not been reflected in a recent case decided by a federal District Court. In Wilson v Ake it was held that the full faith and credit clause did not preclude the courts of a State from refusing to recognise marriages celebrated in accordance with the laws of a sister State on grounds of public policy. And, so far, there does not seem to be any case decided by the United States Supreme Court which denies the correctness of that view. Until that Court accepts that the Full Faith and Credit Clause prevents reliance on public policy as a ground for refusing recognition to a marriage contracted in a sister State, there seems little likelihood that that clause will have the effect of requiring the recognition of same-sex marriages.
In Australia the High Court has, however, recently decided that such a qualification cannot operate in the Australian setting as between the laws and judgments of sister States. The National Cross-vesting Scheme also offers fascinating and interesting possibilities for the application and enforcement of legislation which is otherwise valid and provides for same-sex marriages or civil unions, in the courts of Australian jurisdictions other than the State in which the legislation was enacted. If sound, those possibilities may obviate the necessity for reliance on s 118 of the Constitution.
Rather than provide an answer to whether the full faith and credit guarantee requires the recognition of interstate same-sex marriages or civil unions the position in the United States is complicated by the enactment in 1996 of Congressional legislation known as the Defense of Marriage Act (‘DOMA’) which was passed to meet the fears generated by the the Hawaiian State court decision recognising same-sex marriages. The legislation purports to relieve rather than oblige States and their courts from any obligation to recognise such marriages or unions. This seems surprising, at least to outside observers, since the legislation was passed as an exercise of the power of Congress to pass ‘general laws’ which are supposed to prescribe ‘the manner’ in which the relevant ‘acts, records, and proceedings ‘ of a sister State ‘shall be proved’ and provide for ‘the effect thereof’.
It has been said of this power that ‘[i]ndeed, there are few clauses of the Constitution, the merely literal possibilities of which have been so little developed as the Full Faith and Credit Clause’. The question therefore arises whether:
(a) DOMA is valid as the kind of law Congress can pass to regulate the manner of proof and the effect that is given to the laws and judgments of sister States; or
(b) whether that Act is invalid, as a noted American commentator and others have argued.
Arguably, the power of Congress is limited to facilitating the enforcement of the full faith and credit guarantee and regulating the procedural requirements that have to be satisfied to attract the benefit of its operation, as distinct from creating substantive qualifications which detract from giving effect to the guarantee in question.
Despite that argument, some State courts have, however, assumed the validity of the Act and the argument was squarely rejected in Wilson. It was held in that case that to accept that the Full Faith and Credit Clause requires the recognition of same-sex marriages in sister States would enable the States which recognised such marriages to create a national policy — a view rejected by the United States Supreme Court in relation to laws dealing with the right to receive compensation for injury. Whether a different policy needs to be accepted for laws that deal with personal status in a federal nation, especially given the general policy of sustaining marriages would seem to be a question that deserves further consideration.
A further point might be made if the effect of the recognition was more restricted and contained so as to deny the resulting creation of the suggested ‘national policy’. The restriction and containment would follow if two conditions could be satisfied. The first would be that the effect of the constitutional guarantee was confined to recognising only those interstate marriages which were required to be recognised under the common law rules of private international law — but without having regard to the public policy of the forum State. The second condition would be that, according to that approach, the mere fact that a marriage was created under a law of a sister State would not be enough to ensure recognition if the parties to the marriage were domiciled in a State which did not recognise same-sex marriages. However, it appears that, although there is a substantial number of cases denying validity to an interstate marriage contracted by parties who are forbidden to enter certain marriages by the law of their domicile, the general rule is that marriages are valid wherever they are entered into without making any distinction between the form and capacity to marry.
It was also held in the Wilson case that DOMA was exactly the kind of legislation the Framers envisaged would be enacted by Congress, although no authority was cited in support of that holding. It therefore seems that, on present indications, the prospects for successfully challenging the validity of DOMA do not seem to be promising, even if the arguments in favour of invalidity seem quite strong.
Similar issues would seem to arise in relation to Australian federal legislation of the same kind under the power to legislate for ‘the recognition of the public Acts and records, and the judicial proceedings of the States’ in s 51(xxiv) of the Constitution. The latter power is contained in s 51 which is, of course, prefaced by the words ‘subject to this Constitution’. For present purposes this must include whatever is encompassed by the obligation of Australian courts to accord full faith and credit to the laws of sister States in s 118. It seems difficult to deny the soundness of the assertion made by Quick and Garran and quoted above, that the ‘Federal Parliament … may assist to effectuate, but … cannot render nugatory’ the obligation contained in s118. That said, the scope of that obligation, as with its United States counterpart, remains unclear. But, at the very least, what is now clear is that the inability of State courts to deny recognition to laws of sister States on the ground that they conflict with the public policy of the forum should severely limit the extent to which same-sex marriages celebrated in one State can be refused recognition under the applicable common law rules of private international law in Australia.
6. Proposed and Actual Federal and State Constitutional Amendments Banning Same-sex Marriages in the United States
Reference was made earlier to the prediction made by Professor Sunstein that judicial vindication of claims to recognise same-sex marriages could well produce calls for a constitutional amendment to overturn judicial decisions which vindicated those claims. This prediction has come true.
There are, of course, ample powers of constitutional amendment at both the national and State levels in the United States. The most important amendment to the United States national Constitution that was proposed for preventing the recognition of same-sex marriages was the Federal Marriage Amendment which enjoyed the support of President George W Bush during the Presidential elections held in November 2004. Although it was proposed in both Houses of Congress and enjoyed strong support in those Houses, it was not passed during the life of that Congress (108th) or the 109th Congress despite the re-election of President Bush in 2004 and his strong support of the proposed amendment. Although it was re-introduced into the 110th Congress in substantially the same terms as the previous House versions, there does not seem to have been any action to have the issue debated at the date this article was written.
It is nevertheless worth considering the terms of the same proposed amendment. In its original form, clause 2 of the Senate version provided that:
Marriage in the United States shall consist only of the union of a man and a woman.
Neither this Constitution, nor the constitution of any State, shall be construed to require that marriage or the legal incidents thereof be conferred upon any union other than the union of a man and a woman.
The House of Representatives version was identical, except for the addition of a further provision in clause 3 which stated:
No State shall be required to give effect to any public act, record, or judicial proceeding of any other State concerning a union between persons of the same sex that is treated as a marriage, or as having the legal incidents of marriage, under the laws of such other State.
There can be little doubt that this version was intended to remove the doubts about the validity of DOMA, although it may have had the interesting effect of assuming the continued ability of States to provide for the recognition of same-sex marriages or civil unions. The latter point may prove significant in view of the doubts that surrounded the meaning and reach of clause 2.
The provisions of clause 2 of the Federal Marriage Amendment gave rise to debate about its meaning and operation. The absence of agreement about its effect probably reflects more than a disagreement about linguistic analysis, although it is easy to see why the wording of the proposed amendment gave rise to a debate about the meaning of its wording. The controversial nature of the issue about the legitimacy of same-sex marriages and civil unions divided even those who objected to the recognition of such relationships. At the minimal end of the spectrum there were those who believed that the proposed amendment would have banned only the recognition of those relationships if they were formally described as ‘marriages’. This position was supported by President Bush. Changes were made to the proposed amendment that were designed to make clear that it was not intended to bar civil unions allowed by State law. Normally it might be thought that it should not seem to matter much as long as relationships are placed on the same footing, however those relationships are described. To quote the well-known words of Shakespeare in Romeo and Juliet quoted in a case mentioned earlier:
What’s in a name? That which we call a rose
By any other name would smell as sweet. …
But this ignores the highly symbolic significance of the term ‘marriage’ and there may be many in the community who feel that the term should be exclusively confined to its traditional and religious meaning, but without objecting to otherwise placing same-sex marriage on the same footing. This interpretation of the proposed amendment is derived from the focus of the proposed amendment on the nature and incidents of the traditional concept of marriage and assumes that the term can be so confined.
