• Specific Year
    Any

Goldsworthy, Jeffrey --- "Interpreting the Constitution in Its Second Century" [2000] MelbULawRw 27; (2000) 24(3) Melbourne University Law Review 677

Interpreting the Constitution in Its Second Century

JEFFREY GOLDSWORTHY[*]

[This article defends a moderately originalist theory of constitutional interpretation. Originalism is shown to be motivated not by ‘ancestor worship’, as Justice Michael Kirby recently claimed, but by the principles of democracy, the rule of law, and federalism. Of course, it must be possible to adapt a constitution to contemporary needs and values. But in Australia, exclusive legal and moral authority to do so is vested in the electors, and their representatives, acting pursuant to s 128 of the Constitution. A radical non-originalist theory would enable the judges to evade s 128, in a haphazard fashion, and the principles of democracy, the rule of law, and federalism. It would also suffer from other serious practical flaws. Such theories are being abandoned in the United States, even by eminent erstwhile proponents. The article discusses problem cases for both originalists and non-originalists, and also compares several alternative versions of moderate originalism.]

I INTRODUCTION

Towards the end of the first century in the life of a constitution, some judges and commentators get restless, and chafe at the constraints of orthodox principles of legal interpretation. They call for greater judicial latitude to interpret the constitution creatively, unconstrained by its founders’ intentions, so that it can be adapted to contemporary needs and values more easily than by formal amendment. They describe it as a ‘living organism’, which spontaneously evolves in response to external social developments. This may be an over-generalisation from only two cases, but this, at least, was the experience in the United States, which is now apparently being replicated in Australia.

In the United States orthodox principles of statutory interpretation were unquestioningly applied to that nation’s Constitution for the first century after its adoption. According to those principles, unless it has been amended since its creation, a statute or constitution means what it originally meant, which is partly determined by the intentions and purposes of those who made it. The application of a constitution necessarily changes as society changes, but its meaning can legitimately be changed only by the process of amendment that it itself prescribes.[1]

The idea of a ‘living’ or ‘organic’ constitution, with an evolving meaning that adapts to social developments, first appeared in the United States in the late 19th century. In the early 20th century it became popular among political progressives hoping to roll back the Supreme Court’s laissez-faire constitutional jurisprudence.[2] Later still, it was deployed to defend the Warren court’s judicial activism. But it never succeeded in displacing the orthodox principles of interpretation, which came to be strongly defended under the banner of ‘originalism’. Recently, originalism has been winning the debate, with many eminent non-originalists admitting to a change of mind.[3]

Constitutional jurisprudence in Australia now seems to be developing along similar lines. Here, too, orthodox principles of statutory interpretation were applied to the Australian Constitution (‘Constitution’) for the first century after its adoption.[4] And here, too, some commentators and judges have recently begun to challenge their legitimacy, and to reject the bindingness or even the relevance of the Constitution’s original meaning. Whether their arguments will be more successful than those of their American counterparts remains to be seen. However, the debate in Australia is likely to be shortened by most lines of argument and counterargument having been thoroughly explored in the United States.

It should be admitted by all concerned that there are some good arguments on both sides. A constitution cannot be rigidly interpreted and applied according to its founders’ intentions, without reference to contemporary needs and values. In a previous article I defended a theory of ‘moderate originalism’, in which a core commitment to original, intended meaning is moderated by five ways in which constitutional law can evolve and be adapted to contemporary circumstances.[5] In a similar spirit of compromise Jeremy Kirk has proposed a theory of ‘evolutionary originalism’.[6] But not everyone acknowledges the virtues of moderation. Justice Michael Kirby, for example, ridicules originalism as ‘a quaint ritual of ancestor worship’ — a ‘primitive form of ancestor worship inappropriate to constitutional interpretation in a modern state.’[7] He advocates an extreme and radical version of non-originalism, which concedes almost no relevance at all to either the Constitution’s original meaning or its founders’ intentions. I will first discuss his theory and expose its deficiencies, and then turn to more moderate alternatives.

II JUSTICE KIRBY’S RADICAL NON-ORIGINALISM

Justice Kirby rightly emphasises the importance of the question of how the court should interpret the Constitution. It ‘arises at the very threshold of every case in which the constitutional text must be elucidated’,[8] and ‘[t]here is no task performed by a Justice of the High Court which is more important than the task of interpreting the Constitution.’ To ensure that decisions are properly justified and consistent, ‘it is vital that each Justice ... should have a theory of constitutional interpretation.’[9]

He compares three possible approaches that the Justices might take to the Constitution. The first is to ‘confine themselves to ascertaining what the words meant according to the original intentions of the founders’. This is the approach known as ‘originalism’, although there are different versions of it. The second is to ‘accept those intentions as being relevant to their task; but decline to view them as necessarily determinative’. The third is to ‘regard the constitutional document as having been set completely free in 1901 from the intentions, beliefs and wishes of those who drafted it so that it is viewed by each succeeding generation of Australians with the eyes of their own times’.[10] I will call this third position ‘radical non-originalism’.

The second approach effectively collapses into the third. If the founders’ intentions can be taken into account, but never have to be followed, then in effect the Constitution is ‘set completely free’ of them. A judge who adopts that approach is just as free to interpret the Constitution according to contemporary values, as a judge who follows the third approach. That may be why, throughout the remainder of his article, Justice Kirby ignores the second approach, and reduces the issue to a choice between the first and third ones.[11] The final heading of his article is: ‘Original Intent or Text Set Free?’[12] In other words, the choice he poses is between originalism and radical non-originalism.