A further position along the spectrum focuses on the injunction against construing the Federal and State Constitutions as requiring the recognition of same-sex marriages and for these purposes the construction to which the amendment is addressed is that undertaken by the judges and courts of the land. In other words, it is concerned with the judicial rather than legislative recognition of same-sex marriages. The concern here is to curb a perceived judicial activism and leaves the legislatures free to provide for the recognition of same-sex marriages. Again, however, the question is whether the proposed amendment can be so confined since its provisions go beyond providing for the injunction mentioned.
There was, in fact, a chance that the provisions were so worded as to go further and also ban civil unions even if they were not called marriages. This reflects the most extreme position along the spectrum, which not only guards against the ‘misuse’ of the term ‘marriage’ but wishes to ensure that the rights which flow from marriage are confined to the partners of a marriage — in other words persons of the opposite sex who are husbands and wives. From a linguistic point of view this interpretation of the proposed amendment derived some support from the reference in clause 2 to the ‘legal incidents’ of ‘marriage’ not being ‘conferred upon any union other than the union of a man and a woman’. Although this was hardly likely to figure in the debate in the United States, it is also a view which interestingly derives support from the literal possibilities opened up by the judicial interpretation accorded to the subject-matter of marriage in the power of the Commonwealth Parliament to make laws with respect to that subject. It will be recalled that the High Court has long accepted the view that the marriage power is not confined to its celebration or creation and extends to dealing with the rights and duties which arise out of a marriage.
At the time this article was written at least 26 States had passed amendments to their State Constitutions seeking to ban same-sex marriages. A number of these amendments were adopted by sweeping majorities in the November 2004 elections. The ease with which such amendments were carried issues a powerful message about the mood of the American population. The theoretical possibility exists that such amendments are themselves open to attack on the ground that they violate the Equal Protection (and possibly the Due Process) Clauses of the Fourteenth Amendment to the United States Constitution, which applies to provisions of State constitutions even when they represent the will of the people of the State. But such a challenge would be the very kind of judicial challenge in the United States Supreme Court which Professor Sunstein cautioned against.
A challenge to the validity of a State constitutional amendment seeking to ban same-sex marriage based on the Fourteenth Amendment did in fact meet with initial success. But despite its subsequent reversal on appeal, it is worth noting that the initially successful basis of the challenge was not grounded on whether the denial of the recognition of same-sex marriage or civil unions was itself a denial of the Equal Protection and Due Process Clauses of the Fourteenth Amendment. The case in question involved a challenge to the validity of the amendment of the State Constitution of Nebraska which provided that marriage was confined to unions between persons of the opposite sex and that persons of the same sex could not unite in a ‘civil union, domestic partnership, or other similar same-sex relationship’. Such a provision was initially held to deny the equal protection of the law because it focussed primarily on a particular class of persons and made them unable to compete on an equal footing in being able to lobby for changes in the law to confer benefits and protections enjoyed by other persons. The amendment interfered with the fundamental right of access to the political process by erecting a barrier that made it difficult for the members of the targeted class to obtain a benefit that was enjoyed by members of other classes of persons in the community. Unlike the latter class of persons, the targeted class could not obtain the same benefits by seeking to obtain the enactment of a law in the ordinary way by the Nebraska State legislature but had to obtain the approval of electors to amend the Nebraska State constitution. These grounds were the same grounds relied on by the United States Supreme Court in a case where that Court held invalid an amendment to the Colorado State Constitution which repealed and prohibited measures designed to protect homosexual and lesbian persons as a violation of the Equal Protection Clause in the United States Constitution.
It was held on appeal that the State amendment was rationally related to legitimate State interests and, therefore, did not violate that clause. Nor did it violate that clause by making it more difficult for a group of persons who retained full access to the political process to successfully advocate their views since, although the First Amendment guaranteed the right to advocate their views, it did not guarantee them political success.
In addition there is already some State judicial authority which has upheld, and had to interpret, the scope of an amendment to a State constitution which sought to ban same-sex marriages. Not surprisingly, the question whether such an amendment effectively adopts the extreme position on the spectrum mentioned above was raised, but left open on this occasion as not being properly raised before the court.
It is worth pausing to reflect at this point on how far courts can, and should, act in advance, or in the face, of hostile public opinions — however rational and principled the decisions of the State courts upholding the recognition of same-sex marriage may have been. To this may be added the opposite reaction in Canada — in particular, the decision of the Canadian Governments not to appeal against the decision in the Halpern case and the general, even if not enthusiastic, public acceptance of that case and the other Provincial cases that followed the decision in that case. This culminated in the enactment of Dominion legislation making it possible to enter into and have recognised same-sex marriages even if that enactment met with some opposition in the Canadian Parliament.
The courts in both countries have had to grapple with a judicial guarantee of equality and its reach in relation to homosexual relations following the removal of criminal sanctions which previously attached to those relations. Yet the public reaction in the two countries to the relevant judicial decisions have been different, even though it might have been thought that the comparatively recent acceptance of the judicial enforcement of individual guarantees of liberty in Canada might have suggested otherwise. In my view, that difference can only be satisfactorily explained by the different underlying social attitudes and cultural values that prevail in the United States and Canada.
New Zealand does not of course have a federal constitution, but is instead a unitary state. The issue there is whether same-sex marriages were recognised as a result of the protection provided against discrimination by the New Zealand Bill of Rights Act 1990 (‘Bill of Rights’). That instrument is a statutory, and not a constitutional, Bill of Rights and therefore only enjoys the status of an ordinary statute. The attempt to have same-sex marriages recognised through a parliamentary enactment of that kind provides a good illustration of the limits of such an instrument. As will be seen, the attempt did not meet with success and shows that there was no judicial adventurism, as had occurred with other aspects of the Bill of Rights.
In order to understand how the issue was resolved, it is necessary to give a brief outline of the relevant provisions of the Bill of Rights. The legislation sets out to affirm and give statutory protection to the rights and freedoms contained in the legislation which were based on the International Covenant on Civil and Political Rights (‘ICCPR’). This was done by providing that legislation should wherever possible be construed as being consistent with the rights and freedoms contained in the Bill of Rights — in other words, by creating an interpretative principle of preferred consistency with those rights and freedoms (s 6). At the same time, the legislation was not to be taken as having impliedly repealed any provision which was enacted either before or after the enactment of the Bill of Rights by reason of any inconsistency with a provision in the Bill of Rights. Nor could a court decline to apply any provision of an enactment by reason only of such inconsistency (s 4). This ensured that there was no derogation from the parliamentary supremacy accorded to the New Zealand Parliament. The rights and freedoms contained in the Bill of Rights were further qualified by being subject to ‘to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society’ (s 5). The provisions of s 19 of the Bill of Rights provided for freedom from discrimination based on various grounds. These included freedom from discrimination based on sex or sexual orientation as a result of s 21 of the Human Rights Act 1993 (NZ).
In Quilter v Attorney-General (1998)  the New Zealand Court of Appeal held by way of dicta that the failure of the New Zealand Parliament to recognise same-sex marriages in the Marriage Act 1955 (NZ) did not constitute discrimination contrary to s 19 of the Bill of Rights. The reasoning used by the judges added little to the kind of reasons given in dissenting judgments in the American State cases, although no reliance seems to have been placed by two of the three majority judges on the formal absence of discrimination argument which was mentioned earlier and rejected in a different context in Loving v Virginia. On the other hand, the case contained an extensive discussion of the international human rights dimension of the problem. This is understandable given the nature of the Bill of Rights and its aim of affirming New Zealand’s commitment to the ICCPR. A special emphasis was placed on what was perceived to be the need for greater particularity to substantiate the assertion that international law human rights instruments now required the recognition of same-sex marriages. Freedom from discrimination based on sex or sexual orientation was thought to be insufficient by itself to prohibit the non-recognition of same-sex marriage. This turned on the interpretation of the relevant articles in the ICCPR which are beyond the scope of this article. No attempt was made to invoke the qualifications to the rights and freedoms contained in the Bill of Rights created by s 5 of that Act. In the view of the writer, the discussion regarding the existence or non-existence of discrimination did not at times clearly distinguish between two separate things: first, whether the non-recognition of same-sex marriage constituted the relevant kind of discrimination; and second whether, even if it did, the New Zealand Parliament intended to depart from the traditional concept of marriage contained in the Marriage Act — legislation which pre-dated the enactment of the Bill of Rights.
All five judges in the case agreed that, as a matter of parliamentary intention, the Marriage Act 1955 (NZ) adopted the traditional view of marriage and that Parliament could not, by enacting s 19 of the Bill of Rights (as amended by the Human Rights Act 1993 (NZ)), be taken as having ‘effected such a major change to a fundamental institution … in such an indirect manner’. Thus it was said to be ‘highly unlikely that Parliament would have intended to make such a substantial change to one of society’s fundamental institutions by the indirect route of s 19 and s 6 of the Bill of Rights’. This was found to be so especially in the light of other post-1990 legislation which showed Parliament’s intention to adhere to the traditional view of marriage. It will be recalled that the Bill of Rights does not entrench the rights and freedoms contained in that instrument. The essential limitation to the rule contained in s 6 is that it is only an interpretative principle. It requires legislation to be given a meaning that is consistent with those rights and freedoms only in certain circumstances. Those circumstances are that such a meaning is to be preferred only where the relevant statutory provisions are in effect open to such a meaning. Given the gravity of the change sought to be made to the meaning of ‘marriage’, the relevant provisions of the Marriage Act were found not to be open to that meaning, particularly having regard to the traditional meaning of that term. In short, if such a change was to be made at all, it would have to be made by the Parliament of New Zealand and not the courts of that country.
It can be seen that the issue in New Zealand is shorn of any constitutional implications. The issue there was simply whether Parliament could be taken to have impliedly amended its Marriage Act to recognise same-sex marriages merely because it had in subsequent legislation provided for statutory protection against discrimination based on sex and sexual orientation — especially when the subsequent legislation made no specific reference to the recognition of same-sex marriages. At bottom, this depends on the interaction of two statutes of equal status, and thus ultimately on whether there was an intention on Parliament’s part to impliedly amend earlier legislation by the enactment of potentially inconsistent later legislation. This is, of course, a familiar problem of statutory interpretation. It does, however, illustrate how a controversial issue such as same-sex marriage can present itself to the courts even in a country which has a unitary constitution based on parliamentary supremacy. Subsequently the New Zealand Parliament enacted legislation to provide for civil unions.
The developments surveyed in this article show how certain societies have responded to changes in fundamental social values and attitudes which divide their communities — in particular, how the legislatures and the courts of those societies have coped, and are coping, with changes of that nature. The recognition or non-recognition of same-sex marriages provides an obvious illustration of this kind of situation. The result of the survey is to reveal an acute and underlying tension between two intertwining and conflicting values. Not surprisingly, the tension is expressed in different ways that reflect the different constitutional arrangements in the countries discussed. The tension is whether the recognition of same-sex-marriages should be decided by only the elected representatives of the people or by those representatives subject to the ultimate determination of unelected judges. (Even where judges are elected, as is the case with the judiciary in some States of the United States, it is arguable that the role for which they are elected will be constrained by the separation of judicial from other governmental powers.)
In the United States the potential for this issue to be decided by the courts is greatly increased by the American commitment to the equal protection of the law and other constitutional guarantees — a commitment that represents a unique experiment in the judicial protection and entrenchment of fundamental guarantees of individual rights and liberties. In that regard the United States has, of course, been joined by Canada with the adoption of a similar guarantee of equality in the latter part of the last century. This lays the ground for a clash between the courts and majority public opinion of the kind that has been particularly evident in the United States but not, it seems, Canada or at any rate at least to the same extent. The potential clash with public opinion seems to be a necessary and inevitable price to pay for the protection of the rights of unpopular minorities. In the United States the issue is also complicated by federalism to the extent that marriage is a matter for determination by the State legislatures and their courts. But, even here, the driving force has been the courts through the judicial enforcement of the guarantee of equality (and in some cases due process) contained in State Bills of Rights. In this regard both the Federal and State Constitutions of the United States can be regarded as an ‘exercise in self restraint’.
The operation of federalism in Australia and Canada is different in that the power to legislate with respect to marriage is vested in the national Parliaments of those countries. In Australia, the role of the courts is confined to determining whether that power encompasses the ability to provide for both the recognition and non-recognition of same-sex marriage. The issue is which Parliament can legislate on this matter under the federal distribution of legislative powers — not the legislative outcome that results from that distribution. It would be extremely surprising if the framers of the Commonwealth and Canadian Constitutions could have anticipated the fundamental changes in social attitudes that have given rise to the modern day possibility of recognising same-sex marriages as marriages. But the role of judicial review in a federation is to confront the effect of such a change in the quest for determining the meaning of words used to describe the subject matters of national legislative power and this attracts the familiar and associated debate about originalism and original intent.
The experience in Australia has illustrated the uses to which national legislative powers can be put where a Parliament and government are unsympathetic to the recognition of same-sex marriage. The Commonwealth Parliament used its powers to put the traditional meaning of ‘marriage’ beyond judicial doubt in its marriage legislation and perhaps also to ensure that any civil unions provided by State legislation would not be confused with marriage as a national legal institution. I have argued that the Commonwealth Parliament has the power to provide the latter, even if the subject matter of ‘marriage’ is confined to unions of the opposite sex. But if the subject matter is construed broadly and generously to accommodate same-sex marriages — as I think it should and ultimately will be — this will ironically make it easier for a national Parliament to ban not only same-sex marriages but also civil unions, even if they do bear the label of ‘marriage’. That said, the present Commonwealth Government has eschewed such an intention, except perhaps in the Territories over which it has largely unfettered legislative control.
By contrast the experience in Canada has been to show what can happen when a national government and Parliament is, if not supportive of the recognition of same-sex marriages, at least not hostile to such recognition. What has been particularly interesting has been the role of the courts in that country in acting as a catalyst for the national legislation which now recognises same-sex marriage since it is doubtful whether the legislation would otherwise have been passed, at least in current times.
The United States and Australia share, or may have cause to share, at least one problem which results from the adoption of a federal distribution of legislative powers, and that is the problem of full faith and credit, which will arise if some States legislate to recognise same-sex unions when others refuse to do so. If federalism can be seen as enabling States to be used as social and political laboratories, both countries have unresolved questions about the extent to which the results of such laboratories must be recognised in other States which are not sympathetic to such social and political experiments.
By contrast to the three countries mentioned above, New Zealand provides an obvious example of the constitutional issues which arise from the attempt to recognise same-sex marriages in the courts of countries which operate under a unitary constitution based on the legal supremacy of Parliament. Not surprisingly, there is much less potential for courts to play a meaningful role and much greater justification for such courts to affirm that the issue must be resolved by the legislature and not the courts. The doctrine of implied repeal or amendment is likely to make it difficult to come to any other conclusion.