Having posed the issue in these stark terms, he emphatically endorses radical non-originalism: ‘In my opinion, a consistent application of the view that the Constitution was set free from its founders in 1901 is the rule that we should apply.’[13] Both in the article and elsewhere, he dismisses the original, intended meaning of the Constitution as almost completely irrelevant. ‘Reference to 1900, if made at all, should be in the minor key and largely for historical interest.’[14] The ambit of the Constitution’s words in 1900 should not ‘be thought to control, or even significantly to influence, [their] contemporary meaning.’[15]

[I]t is inaccurate ... to govern the meaning of the Australian Constitution wholly, or even in part, by reference to what was in the minds of the Imperial legislators in 1900 or of the Australian colonists who proposed the Constitution to them at that time.[16]

I do not believe the ascertainment of that meaning in 1900 is crucial or even important.[17]

According to Justice Kirby, the meaning as well as the application of the Constitution can change without any formal amendment:

[T]he meaning and intent of those who had drafted the Constitution ... [should not] ... limit subsequent developments, whether in the understanding of legal terms, a change in the meaning of language or radically different social circumstances to which the language would apply.[18]

[T]he court’s search has become one for the contemporary meaning of constitutional words, rather than for the meaning which those words held in 1900.[19]

The words remain the same. The meaning and content of the words take colour from the circumstances in which the words must be understood and to which they must be applied.[20]

The Constitution is read by today’s Australians to meet, so far as its text allows, their contemporary governmental needs.[21]

It is worth mentioning a secondary principle of constitutional interpretation that Justice Kirby endorsed in Kartinyeri v Commonwealth:

Where the Constitution is ambiguous, this Court should adopt that meaning which conforms to the principles of universal and fundamental rights rather than an interpretation which would involve a departure from such rights.[22]

In itself, this principle of interpretation could be accepted by originalists. But its application and effect depend on what is meant by ‘ambiguous’. Originalists hold that the Constitution should be regarded as ambiguous only after all admissible evidence of its original, intended meaning has been exhausted. Since Justice Kirby dismisses such evidence as irrelevant, he must regard a bare linguistic ambiguity as sufficient to justify application of the principle. On his view, the Constitution contains many more ambiguities, and the principle of interpretation should therefore be applied much more often, than originalists would allow.

III RADICAL NON-ORIGINALISM APPLIED: A CASE STUDY

Justice Kirby’s latest comments confirm earlier suspicions that he believes ‘that constitutional interpretation should be cut wholly adrift from historical context and intended meaning, a view which would venture far into the dangerous drifting currents of non-originalism.’[23] To illustrate the radical consequences of his approach, consider the opening words of s 51 of the Constitution, which confers on the Commonwealth Parliament most of its legislative powers. Those words grant the Parliament power to make laws for the ‘peace, order, and good government’ of the Commonwealth with respect to the subject-matters listed.[24] Read literally, these words would appear to limit Parliament’s powers by authorising judges to invalidate legislation that in their opinion is inimical to the peace, order and good government of the Commonwealth. But when the Constitution was enacted in 1900, the words were understood by lawyers to have the opposite meaning. It had previously been decided by the Privy Council that they did not impose judicially enforceable limits on the legislative power they granted.[25] They were used by the Imperial Parliament to confer unlimited, sovereign power with respect to the subject-matters listed.[26]

Now, according to Justice Kirby, the original, intended meaning of the Constitution’s words should not control, or even significantly influence, their interpretation. The important question is: what interpretation best serves the governmental needs of today’s Australians? It would seem to follow that any judge who believes that the governmental needs of today’s Australians would best be served if judges could invalidate legislation inconsistent (in their opinion) with important human rights, should interpret the words ‘peace, order, and good government’ as authorising them to do so. After all, their current literal meaning either directly supports that conclusion, or is ambiguous. If it is ambiguous, then the principle of interpretation that Justice Kirby laid down in Kartinyeri (‘adopt that meaning which conforms to the principles of universal and fundamental rights’) leads to the same conclusion.[27]

It is difficult to find any reasons of principle (that is, reasons other than fear of political opprobrium) why someone who follows this approach should not reach that conclusion. It is true that Justice Kirby might give some weight to the principle of stare decisis, and therefore to earlier decisions affirming the original, intended meaning of the words. But they would be entitled to very little weight. After all, they were almost certainly based on the very theory of interpretation — originalism — he so scornfully rejects. The principle of stare decisis accords little weight to previous decisions that were fundamentally wrong.[28] Even if they were not based on originalist reasoning — if (an admittedly unlikely possibility) they were based on the earlier judges’ own opinions about the governmental needs of their day — any judge who disagrees with those opinions should still regard the decisions as wrong. And even if they were not wrong when they were decided — because Australia’s governmental needs have subsequently changed — how could they possibly outweigh the governmental needs of today, as evidenced by contemporary principles of the international law of human rights?[29]

It might be responded that the court would probably decline to change the meaning of these words, even if it endorsed Justice Kirby’s theory, because it could not be confident that such a change would meet with popular approval. After all, surely non-originalists think that judges interpreting the Constitution should be guided by the values of ‘today’s Australians’ rather than their own opinions about governmental needs. But this response assumes that, according to Justice Kirby’s theory, the Constitution means what it was originally intended to mean unless and until the court changes its meaning. Only then would it follow that the court should be reluctant to do so unless it is confident that the general public would approve the change. In other words, the response assumes that there is a presumption in favour of the original, intended meaning.