It remains to offer some brief concluding remarks about the powerful judicial effect of the concept of equality, whatever debate may exist about its definition and operation. The fundamental nature of the concept means that it has ancient roots. It has been observed that Aristotle said that justice demanded the equal treatment of those equal before the law, but that it remains for each political order to determine whom to treat as equal or otherwise. This article has shown that in countries which entrench guarantees of individual rights and freedoms the ‘political order’ can comprise the courts even if — as has been seen in the United States — their decisions can, and sometimes do, incur the serious disapproval of the majority of electors. The Canadian experience, on the other hand, shows how such decisions can play a leading role in influencing the development of public opinion in the shaping of what constitutes equality.
The changes that have occurred in public and community attitudes towards same-sex marriages are not unlike those that have occurred in relation to the position of women. Professor Chaim Perelman has rightly observed that ‘[w]hat is considered reasonable in one society, at one period, may not be regarded as such at another period or in another society.’ He went on to state:
An example taken from the law of Belgium illustrates this. On 11th November, 1889, the Belgian Supreme Court refused a Belgian woman access to the bar although she was a doctor of laws who fulfilled all the requirements for admission set out in the law. While article 6 of the Belgian Constitution proclaims the equality of Belgians before the law, the Supreme Court nonetheless ruled that if the legislator had not explicitly excluded women from the bar, it was ‘an axiom too evident to require legal pronouncement that the administration of law was reserved for men’. What was evident and therefore reasonable in those days, can appear unreasonable and even ridiculous today.
In the author’s view this is similar, although not identical, to the position that was reached in the countries discussed in this article regarding the recognition of same-sex unions once it was agreed to lift criminal sanctions on homosexual behaviour.
This leads to an interesting question as to why judges and lawyers can sometimes appear to be out of step with public opinion on issues of equality. Although by no means inevitable, decisions like Goodridge and Lewisdiscussed earlier in this article reflect, in my view, a commitment to notions of equality which form part of a lawyer’s basic training. It is, after all, no accident that equality before the law forms a basic tenet of the rule of law. There are here parallels with the developments that occurred in relation to the cases on the recognition and efficacy of native title in Australia, even though those cases did not strictly involve a constitutional guarantee of equality.
It has been suggested that ‘[d]espite conceptual muddle over its positive content, the principle of equality has been negatively of great value in placing the onus of justification firmly on its opponents’. In my view, the arguments advanced to justify the differential treatment accorded to same-sex unions on the assumption that the treatment was prima facie unequal, were not persuasive and bear out the force of the observation just quoted, despite the judicial reaction in other States provoked by cases such as Goodridge in Massachusetts. It can hardly be surprising if at least some courts prove more rigorous than legislatures in addressing that onus.
The picture painted in this survey is incomplete. It remains to be seen whether, and when, public opinion comes to accept the recognition of same-sex marriage so as to eliminate the underlying tension between whether the issue should be decided only by the elected representatives of the people or by those representatives subject to the ultimate determination of judges. Until that time arrives, the tension will remain and continue to require resolution.
[∗] Professorial Fellow, Melbourne Law School and Adjunct Professor of Law, Adelaide University and the Australian National University.
 William Shakespeare, Hamlet, Act 3, Scene 1 (‘To be, or not to be: that is the question …’).
 Commonwealth Constitution s 51(xxi).
 Constitution Act 1867 (UK) s 91(26) which gives the Dominion Parliament exclusive power to legislate with respect to ‘Marriage and Divorce’ and compare s 92(12) which gives the legislatures of the Provinces exclusive power to legislate with respect to the ‘The Solemnization of Marriage in the Province’.
 The Australian Concise Oxford Dictionary (2nd ed, 1992; reprinted 1993) at 692.
 Webster’s Third New International Dictionary of the English Language, Unabridged (2002) at 1384.
 Quilter v Attorney-General of New Zealand  NZCA 207;  1 NZLR 523 at 549 (Thomas J).
 James Hastings (ed), Encyclopaedia of Religion and Ethics (1971) vol viii at 445 (‘Marriage (Greek’)).
 The New Encyclopaedia Britannica: Micropaedia (15th ed, 1995) vol 7 at 871 (‘marriage’).
 Marriage Act 1961 (Cth) ss 5(1) (definition of ‘marriage’), 88B(4), as amended by the Marriage Amendment Act 2004 (Cth).
 United States Constitution, Amendment 14 § 1.
 Cass R Sunstein, Designing Democracy: What Constitutions Do (2001) at 206.
 Id at 193. See generally ch 8 (‘Homosexuality and the Constitution’) of the same book at 183–208 and also for an earlier similar expression of his views: Cass R Sunstein, ‘Foreword: Leaving Things Undecided’ (1996) 110 Harvard Law Review 4 at 96–99, especially 97.
 Hawaii: Baehr v Lewin 852 P 2d 44 (1993); Vermont: Baker v State 744 A 2d 864 (1999); Massachusetts: Goodridge v Department of Public Health 798 NE 2d 941 (2003) and In re Opinions of the Justices to the Senate 802 NE 2d 565 (2004). See also Graham Gee, ‘Same-sex Marriage in Massachusetts: Judicial Interplay Between Federal and State Courts’  Public Law 252.
  USSC 4776; 539 US 558 (2003) (Stevens, O’Connor, Kennedy, Souter, Ginsburg & Breyer JJ; Rehnquist CJ, Scalia & Thomas JJ dissenting). Only O’Connor J relied on the Equal Protection Clause of the Fourteenth Amendment but she was careful to reserve her position on the same-sex marriage issue.
 William J Brennan, ‘The Constitution of the United States: Contemporary Ratification’ (1985– 1986) 27 South Texas Law Review 433 especially at 436–437 (speech delivered at Georgetown University, 12 October, 1985). See also Mello below n 18 at 42–44.
 Goodridge v Department of Public Health 798 NE 2d 941(2003) (‘Goodridge’).
 In re Opinions of the Justices to the Senate 802 NE 2d 565 (2004) (‘In re Opinions’).
 Such a law has been passed in Vermont, as to which see generally Michael Mello, Legalizing Gay Marriage (2004). There is an obvious analogy here with the ‘separate-but-equal’ concept: id at 23–4. The inadequacy of such a measure in meeting the guarantee of equality is rejected by Mello: id, especially ch 5 at 142–192. In this article a legal relationship under which same-sex partners enjoy the same rights and duties as those that are conferred or imposed on persons of the opposite sex in a marriage without being referred to as a ‘marriage’, is described as a ‘civil union’ as distinct from ‘same-sex marriage’. An alternative description of civil unions is ‘domestic partnerships’.
 Halpern (2003) 225 DLR (4th) 529. Courts in several other Provinces followed suit. After the Canadian Supreme Court upheld the ability of the Dominion Parliament to recognise same-sex marriage in an advisory opinion in Reference re Same-Sex Marriage (2004) 246 DLR (4th) 193 that Parliament subsequently passed legislation to give effect to such recognition: Civil Marriage Act 2005 (Can).
 Halpern (2003) 225 DLR (4th) 529 at 554–562 –, applying the test established in Law v Canada (Minister of Employment and Immigration)  1 SCR 497 at 535 quoted in Halpern at 555 .
  USSC 168; 388 US 1 (1967) and see Sunstein, above n11 at 198–200.
 Halpern (2003) 225 DLR (4th) 529 at 553 .
 Goodridge 798 NE 2d 941 (2003) at 961 n 23.