But Justice Kirby’s radical non-originalism does not include any such presumption. On his view, the words of the Constitution do not mean what they were originally intended to mean, unless and until the court changes their meaning. Rather, the original, intended meaning is irrelevant. All the court has to work with are (1) the Constitution’s words, (2) its own earlier decisions interpreting those words, and (3) its perceptions of contemporary values and governmental needs. Putting (2) aside (because, as previously noted, earlier decisions based on originalist reasoning are of little weight), whatever decision the court reaches, it must be guided by its own judgment of contemporary values or needs. It may be extremely difficult for the court to have confidence in its judgment, but that is no reason to favour one alternative rather than the other. There is no reason to favour one meaning on the ground that it was the original, intended meaning, because that fact remains irrelevant. In the case of the words ‘peace, order, and good government’, the court must judge whether the literal meaning, or the original, intended meaning, is more consistent with contemporary values or needs. That is a very difficult judgment to make, and the court could have little confidence in its ability to get it right. But that lack of confidence would not be a reason to favour the original, intended meaning over the literal meaning.

Of course, my point is not that the court would inevitably adopt this radical re-interpretation of the words ‘peace, order, and good government’, even if it were to adopt Justice Kirby’s radical non-originalism. My point is that, according to his theory, there would be no strong reasons of principle for not doing so.[30] My point is also that there obviously are such reasons — and if so, there must be something very wrong with his theory. Let us turn to consider what those reasons might be.

IV ANCESTOR WORSHIP OR DEMOCRACY WORSHIP?

Throughout his article, Justice Kirby ridicules originalism as a ‘quaint’ and ‘primitive’ form of ‘ancestor worship’.[31] With respect, this is what Americans call a ‘cheap shot’. It is glib, and sufficiently specious to be rhetorically effective, but it trivialises the debate and completely misrepresents the case in favour of originalism.

Originalism is principally motivated by two principles: democracy, and the rule of law. In the Australian context, we can add a third: federalism. Ancestor worship has nothing whatever to do with it. Originalism is motivated not by misplaced respect for people in the past, but by a proper respect for people in the present — namely, the electors of Australia and their elected representatives, who, pursuant to s 128 of the Constitution, have exclusive authority to change their own Constitution. Originalism is concerned to ensure that their authority is not usurped by a small group of unelected judges, who are authorised only to interpret the Constitution, and not to change it. It is concerned to ensure that if the Constitution is to be changed, the consent of a majority of the electors must first be directly and expressly obtained, and not taken for granted by a presumptuous elite purporting to read their minds or speak on their behalf. In this respect, originalism is also motivated by respect for the rule of law. Section 128 should not be evaded by lawyers and judges disguising substantive constitutional change as interpretation. In addition, originalism is concerned to ensure that the consent of a majority of electors in a majority of states be expressly obtained, as s 128 also requires. This federal constraint on constitutional change would be all too easily bypassed by judges presuming to divine the ‘contemporary needs and values’ of the nation as a whole. The principle of federalism tempers the principle of democracy, but not — in a federal democracy — unduly.[32]

Originalists agree with non-originalists that no constitution should — or can — be frozen in the past, permanently entrenching the intentions of a founding generation that was necessarily ignorant of future needs and values. It must be possible to change a constitution to adapt to those needs and opinions. But the crucial question is: who has the right to decide whether contemporary needs and values have so changed that a consequential constitutional change is desirable? Originalists insist that in Australia, the Constitution itself provides a clear and morally compelling answer to that question. It provides that ‘this Constitution shall not be changed except’ by the electors of Australia and their elected representatives. Before their Constitution is changed, ordinary Australians must be consulted.

Non-originalists might reply that this begs the crucial question. Section 128 says that ‘this Constitution cannot be changed’ except by a process involving a referendum, but what is ‘this Constitution’? They concede that the words of the Constitution cannot be changed except according to s 128 (although it is not clear why that is not a demeaning form of ancestor worship). As Justice Kirby acknowledges, ‘the Constitution ... is expressed in a form the text of which cannot be altered except with the authority of the electors qualified to vote.’[33] But non-originalists deny that it follows from this that the meaning of the text cannot be altered by other means.

It never ceases to surprise me that anyone can regard this argument as anything more than a rationalisation for the blatant evasion of s 128. As I argued previously:

A law necessarily means something — nothing meaningless can be a law — and its meaning is part of what it is. Therefore to change the meaning of a law is to change the law. To hold that a law no longer means what it meant when it was first enacted would be to hold that it is no longer the same law — and therefore that it has been changed, without the constitutionally prescribed method of change having been employed.[34]

Taken to the extreme, this non-originalist argument entails that ‘this Constitution’ is a meaningless text, a mere assemblage of marks on paper, which can be given any meaning at all by those entrusted with its ‘interpretation’. That would be absurd. It would be pointless to forbid the marks to be changed except by a special process, but permit them at any time to be given any meaning at all by their ‘interpreters’. The ‘interpreters’ would have unlimited power to change their ‘meaning’ at any time and in any way they should please.

Non-originalists would no doubt disown that absurd opinion. If so, they must agree that a constitution must have some meaning that cannot be changed except by the prescribed process of amendment. But in their view, that unchanging meaning has minimal content, stripped of all evidence of the founders’ intentions and purposes other than what their words mean according to current rules of written English. The Constitution can then be interpreted in any way that is consistent with a current standard English dictionary.

This view is not absurd, but it is open to a similar objection. It makes a requirement that the Constitution be changed only by a special procedure, protecting the democratic rights of its citizens, hostage to linguistic fortune. It is liable at any time to be undermined by accidental ambiguity or linguistic mutation, enabling the ‘interpreters’ to give very different meanings to particular provisions than those originally endorsed.[35] The only difference between this view and the absurd one, is that instead of the ‘interpreters’ having unlimited power to change the Constitution as they please, they only have power to change it occasionally, when permitted by the vagaries of linguistic ambiguity and mutation. As I argued previously:

[N]ew principles can be smuggled into a constitution if the current dictionary definitions of its words are sufficiently pliable, but not otherwise, and whether or not they are has nothing whatsoever to do with their underlying purpose. Such a haphazard and opportunistic approach to constitutional interpretation has little normative appeal.[36]

The example discussed in the previous section, concerning the words ‘peace, order, and good government’, illustrates all these points. The Constitution does not include anything like a Bill of Rights granting the judges broad authority to invalidate legislation that (in their opinion) violates human rights. Given s 128, one would surely expect that, before the judges could acquire such authority, the Constitution would have to be formally amended by referendum. The law itself, as well as the principles of democracy and federalism, require that such a major constitutional change first be approved by a majority of electors nationwide, and a majority of electors in a majority of states. But not so, the radical non-originalist says (or logically should say). The founders included the words ‘for the peace, order and good government of the Commonwealth’ in s 51 of the Constitution. They intended those words to grant plenary and unlimited power to Parliament, but that intention is not even significant, let alone controlling. The literal meaning of the words is (fortuitously) capable of bearing the opposite meaning. The court is therefore at liberty to interpret them as conferring on the judges precisely that broad authority to invalidate legislation on human rights grounds that the founders intended to withhold. Indeed, since the original, intended meaning of the words is irrelevant, the court is required to decide which of the alternative interpretations best serves the governmental needs of today’s Australians. It must read the words in the light of contemporary needs and values, and not in the light of their meaning in 1901. So there is no need for any referendum. The words can legitimately be held to have a new meaning — the opposite of their original, intended meaning — without the words themselves having to be changed. Section 128 is not violated because it requires a referendum only if the words are to be changed.

I previously considered the suggestion that the court should decline to change the meaning of these words, even if it endorsed Justice Kirby’s new theory, because it could not be confident that such a change would meet with popular approval. I replied that the suggestion underestimated the radical nature of his theory.[37] It should also be noted that, even if my reply is mistaken, the suggestion does not meet the objections from democracy, the rule of law, and federalism. Section 128 requires that the electors must be consulted before the Constitution is changed, and that a majority of electors both nationwide and in a majority of states must approve the change. But according to this suggestion, if the Constitution can be changed by creative ‘interpretation’, the electors should be consulted only if the judges deem it necessary to do so, because they lack confidence in their own ability to divine popular values. The dangers of a tiny, unrepresentative elite presuming to decide whether or not the electors need to be consulted before it makes a major constitutional change do not need to be spelt out. As unexpected referendum and election results frequently demonstrate, it is extremely difficult to divine electors’ views without actually asking them. It is even more difficult to divine their views about a constitutional change that is slipped quietly through without the extended public debate that accompanies a referendum campaign.

Non-originalists typically ignore the crucial difference between judicial and democratic decision-making. The following passage from one of Justice Kirby’s recent judgments is characteristic: ‘The Constitution is read by today’s Australians to meet, so far as its text allows, their contemporary governmental needs.’[38] This has an impressively democratic ring to it. But it deftly glosses over the fact that it is the judges who authoritatively interpret the Constitution, and ‘today’s Australians’ are stuck with their decisions whether they like them or not — indeed, without ever being consulted. Whenever non-originalists trot out the tired old refrain that ‘we’, ‘today’s Australians’, ‘the present generation’, etc, should not be bound by ‘the dead hand of the past’, they really mean that the judges should not be bound by it. They assume that the judges speak for ‘us’, and imply that to limit the judges’ ability to change the Constitution by pseudo-interpretation is to limit ‘our’ ability to do so democratically. The assumption is highly questionable, and the implication plainly false.[39]

The words ‘peace, order, and good government’ provide the most vivid example of the way that Justice Kirby’s radical non-originalism would reduce at least some constitutional provisions to indeterminacy, allowing them to be reshaped according to the political philosophies of the judges. But many other examples could be provided. For example, if in A-G (Vic) ex rel Black v Commonwealth[40] the original, intended meaning of the words ‘for establishing any religion’ had been dismissed as irrelevant, the case would have turned on little more than whether a majority of judges approved or disapproved of Commonwealth financial aid to religious schools. Stripped of their intended meaning, such words are almost empty husks, able to be filled with whatever content the judges think best serves ‘the governmental needs of contemporary Australians’. Another example is s 92 of the Constitution. It would surely have been preferable for the judges in the 1930s and 1940s to have been guided by evidence of its original intended meaning, rather than by its bare words, which were perfectly consistent with the laissez-faire interpretation that most of them no doubt found very congenial.

I regard the ‘peace, order and good government’ example as virtually a reductio ad absurdum of radical non-originalism. Non-originalists have two options: they can either agree, or disagree, that their theory is at least compatible with (if it does not positively invite) radical reinterpretation of those words along the lines I have suggested. If they agree that it is, they must answer the objections I have raised, based on the principles of democracy, the rule of law, and federalism. If they do not agree, they must show how non-originalism avoids that consequence without inadvertently resorting to originalist principles, and collapsing into a moderate version of originalism. For example, any suggestion that the ‘evolution’ of the meaning of constitutional words can only take place from a starting point, and within parameters, fixed by their original, intended meaning, would fall into that trap.

V WORDS SET FREE OF THEIR ORIGINAL PURPOSE

Justice Kirby’s radical non-originalism must overcome other, more general, difficulties that he fails to acknowledge.