 Id at 961–962. Even though impotence can, in certain circumstances, be a ground for subsequently nullifying a marriage at the election of a disaffected party: at 961 n 22. The point made in the text is also underlined by the view adopted by the United States Supreme Court in Turner v Safley  USSC 100; 482 US 78 (1987) when it upheld the fundamental right of prison inmates to marry because most inmates would eventually be released and such marriages were formed in the expectation that they would ultimately be consummated. The reference to ‘most inmates’ implies a recognition that not all inmates would eventually be released to enable the consummation of their marriage. The marriages in question were treated as ‘expressions of emotional support and public commitment’: at 95 and see, generally at 94–97. To similar effect are the remarks of the House of Lords in Bellinger v Bellinger  UKHL 21;  2 AC 467 which suggest that the traditional emphasis of the institution of marriage on procreation has given way to emphasis on viewing marriage as also being based on the ‘mutual society, help and comfort’ that partners to a marriage should provide each other: at 480  per Lord Nicholls of Birkenhead.
 Goodridge 798 NE 2d 941 (2003) at 962–963.
 Id at 961, 963.
 Id at 963–964.
 Id at 967.
 Id at 965.
 Id at 964. An associated argument is that the recognition of same-sex marriages would encourage marriages of convenience to gain the benefits that attach to that institution: Baker v State 744 A 2d 864 (1999) at 885 n 14, 911.
 In Australia, see for example the effect of s 117 of the Commonwealth Constitution discussed in Street v Queensland Bar Association  HCA 53; (1989) 168 CLR 461 at 486, 502–503, 504 and 520 and compare Goryl v Greyhound Australia Pty Ltd (1994) 179 CLR 463. See also, in relation to s 109, McBain v Victoria  FCA 1009; (2000) 99 FCR 116.
 798 NE 2d 941 (2003) at 968–970.
 Halpern (2003) 225 DLR (4th) 529 at 569–572 –.
 Above [pp 000] (at p 1).
 Lockyer v City and County of San Francisco 95 P3d 459 (2004) at 464.
 Id at 485.
 Arizona: Standhardt v Superior Court, ex rel County of Maricopa 77 P3d 451 (2003) (‘Standhard’)); Florida: Frandsen v County of Brevard 800 So 2d 757 (2001) especially at 759, review denied 828 So 2d 386 (2002); Indiana: Morrison v Sadler 821 NE 2d 15 (2005) (‘Morrison’); New Jersey: Lewis v Harris 908 A 2d 196 (2006) (‘Lewis’); New York: Hernandez v Robles 855 NE 2d 1 (2006) (‘Hernandez’); Washington: Andersen v King County 138 P 3d 963 (2006) (‘Andersen’). In California the Court of Appeals has also declined to recognise same-sex marriages: In re Marriage Cases 49 Cal Reptr 3d 675 (2006). Although this ruling reversed a lower court decision on the matter, the decision of the Court of Appeals has been appealed to the State’s highest appellate court which, as at the date of writing, had yet to determine the appeal. So far as New York is concerned, and shortly before this article went to print, the Appellate Division (4th Department) of the Supreme Court of that State is reported to have ruled on 1 February 2008 in favour of recognising a same sex marriage celebrated in Canada under what was described as the State of New York’s ‘“long standing ‘marriage recognition rule’” in Martinez v County of Monroe, 2008 NY Slip Op 00909: R McFadden, ‘State Court Recognizes Gay Marriages From Elsewhere’ The New York Times 2 February 2008 - <http://www.nytimes.com/2008/02/02/nyregion/02/nyregion/02samesex.html?_r=l & sq=same sex ma.> [available to me as at 3 February 2008]. However this article is only concerned with conflict of laws issues so far as they are relevant to the constitutional aspects of the recognition of same sex marriage.
 See for example Hernandezs 855 NE 2d 1 (2006) at 10–11; and Andersen 138 P3d 963 (2006) at 989–990.
 See for example Hernandezs 855 NE 2d 1 (2006) at 7–8; and Andersen 138 P 3d 963 (2006) at 969, 982–984.
 Hernandezs 855 NE 2d 1 (2006) at 7.
 Andersen 138 P 3d 963 (2006) at 969.
 Hernandezs 855 NE 2d 1 (2006) at 7–8.
 Id at 8.
 The point was neatly put by Fairhurst J in dissent by emphasising that the issue should have been whether not recognising same-sex marriages furthered that interest: Andersen 138 P 3d 963 (2006) at 1012–1013 and compare at 969 n 2. For a clear and more sympathetic analysis of the considerations relied on to justify the non-recognition of same-sex marriages, even though it failed to persuade me, see Frank Brennan, Acting on Conscience: How can we responsibly mix law, religion and politics? (2007) at 183–214 (ch 8).
 Andersen 138 P 3d 963 (2006) at 980 and see also at 969, 983 and 984; and Hernandez 855 NE 2d 1 (2006) at 10–12.
 Id at 9–10; Andersen 138 P3d 963 (2006) at 973–980.
 Hernandez 855 NE 2d 1 (2006) at 9–10.
 Lewis 908 A 2d 196 (2006) at 208–212 and see also at 222–23.
 Id at 221–224.
 Id at 217.
 Id at 219–220.
 See for example Adams v Howerton,  USCA9 302; 673 F 2d 1036 (1982), certiorari denied 458 US 1111 (1982) — same-sex marriage partners held not to qualify as a citizen’s spouse within the meaning of the Immigration and Nationality Act (US). See also Smelt v County of Orange 374 F Supp 2d 861 (2005) where a Federal District Court in California upheld certain provisions of the Defense of Marriage Act 1996 (US) (s 3) cited and explained below in text accompanying nn 91–92 below. The provisions defined ‘marriage’ as a legal union between one man and one woman for the purposes of federal legislation.
 Compare the introduction of statutory Bills of Rights at the State and Territory levels of government: Charter of Rights and Responsibilities Act 2006 (Vic) and the Human Rights Act 2004 (ACT).
 The Commonwealth Parliament does have the power to make laws for the recognition (and the denial of recognition) of same-sex marriages and civil unions in the Territories under s 122 of the Commonwealth Constitution. It also has legislative power to define the meaning of the term ‘marriage’ in any valid legislation passed in the exercise of other powers apart from that contained in s 51(xxi). Despite the extensive grant of self-government to the two internal Australian Territories, the Governor-General in Council retains the power to disallow legislation enacted by the legislatures created for those Territories: see Australian Capital Territory (Self-Government) Act 1988 (Cth) s 35 and the Northern Territory (Self-Government) Act 1978 (Cth) s 9. The reference to the ‘Governor General’ is usually taken to mean the Governor-General acting with the advice of the Federal Executive Council by reason of s 16A of the Acts Interpretation Act 1901 (Cth). Effectively this means that the power is exercised on the advice of the Federal Government. The power was exercised in the case of the Australian Capital Territory in relation to the Civil Unions Act 2006 (ACT) which made provision for civil unions because the Federal Government believed that it was an indirect attempt to widen the definition of marriage which was seen as the exclusive preserve of the Federal Parliament: see Commonwealth of Australia Special Gazette, No S 93, 14 June 2006 and Federal Attorney-General, Philip Ruddock, ‘Government moves to protect the status of marriage’ (Press release 106/2006, 13 June 2006).