If the Constitution’s original, intended meaning is irrelevant, there are only two alternative methods of interpreting it.[41] First, it could be interpreted to mean anything that is consistent with the current literal meaning(s) of its words. I will call this ‘contemporary literalism’. Secondly, the Constitution could be interpreted as if it expresses the intentions or purposes of some group of people other than its actual founders, such as a majority of ‘today’s Australians’. I will call this ‘contemporary intentionalism’. The first alternative gives rise to difficulties that the second one can overcome, although only by running into other difficulties of its own.

In few if any other areas of law has it been thought reasonable for judges, in seeking the meaning of a legal document, to restrict their attention to the literal meanings (either original or contemporary) of its words. ‘Literalism’ has long been condemned, for very good reasons. It is a byword for a narrow, formalistic, and obstructive approach to interpretation. The authors of legal documents, whether they are parties to a contract, or legislatures, inevitably fail to express themselves with perfect clarity and comprehensiveness. To limit the meaning of their document to the literal meanings of its words maximises indeterminacy, absurdity, and the frustration of their intentions or purposes. It maximises indeterminacy because it ignores evidence of authorial intention or purpose that can often resolve problems caused by semantic ambiguity, vagueness, incompleteness, or inconsistency. The resulting indeterminacies must then be resolved by judicial discretion, which may be guided by values and goals very different from those of the law-maker. This is why literalism enables the proverbial ‘coach-and-four’ to be driven through a tax Act, frustrating the public policy it was designed to serve. For those who like to reduce everything to politics (and they are many), literalism enables conservative judges to thwart tax laws, labour laws, or other progressive legislation enacted by a reformist legislature. By the same token, of course, it allows reformist judges to obstruct laws enacted by conservative legislatures. But judges cannot apply principles of interpretation selectively, depending on whether they approve or disapprove of the political complexion of the legislature. Principles of interpretation must be bought wholesale, not retail.

These are the reasons why literalism in statutory interpretation is almost universally condemned. But not in constitutional interpretation. Non-originalists in effect argue that the most objectionable features of literalism in the case of statutes are positively desirable in the case of constitutions. The same people who would condemn judges for ignoring a legislature’s intentions would applaud them for ignoring the intentions of the Constitution’s founders. Of course, they would argue that it is perfectly reasonable to interpret statutes differently from constitutions. Democracy requires that judges, when exercising legitimate discretion, should be guided by the beliefs and values of a majority of their fellow citizens. It follows that judges have good reasons to respect the intentions of a majority of recently elected legislators, but not those of long-deceased ‘founding fathers’.

But this argument from democracy overlooks the fact that old constitutions were once young ones. Immediately after a constitution is established by reasonably democratic processes, there is no gap between the purposes and values of its founders and contemporary ones. At that time, its provisions should be held to mean what they were intended to mean. To interpret them literally, contrary to their intended meanings, would be open to all the objections just mentioned. It would permit the judges to thwart democracy, by substituting their purposes and values for those that have just received popular endorsement. The question then becomes whether, as time passes and the original meanings of some provisions lose touch with contemporary values, they can be changed by judicial interpretation. The argument that they can runs into all the objections made in the previous section. If the Constitution itself requires that it be changed only by a special, democratic process, then its original, intended meaning can legitimately be changed only in that way. The Constitution is what it means. Judges who change its meaning change the Constitution, contrary to the democratic process that it prescribes. The non-originalist argument from democracy is therefore trumped by democracy.

In any event, literalism comes at an unacceptable cost: it logically excludes implications, other than those of strict logical entailment. As I previously argued:

If the meaning of a statute or constitution is exhausted by the literal meanings which its express words currently possess, then it cannot include implications, other than those of strict logic, because the former necessarily depend on factors in addition to literal meanings ... If judges who adopted the first, literalist version of non-originalism were to say that something was implied, they would be either lying or confused. This consequence would be a disaster for legal interpretation. As I have explained elsewhere, any attempt to read a text literally, disregarding all evidence of underlying purposes and intentions, will inevitably confront massive problems of indeterminacy and absurdity, partly because the proper understanding of any text depends on understanding unexpressed assumptions. As Felix Frankfurter remarked, the most fundamental question in legal interpretation is: ‘What is below the surface of the words and yet fairly a part of them?’ The literalist version of non-originalism is forced to reply: ‘nothing’. It follows that an absurdity generated by a literal interpretation of a provision cannot be avoided, as it can be when orthodox principles are applied, by assuming that it could not have been intended and is therefore excluded by implication. The loss of genuine implications would be a particular disaster for those who approve of the judicial discovery of implied rights and powers. To begin with, the High Court’s own power of judicial review rests on an implication![42]

This objection cannot be overcome by arguing that implications can be based on some kind of practical necessity, regardless of underlying intentions and purposes. Any argument that an implication is practically necessary for constitutional provisions to function effectively depends on what their ‘effective functioning’ consists of. And since that is not specified by the provisions themselves, it can only be determined by their purposes, which in turn depend on someone’s intentions. Strictly speaking, inanimate entities such as words do not have purposes — only the people who use them do. Therefore, when we attribute a purpose to a provision, we must really be attributing it either to the people who enacted the provision, or to the people who apply or benefit from it. In either case, we must go beyond literalism.

If originalism and literalism are both rejected, we are left with contemporary intentionalism, the theory that the Constitution expresses the intentions and purposes of some group of people other than those who actually created it, such as ‘today’s Australians’. It is ironic that people who often dismiss original intentions as fictions may be forced to rely on contemporary intentions that by their lights must be equally if not more fictional, but I put that aside. I also put aside most of the arguments I have previously made against this theory.[43] I will mention just two further objections.

One major problem for contemporary intentionalism is that it undermines the non-originalist rejoinder to the objection that according to s 128, ‘this Constitution’ cannot be changed except by referendum. That rejoinder is the argument that, although the words of the Constitution can be changed only by referendum, their meanings can be changed by other means, except insofar as they are fixed by the rules of the English language. The rejoinder can succeed only if, in s 128, ‘this Constitution’ denotes only the words of the Constitution, shorn of all content other than their standard meanings in English. If, in 1901, the meaning of the Constitution was partly determined by the founders’ intentions (which were the ‘contemporary intentions’ of that time), then that original meaning was ‘locked in’ by s 128. In short, the non-originalist rejoinder can succeed only on a literalist conception of what ‘this Constitution’ is. And that takes us back to the deficiencies of literalism. Non-originalists are trapped on the horns of a dilemma.

A second major problem for contemporary intentionalism concerns implications. I have already shown that non-originalism would give judges excessive discretion to change the meaning of the Constitution’s express words, allowing them occasionally to evade s 128. Applied to implications, contemporary intentionalism would diminish the role of s 128 even more drastically.

Let us imagine that according to a majority of the High Court, ‘today’s Australians’ now expect the Constitution to serve some new purpose, which previous generations did not intend it to serve. Let us also imagine that applying the words of the Constitution in their strict and literal sense cannot achieve that purpose. In other words, some clause or provision in addition to what the words expressly provide is practically necessary to achieve that purpose. According to contemporary intentionalism, that clause or provision does not need to be added to the Constitution by formal amendment — even though the purpose it is needed to achieve is ex hypothesi a new one. The court can legitimately hold that it is already implied by the Constitution, because it is practically necessary to enable the Constitution to fulfil one of its (contemporary) purposes. In other words, whenever it is thought that a majority of Australians expect their Constitution to serve some new purpose, whatever is practically necessary to achieve that purpose can be regarded as having been automatically added to the Constitution. The Constitution spontaneously sprouts whatever new implications are necessary to achieve what a majority of today’s Australians supposedly expect it to achieve. No referendum must be held. Today’s Australians need not be asked whether they really want the Constitution to achieve that purpose, or whether, on reflection, they really want it to include the new clause or provision that it requires in order to do so.

It is unnecessary to spell out how blatantly this would violate the principles of democracy, the rule of law, and federalism. It should be stressed that this is a logical consequence — and not a fanciful and unfair caricature — of contemporary intentionalism. If that consequence is too hard to swallow, incorrigible non-originalists must swallow contemporary literalism instead — if they can do so without choking.

VI ORIGINALISM AS A NOVEL AND ALIEN THEORY

Justice Kirby suggests that originalism is an alien theory — a ‘quaint American ritual’ — that should not be allowed ‘to travel to the Antipodes and capture the imagination of Australia’s judges’.[44] He also suggests that it is novel: ‘the Australian High Court has, for a long time, turned its back upon originalism’.[45] ‘In the beginning, the High Court was extremely cautious about the use of Australian legal history, at least in constitutional interpretation’.[46] According to his account, it was not until Cole v Whitfield,[47] ‘[n]early 90 years later, [that] legal history came to the rescue of constitutional interpretation ... The pretence that constitutional interpretation required nothing but a close and prolonged study of the text of the Constitution was abandoned.’[48]

Apart from one reference in a footnote, Justice Kirby ignores the extensive evidence I compiled in an earlier article showing that although the High Court traditionally refused to consult the Convention Debates, it was always willing to take into account other historical information, concerning the legal and social circumstances at the time the Constitution was enacted, as evidence of the founders’ intentions.[49] My argument is corroborated by others who have also investigated the matter. Henry Burmester, for example, refers to

one of the basic principles of constitutional interpretation in Australia, namely that the interpretation of the Constitution can take account of historical understandings, and should be made in the light of circumstances existing at the time when the Constitution was made.[50]

Jeremy Kirk remarks that:

Australian literalist orthodoxy falls within the realm of originalism ... [which] indicates that constitutional words are to be given their full, natural or literal meaning as understood in their textual and historical context ... Provisions are to be understood according to their essential meaning at the time they were enacted in 1900.[51]

In adopting this approach to constitutional interpretation, the High Court applied well-established British principles of statutory interpretation. Even during their most literalist phase, in the late 19th century, British courts held that they were bound to seek

the intent of the legislature, to be collected from the cause and necessity of the Act being made, from a comparison of its several parts, and from foreign [extraneous] circumstances, so far as they can justly be considered to throw light upon the subject.[52]

As Kirk correctly observes, ‘the most basic principle of statutory interpretation is the adoption of an originalist approach’,[53] and the Court in Cole v Whitfield ‘essentially reiterates (indeed, almost quotes) an interpretational principle established since at least 1852, as restated in the constitutional context by Griffith CJ in 1904.’[54] Justice Kirby’s discussion therefore conveys a quite misleading impression of the principles and methodology of legal interpretation traditionally accepted by courts throughout the British Empire, including the High Court of Australia. The idea that originalism is a novel theory of American origin is quite wrong. Indeed, the reverse is true: it is non-originalism that is a relatively new American invention.