 See for example Grain Pool of WA v Commonwealth (2000) 202 CLR 479 at 491–496, 511–513 –, – (Gleeson, Gaudron, McHugh, Gummow, Hayne & Callinan JJ) and compare at 515, 518–532 , – (Kirby J) (‘Grain Pool’). In that case the High Court dealt with the meaning of the terms used in s 51(xviii) of the Constitution when it upheld the validity of the Plant Variety Rights Act 1987 and Plant Breeder’s Rights Act 1994 (Cth) which provided exclusive rights over new plant varieties. See also Eastman v The Queen  HCA 29; (2000) 203 CLR 1 at 41–51 – (McHugh J); and Singh v Commonwealth  HCA 43; (2004) 222 CLR 322 at 385–386 – (Gummow, Hayne & Heydon JJ) and compare at 411 , 413 , 417 , 418  (Kirby J) (‘Singh’). For a sophisticated and illuminating analysis of the meaning of constitutional terms which seeks to draw on theoretical perspectives, see Simon Evans, ‘The Meaning of Constitutional Terms: Essential Features, Family Resemblance and Theory-Based Approaches’  UNSWLawJl 24; (2006) 29 University of New South Wales Law Journal 207; and also the conventional treatment of the issue in Patrick H Lane, Lane’s Commentary on the Australian Constitution (2nd ed, 1997) at 909–913.
 R v Brislan; Ex parte Williams  HCA 78; (1935) 54 CLR 262 (radio): Jones v Commonwealth (No 2)  HCA 6; (1965) 112 CLR 206 (television).
 The Queen v L  HCA 48; (1991) 174 CLR 379 at 404 (Dawson J) and compare Attorney-General for NSW v Brewery Employees Union of NSW  HCA 94; (1908) 6 CLR 469 at 610 (Higgins J) who suggested that Parliament had the power under s 51(xxi) to ‘prescribe what unions are to be regarded as marriages’.
 Above text and note accompanying n 9, and see Bethell v Hilyard  UKLawRpCh 37; (1887) 38 Ch D 220 cited in John Quick & Robert R Garran, The Annotated Constitution of the Australian Constitution (1901) at 608; and see also Hyde v Hyde (1866) LR 1 P & D 130. For a modern re-affirmation of the traditional meaning of the term marriage by the House of Lords see Bellinger v Bellinger (Lord Chancellor Intervening)  UKHL 21;  2 AC 467 at 480  (Lord Nicholls of Birkenhead).
 Brennan J seemed to support this view in Fisher v Fisher  HCA 41; (1986) 161 CLR 376 at 455–456 and probably also in The Queen v L  HCA 48; (1991) 174 CLR 379 at 392. Compare the refusal of Windeyer J in Attorney-General (Vic) v Commonwealth  HCA 37; (1962) 107 CLR 529 at 576–577, to confine the meaning of ‘marriage’ in the Constitution to the definition in Hyde v Hyde (1866) LR 1 P & D 130 and, in particular, to decide whether the term might encompass polygamous marriage. Those authorities were referred to in Commonwealth Information and Research Services, Parliamentary Library Bills Digest No 155, 2003-04: Marriage Legislation Amendment Bill 2004 at 3–4 (‘Bills Digest’). See also Brennan, above n44 at 190–192 who also points to the difficulty mentioned in the accompanying text.
 See McHugh J in Re Wakim; Ex parte McNally (1999) 198 CLR 511 at 553 , and Singh  HCA 43; (2004) 222 CLR 322 at 343–344 n 73 ; and Jeffrey Goldsworthy, ‘Interpreting the Constitution in its Second Century’  MelbULawRw 27; (2000) 24 Melbourne University Law Review 677 at 699. It seems likely from his observations in Grain Pool (2000) 202 CLR 479 at 529  and Singh  HCA 43; (2004) 222 CLR 322 at 413  that Kirby J would uphold the possibility, but he did not subscribe to the orthodox principles of constitutional interpretation and, similarly, Dan Meagher, ‘“The Times are They a-changin’? Can the Commonwealth Parliament Legislate for Same Sex Marriages?’ (2003) 17 Australian Journal of Family Law 134. Meagher has adopted a modified approach to those principles.
 Halpern’s case (2003) 225 DLR (4th) 529 at 547 .
 Reference re Same Sex Marriage (2004) 246 DLR (4th) 193 especially at 203–207 –. The Court placed strong emphasis on the principle whereby the Canadian Constitution is treated as ‘a living tree which, by way of progressive interpretation, accommodates and addresses the realities of modern life’: at 204. The Court also concluded that such legislation would not trench upon Provincial legislative power: at 207–208 –.
 Id at .
 See also Brennan, above n44 at 192–193 who also takes the same view as that expressed in the accompanying text.
 See Russell v Russell (1976) 134 CLR 495 at 546 (Jacobs J); Cormick and Cormick v Salmon  HCA 79; (1984) 156 CLR 170 at 178 (Murphy J); and Quick & Garran, above n58 at 608, 610.
 (1976) 134 CLR 495 where it was held that the power was not confined to the celebration of marriage and the creation of rights and duties arising out of that relationship, but also extends to the enforcement of those rights and duties: per Stephen, Mason and Jacobs JJ; Barwick CJ and Gibbs J dissenting.
 Sub-section 46(1).
 Above n9. Despite the availability of divorce under the Family Law Act 1975 (Cth), the current statutory definition of marriage surprisingly defines the union as one ‘entered into for life’. It has been said that the reference to the similar common law definition of ‘marriage’ in the Family Law Act ‘can only be regarded as propaganda contradicted by the substantial provisions of the [Family Law] Act which, except for the creation of counselling facilities, are directed to the speedy termination of the married state’: Seidler v Schallhlhofer  FLC 91–273 at 77, 552–1 per Hutley JA quoted in Bills Digest, above n59 at 17–18 n 55.
 Attorney-General v Kevin  FamCA 94; (2003) 30 Fam LR 1. Compare the refusal of American State courts to accept this view. See for example Kantaras v Kantaras 884 So 2d 155 (Fla Dist Ct App, 2004) where the Kevin case was specifically drawn to the court’s attention and rejected: at 160–161: review denied 898 So 2d 80 (2005).
 See Kruger v Commonwealth  HCA 27; (1997) 190 CLR 1 and compare Leeth v Commonwealth (1992) 174 CLR 455.
 Constitution s 51(xxix). It was established that under that power the Parliament could legislate with respect to places, persons and things physically external to Australia: Polyukhovich v Commonwealth  HCA 32; (1991) 172 CLR 501 at 528–31, 599–604, 632, 696 and 712–14 and Horta v Commonwealth  HCA 32; (1994) 181 CLR 183 at 193–96. This view of the external affairs power was recently followed and applied, but only by a majority, in XYZ v Commonwealth  HCA 25; (2006) 227 CLR 532 (Gleeson CJ, Gummow, Hayne & Crennan JJ). The recognition of same-sex marriages celebrated overseas would also have satisfied the narrowest definition of an external affair as formulated by Gibbs CJ in Koowarta v Bjelke-Petersen  HCA 27; (1982) 153 CLR 168 which he thought encompassed ‘a relationship with other countries or with persons and things outside Australia’. In his view a law would be valid under this power if it regulated ‘transactions between Australia and other countries, or between residents of Australia and residents of other countries … whatever its subject matter’: at 201–202.
 (1866) LR 1 P & D 130.
 This is distinguishable from the provisions which purported to prohibit the use of expressions in everyday use such as ‘1788’, ‘1988’ in connection with a business, trade or occupation or in combination with ‘Melbourne’ or ‘Sydney’ and ‘Family Law Conference Melbourne 1988’. These were held invalid in Davis v Commonwealth  HCA 63; (1988) 166 CLR 79 especially at 99–100 as an exercise of the implied nationhood power. They were not legal terms used to describe the subject matters denoted by express heads of federal legislative power. Presumably the same ability to prevent confusion could apply to ‘copyright, patents … and trade marks’ and ‘bankruptcy’ in Constitution ss 51(xviii) and (xvii) respectively.