To bolster the thesis that originalism is not part of our constitutional tradition, Justice Kirby repeats the claim that Andrew Inglis Clark rejected originalism in his book Studies in Australian Constitutional Law.[55] That book is ‘particularly important because of Clark’s leading part in the committee which prepared the original draft of the Constitution.’ As ‘one of the most influential writers on Australian constitutional law’, his theory of constitutional interpretation ‘has had an influence from the beginning.’[56] Justice Kirby quotes Clark’s statement that the language of the Constitution

must be read and construed, not as containing a declaration of the will and intentions of men long since dead ... but as declaring the will and intentions of the present inheritors and possessors of sovereign power, who maintain the Constitution and have the power to alter it.[57]

But as I previously pointed out, in the same article elsewhere footnoted by Justice Kirby, this illustrates the danger of quoting passages out of context. Clark immediately went on to say:

But so long as the present possessors of sovereignty convey their commands in the language of their predecessors, that language must be interpreted by the judiciary consistently with a proper use of it as an intelligible vehicle of the conceptions and intentions of the human mind, and consistently with the historical associations from which particular words and phrases derive the whole of their meaning in juxtaposition with their context. If the present possessors of sovereignty discover that the result so produced is contrary in particular cases to their will in regard to future cases of a like character, they will amend the language which they previously retained as the expression of their will.[58]

As Kirk puts it, a ‘careful reading of Clark suggests that he was essentially an originalist.’[59] Just before the passages just quoted, Clark said that ‘[i]t has been repeatedly stated that the fundamental rule for the interpretation of a written law is to follow the intention of the makers of it as they have disclosed it in the language in which they have declared the law.’ When the circumstances of the case before the court ‘could not have been in the contemplation of the makers of the law’, it is necessary to examine the language of the law ‘for the purpose of ascertaining whether it is such as we may reasonably believe the makers of the law would have regarded as sufficient to embrace the particular act or set of circumstances in question if it had been foreseen by them.’ If the answer is that they would have so regarded it, ‘then the language of the law is construed to embrace’ that act or set of circumstances.[60]

Justice Kirby frequently cites in his support the well-known passage in which Windeyer J said that in the Engineers’ Case[61] ‘the Constitution was read in a new light’ that reflected the growth of Australian nationhood over the previous 20 years.[62] But whatever he meant in that passage, Windeyer J clearly believed that it was perfectly consistent with orthodox, originalist principles of interpretation. For example, in Bonser v La Macchia he said:

Doubtless the words of the Constitution should be given the meaning that they had when it was enacted as a statute of the Imperial Parliament. But that does not mean that its words can now denote only what they then denoted, or that we are to ignore the fact that they appear in an instrument of government for a new nation then called into being.[63]

Elsewhere, Windeyer J said that ‘[w]e are not to give words a meaning different from any meaning which they could have borne in 1900. Law is to be accommodated to changing facts. It is not to be changed as language changes.’[64]

Justice Kirby cites two decisions to support his claim that the High Court has eschewed originalism. He fails to mention the ease with which originalist explanations of those decisions can be supplied. Originalism does not entail a static and rigid application of the Constitution that is oblivious to changing circumstances. As I have previously explained, it is compatible with a variety of legitimate interpretive techniques that add a great deal of flexibility to our constitutional law, without going to the extreme of permitting the Court to change the essential meaning of the Constitution itself.[65]

For example, Justice Kirby mentions the Court’s decision in Sue v Hill,[66] that the United Kingdom is now a ‘foreign power’ for the purposes of s 44(i) of the Constitution, although it was not in 1901. He claims that this decision is ‘a particularly clear instance of the way in which Australian constitutional jurisprudence has freed itself from the doctrine of original intent’.[67] But as he must know, an originalist can easily justify that decision on the orthodox ground that the Court recognised that the denotation, but not the connotation, of the words ‘foreign power’ has changed as a result of changed circumstances. In non-technical language, the meaning of the words ‘foreign power’ has not changed, but the facts to which they must be applied have. The relationship between Australia and the United Kingdom has changed, and consequently the United Kingdom has become a foreign power. There is nothing at all puzzling about this. As I previously explained:

The denotations of terms are constantly changing: for example, every time an old lighthouse is demolished, or a new one is built, the denotation of the term ‘lighthouse’ changes. The Constitution was obviously not intended to give the Commonwealth power only over those lighthouses which existed when it came into force on January 1, 1901.[68]

Nor was it intended to refer only to ‘foreign powers’ that existed in 1901. As Gaudron J put it,

‘foreign power’ is an abstract concept apt to describe different nation States at different times according to their circumstances. For example, Papua Nuigini would not properly have been described as a ‘foreign power’ prior to the grant of independence, although it now is ... To acknowledge that, in some constitutional provisions, some words and phrases are capable of applying to different persons or things at different times is not to change the meaning of those provisions.[69]

For the same reason, if New Zealand joined the Commonwealth of Australia, it would cease to be a ‘foreign power’. It is irrelevant (even if it is true) that the idea of Britain ever becoming a foreign power ‘would certainly not have been in the minds and subjective intentions of the drafters of the clause and the delegates to the Constitutional Conventions.’[70] The relevant question is what they meant by the words they used. That meaning must be applied in today’s much changed world, whether or not they anticipated every change.

Justice Kirby acknowledges in a footnote that the majority in Sue v Hill ‘explained their opinion in a way consistent with the distinction’ between connotation and denotation.[71] He later comments that ‘[r]esort to formulae such as ... “connotation and denotation” may sometimes disguise rather than clarify the real reasons why one choice is preferred in a particular case and another is rejected’.[72] This is no doubt true.[73] But even if it is sometimes true, it is not true of the decision in Sue v Hill, which is straightforward.