 Geoffrey Lindell, ‘State Legislative Power to Enact Same-Sex Marriage Legislation, and the Effect of the Marriage Act 1961 (Cth) as amended by the Marriage Amendment Act 2004 (Cth)’ (2006) 9 Constitutional Law and Policy Review 25 at 27–28 –, 29  and 30–31 –. Much of the analysis that follows in the text on the capacity of State Parliaments to legislate for same-sex marriage is taken from this article. A contrary view was expressed by Professor George Williams in an opinion provided to the Tasmanian Greens Party which dealt with the validity of a Same-sex Marriage Bill tabled in the Tasmanian Parliament in 2005: ‘Advice regarding the proposed Same-Sex Marriage Act’ (2006) 9 Constitutional Law and Policy Review 21.
 Commonwealth, Parliamentary Debates, House of Representatives, 24 June 2004 (Philip Ruddock) at 31460 and Senate, 12 August 2004 (Ian MacDonald) at 26504 and see also at 26555.
 Marriage Act 1961 (Cth), as amended in 2004, s 88EA.
 Houssein v Under Secretary, Department of Industrial Relations and Technology (NSW)  HCA 2; (1982) 148 CLR 88 at 94; Balog v Independent Commission Against Corruption  HCA 28; (1990) 169 CLR 625 at 632, Ainsworth v Criminal Justice Commission  HCA 10; (1992) 175 CLR 564 at 575; and see generally Dennis Pearce & Robert Geddes, Statutory Interpretation in Australia (6th ed, 2006) at 140 [4,28], 142 [4.30].
 See text in the para containing n66 above.
 Commonwealth, Parliamentary Debates House of Representatives, 24 June 2004 (Philip Ruddock) at 31463 and (Senate), 12 August 2004 (Helen Coonan) at 26570. Compare however the actions of the Federal Government in bringing about the disallowance of the legislation mentioned above at n54.
 See also 28 USC §§ 1738, 1739 and the Evidence Act 1995 (Cth) s 185.
 Above n 58 at 961.
 Id at 962, and see Martin Davies, Sam Ricketson & Geoffrey Lindell, Conflict of Laws: Commentary and Materials (1997) at 45 [2.2.9]. I indicated there that it is perhaps more likely that the above concern would now be seen as coming within the province of s 117 which has in recent times assumed growing importance in the field of Australian private international law: ibid.
 Gee, above n13 at 253–254 n 15.
 For an example of a case where an interstate same-sex union was not recognised, see Rosengarten v Downes 802 A 2d 170 (2002) certiorari granted: 806 A 2d 1066 (2002) but the appeal was later dismissed as moot. In that case, a Connecticut court refused to recognise a civil union celebrated in Vermont in an action to dissolve that union, both because it could not have qualified as a marriage given the sex of the parties and also because it was not regarded as a marriage in the State where it was celebrated but only a civil union. See also Eugene F Scoles et al, Conflict of Laws (4th ed, 2004) at 593.
 David D Siegel, Conflicts in a Nutshell (2nd ed, 1994) at 169.
 354 F Supp 2d 1298 (M D Fla, 2005) (‘Wilson’).
 Id at 1303.
 Pfieffer v Rogerson  HCA 36; (2000) 203 CLR 503 at 533 –, but this may be affected by the inability of the forum to grant curial relief in relation to the application of the law of the sister State noted at 542 , 543 .
 See the combined operation of ss 4, 9, 11 (as a ‘written law of another State or Territory’ in sub-s 11(1) (b)) of the Jurisdiction of Courts (Cross-vesting) Act 1987 (all States and Territories): Davies, Ricketson & Lindell, above n82, 66 [2.2.34]. See also Lindell, above n 74 at 33–34 .
 This would not be the first time that Australian legislative developments have obviated the need to obtain a full elucidation of the operation of s 118. For other such developments see Davies, Ricketson & Lindell, above n82, 44–45 [2.2.8].
 28 USC § 1738C. The same legislation also enacted 1 § 7 which provides that the term ‘marriage’ in federal legislation was intended to be confined to legal unions between men and women and the term ‘spouse’ was given a corresponding meaning. The Hawaiian case is cited above, n13.
 It seems that 38 States had passed legislation to take advantage of this measure by the time of the publication of the article by Gee, above n13 at 254 (Table A).
 The Constitution of the United States of America: Analysis and Interpretation (Library of Congress, 2002), 908 (as updated in 2006 Supplement): http://www.gpoaccess.gov/constitution/pdf2002/014.pdf.
 See for example Standhardt 77 P3d 451 (2003) at 459 n 13. Although a Federal District Court in California upheld the validity of that Act, the holding was confined to the provisions which defined the meaning of ‘marriage’ for the purposes of federal legislation: Smelt v County of Orange 374 F Supp 2d 861 (2005).
 354 F Supp 2d 1298 (2005) at 1303. See also Scoles, above n84, which predated this case and described as ‘persuasive’ the argument that the Full Faith and Credit Clause is irrelevant ‘because it already permits exceptions to its commands for unpalatable results’: at 594 n 10.
 One of the possible views identified in relation to the corresponding Australian clause, as to which see Davies, Ricketson & Lindell, above n82 at 47 [2.2.17 (b) and (c)].
 See Scoles, above n84 at 570–571 and see also 564–566, 570–583. Reference is made to the few States which have adopted legislation approved by the National Conference of Commissioners on Uniform Law in 1943. Amongst other things, the proposed uniform legislation recognises that the law of a person’s domicile is effective to prevent the recognition of marriages which are prohibited under that law: at 579– 580.
 354 F Supp 2d 1298 (2005) at 1303.
 For a succinct summary of the various views that could be taken of the meaning and operation of s 118, see Constitutional Commission: Final Report (1988) vol 2 at 705–706 [10.344] and also set out in Davies, Ricketson & Lindell, above n82 at 47 [2.2.17] and see generally at 44–47 [2.2.8]–[2.2.15].
 See above text accompanying n11.
 See, as regards the United States Constitution, Art V. The most common method invoked requires a proposed amendment to be approved by two-thirds of both Houses of Congress and the subsequent ratification of the proposed amendment by the legislatures of three-fourths of all the States. To date there have been 27 amendments.
 108th Congress, 1st Session, House of Representatives Resolution 56, 21 May 2003 (117 sponsors) and Senate Journal Resolution 26, 25 November 2003. See also Senate Journal Resolution 30, 22 March 2004 proposed in the 2nd Session of the same Congress. It was sometimes referred to as the ‘Musgrave Amendment’ after the name of one of its co-sponsors in the House of Representatives.
 White House, ‘President Calls for Constitutional Amendment Protecting Marriage’ (Press Statement, 24 February 2004) <http://www.whitehouse.gov/news/releases/2004/02/20040224-2.html> .
 109th Congress, 1st Session, Senate Journal Resolution 1, 24 January 2005 (‘S J Res 1 dated 24 January 2005’); House of Representatives Journal Resolution 39, 17 March 2005 (‘H J Res 39 dated 17 March 2005’).
 H J Res 22, 6 February 2007.
 S J Res 1, 24 January 2005. In the current House of Representatives version the provisions in the second sentence of clause 2 would be replaced with provisions which would deprive all courts in the United States of the ‘jurisdiction to determine whether the Federal and State constitutions require that the legal incidents of marriage be conferred upon any union other than a legal union between one man and one woman’.
 H J Res 39 dated 17 March 2005.
 Alan Cooperman, ‘Little Consensus on Marriage Amendment: Even Authors Disagree on the Meaning of Its Text’ The Washington Post (14 February 2004) at A1 and A18; David Von Drehle, ‘Legal Confusion Over Gay Marriage: Who Is, and Who Isn’t, Wed Is Subject of Great Debate’ The Washington Post (27 February 2004) at A8.
 Mike Allen & Alan Cooperman, ‘Bush Backs Amendment Banning Gay Marriage: President Says States Could Rule on Civil Unions’ The Washington Post (25 February 2004) at A1 and A14 (‘Bush said he wants to preserve marriage as a union of one man and one woman but allow States to determine whether same-sex couples should receive various benefits, a formula that apparently would allow the kind of civil unions and domestic partnership arrangements that exist in Vermont and California’).
 Carl Hulse, ‘Backers Revise Amendment on Marriage’ The New York Times (23 March 2004) <http://www.nytimes.com/2004/03/23/politics/23AMEN.html?pagewanted=print & position=> Alan Cooperman, ‘Same-Sex Marriage Proposal Retooled: States Could Enact Gay Civil Unions’ The Washington Post (23 March 2004) at A4.
 In re Opinions of the Justices to the Senate 802 N E 2d 565 (2004) at 570 n 4, 572 n 1.
 William Shakespeare, Romeo and Juliet, Act 2, Scene 2.
 Text in the paragraph containing n66 above and authority cited in that note.
 Anderson 138 P 3d 963 (2006) Appendix A at 1010–1011 which listed 19 States which had adopted such amendments: at 1012, and note that the same list should have indicated that Alabama had also adopted such an amendment: Constitution, Amendment 774 (2006). The same list also listed an additional six States as having had constitutional amendments pending for election in 2006: at 1012. They were subsequently adopted at those elections: Idaho: Constitution, Art III § 28; South Carolina: Constitution, Art XVII § 15; South Dakota: Constitution, Art XXI; Tennessee: Constitution, Art XI § 18; Virginia: Constitution, Art I § 15A; Wisconsin: Art XIII § 13.
 West Virginia State Board of Education v Barnette  USSC 130; 319 US 624 (1943) at 638; Lucas v Forty-Fourth General Assembly of Colorado  USSC 132; 377 US 713 (1964) at 736; Citizens for Equal Protection, Inc v Bruning 368 F Supp 2d 980 (2005) at 1003 (‘Bruning’).
 Bruning 368 F Supp 2d 980 (2005).
  USCA8 505; 455 F 3d 859 (8th Cir, 2006).
 368 F Supp 2d 980 (2005) at 985 n 1 (‘The plaintiffs expressly disclaim an interest in recognition of same-sex marriages, civil unions or domestic partnerships as a remedy in this case. They seek only ‘a level playing field, an equal opportunity to convince the people's elected representatives that same-sex relationships deserve legal protection’ and ‘equal access, not guaranteed success, in the political arena’.); and also 1000 n 18.
 Nebraska Constitution, Art I § 29.
 368 F Supp 2d 980 (2005) at 997–1005. The amendment was also found invalid as breaching the First Amendment and because it was a Bill of Attainder in violation of Art 1 § 9 of the United States Constitution.
 Romer v Evans  USSC 45; 517 US 620 (1996).
  USCA8 505; 455 F 3d 859 (2006) at 867–869. The Court also held that the challenged legislation did not amount to a Bill of Attainder: at 869.
  USCA8 505; 455 F 3d 859 at 870 (2006).
 Li v State 110 P 3d 91 (2005). The State in question was Oregon.
 Id at 98, 102.
 (2003) 225 DLR (4th) 529.
 The attempt by a newly elected national conservative government to repeal the legislation cited above in n19, was defeated: Tenille Bonoguoure, ‘House votes not to reopen same-sex marriage issue’ The Globe and Mail (7 December 2006) <http:www.theglobemail.com/servlet/story/RTGAM.20061207.wsamesex07/BNSt> and Gloria Galloway, ‘Same-sex marriage file closed for good, PM says’ The Globe and Mail (8 December 2006) at A1.
 The other aspects concern the existence of the right to receive damages and the availability of an action for a declaration regarding the infringement of the rights and freedoms contained in the Bill of Rights, probably contrary to what was intended: Simpson v Attorney-General [Baigent’s Case]  3 NZLR 667 and Ellen France, ‘A Bill of Rights?: The New Zealand Experiment’ in Clement Macintyre and John Williams (eds), Peace Order and Good Government: State Constitutional and Parliamentary Reform (2003) 84 at 86–87. See also the controversial case which dealt with the restricting effect of the Bill of Rights and other statutory provisions on subsequent legislation concerning the sentences to be imposed on convicted criminals: R v Pora  NZCA 403;  2 NZLR 37 (CA) and, generally, Janet McLean, ‘Legislative Invalidation, Human Rights Protection and s 4 of the New Zealand Bill of Rights Act’ (2001) 4 New Zealand Law Review 421.
  NZCA 207;  1 NZLR 523 (CA) (‘Quilter’). For a similar attempt directed at the recognition in England of a foreign same-sex marriage entered into in Canada by reference to the need for the English principles of private international law to comply with the Human Rights Act 1998 (UK), see Wilkinson v Kitzinger  EWHC 2022. The attempt failed because it was not shown that the European Convention on Human Rights, which is partially incorporated into United Kingdom law, required the recognition of such marriages.
 Per Richardson P, Gault and Keith JJ; Tipping and Thomas JJ dissenting.
  USSC 168; 388 US 1 (1967). See the discussion above text accompanying nn 20–21 and Quilter  NZCA 207;  1 NZLR 523 at 557 (Keith J) and 537–538 (Thomas J) but compare at 527 (Gault J).
 Bill of Rights, recital in Preamble para (b).
 Quilter  NZCA 207;  1 NZLR 523 at 560–563 (Keith J) with whose reasons Richardson P and Gault J agreed at 526 and 527 respectively.
 Compare the discussion of this issue in the judgment of Thomas and Tipping JJ in their separate judgments: Quilter  NZCA 207;  1 NZLR 523 at 550–554 and 576–577, respectively.
 To quote the words of Keith J: id at 555. See also at 526 (Richardson P and Gault J), 547–548 (Thomas J), 581 (Tipping J).
 Id at 581 (Tipping J).
 Civil Union Act 2004 (NZ).
 The description given in a work of fiction: Louis Auchincloss, The Partners (1974).
 Wolfgang Friedmann, Legal Theory (4th ed, 1960) at 18 and compare the contrast drawn by Aristotle between persons being allowed to rule in their own interests and ‘[t]he magistrate on the other hand [who] is the guardian of justice, and if of justice, then of equality also’: The Nichomachean Ethics (Oxford World’s Classics, paperback re-issue, 1998) at Book V Ch 6, 123.
 Justice, Law, and Argument: Essays on Moral and Legal Reasoning (1980) at 92.
 Id at 92–93.
 798 NE 2d 941 (2003). For the earlier discussion see above text accompanying nn 16–18, 20–32.
 908 A 2d 196 (2006). For the earlier discussion see above text accompanying nn 48–51.
 Mabo v Queensland (No 1) (1988) 166 CLR 186; Mabo v Queensland (No 2)  HCA 23; (1992) 175 CLR 1; Wik Peoples v Queensland (1996) 187 CLR 1.
 Except as regards the inability of State laws to prevail against federal statutory provisions which prevented discriminatory interference with native title because of s 109 of the Commonwealth Constitution.
 Alan Bullock & Stephen Trombley (eds), The New Fontana Dictionary of Modern Thought (3rd ed, 1999) at 280.