Justice Kirby’s second example is the decision in R v Brislan; Ex parte Williams.[74] He says that ‘[n]o clearer illustration ... could be found’ of the ‘fact’ ‘that the High Court has, for a long time, turned its back upon originalism.’[75] But here again, the decision is perfectly consistent with originalism. The question was whether radio broadcasting came within the power conferred on the Commonwealth Parliament by s 51(v) of the Constitution to make laws with respect to ‘postal, telegraphic, telephonic, and other like services’. The Court found that it did, even though — as Justice Kirby points out — radio broadcasting did not exist and was therefore unknown to the founders in 1900.[76] The founders included the words ‘and other like services’ in order to deal with future technological developments of that kind. Whether one agrees with the majority’s decision, or with Dixon J’s dissent, the Court cannot be accused of consciously straying beyond the original, intended meaning of those words in 1900. Justice Kirby must look elsewhere if he is to find an example of a non-originalist decision. But if, as he claims, Brislan’s case is the clearest example, his chances are bleak.

VII THE RETREAT OF NON-ORIGINALISM IN THE UNITED STATES

Given its fatal flaws, it is not surprising that in the United States, radical non-originalism is now being repudiated even by eminent erstwhile proponents. They are now either publicly announcing their conversion to moderate originalism, or quietly adopting positions indistinguishable from it.

For example, in an article titled ‘An Originalism for Nonoriginalists’, Randy Barnett proposes a moderate version of originalism ‘that would be satisfactory to many who consider themselves, as I did, nonoriginalists.’[77] He admits that ‘[o]riginalism has not only survived the debate of the eighties, but it has virtually triumphed over its rivals. Originalism is now the prevailing approach to constitutional interpretation.’[78] One reason for this is that the most influential early critiques of originalism, those of Paul Brest and H Jefferson Powell, left ‘considerable room for [moderate] originalism to survive and flourish.’[79] A second reason is that ‘[i]t takes a theory to beat a theory and, after a decade of trying, the opponents of originalism have never congealed around an appealing and practical alternative.’[80] A third reason is that more careful historical work demonstrated that originalism did not necessarily lead to conservative conclusions, which ‘helped make originalism safe for the politically progressive academic world.’[81] Among other ‘political progressives who appear to have taken the originalist turn, or have at least incorporated a fully-functioning originalism into their approaches’,[82] he lists Thomas Grey, Ronald Dworkin, Bruce Ackerman, Akhil Amar, Michael Perry, Martin Flaherty, Lawrence Lessig, and William Treanor.[83]

Of Michael Perry, one commentator has said:

In the relatively short span of six years, Perry has gone from one of nonoriginalism’s most prominent, eloquent, and ardent defenders to one of its most vigorous critics ... [H]e now argues that originalism is the most defensible approach to constitutional interpretation.[84]

In his most recent book Perry sets out to show that ‘it is a serious mistake to think that the originalist approach to constitutional interpretation is necessarily conservative.’[85] He mounts a lengthy and persuasive defence of a moderate version of originalism, and then argues that the modern Supreme Court’s supposedly most ‘activist’ decisions can be fully justified on originalist grounds.[86]

Ronald Dworkin is an even more telling example.[87] This scourge of conservative originalists such as Robert Bork now contemptuously rejects the thesis that, in his words, constitutional provisions ‘are chameleons which change their meaning to conform to the needs and spirit of new times.’ He says that this thesis is ‘hardly even intelligible, and I know of no prominent contemporary judge or scholar who holds anything like it.’[88] He claims that those American judges who described their Constitution as something ‘living’ were merely using a metaphor to describe the result of interpreting certain constitutional rights not as highly specific rules, but as abstract principles whose application requires today’s judges to make moral judgments.[89] But that depends on the founders having originally intended to enact such abstract principles: Dworkin insists that ‘the Constitution means what the framers intended to say.’[90]

[I]f we read the abstract clauses ... to say what their authors intended them to say ... then judges must treat these clauses as enacting abstract moral principles and must therefore exercise moral judgment in deciding what they really require. That does not mean ignoring ... historical integrity or morphing the Constitution.[91]

As he put the same point many years ago, when enforcing such abstract principles ‘judges must make substantive decisions of political morality not in place of judgments made by the “Framers” but rather in service of those judgments.’[92]

Many other commentators confirm this recent shift of opinion in the American legal academy. James Fleming refers to an ‘emerging strain of broad originalism in liberal and progressive constitutional theory.’[93] Jack Rakove, an eminent historian, notes a ‘turn to originalism ... so general that citation is almost beside the point.’[94] Paul Horwitz says that ‘[i]t has become a commonplace assertion that, to some degree, most or all constitutional lawyers are originalists now.’[95] According to Michael Dorf: ‘Most, if not all, of us are what Paul Brest has called “moderate” originalists; we are interested “in the framers’ intent on a relatively abstract level of generality”.’[96] In a recent book review Cass Sunstein comments:

Are we all originalists now? In the commentaries in this book, Ronald Dworkin and Lawrence Tribe write as originalists, and with suitable qualifications, most of their work can fly comfortably under the originalist banner. For most participants in the continuing debates, the question is emphatically not whether the original understanding is controlling; it is how the original understanding is best understood.[97]

The timing of Justice Kirby’s advocacy of radical non-originalism is unfortunate, given these developments in its country of origin. He is jumping on the bandwagon just as everyone else is jumping off.

VIII PROBLEM CASES FOR MODERATE ORIGINALISM

I have used the words ‘peace, order, and good government’ to support my argument that radical non-originalism has unacceptable consequences. But non-originalists have their own stock of examples of real or hypothetical problems that are supposed to embarrass originalists. The main attraction of non-originalism is that it seems more likely than originalism to deliver just outcomes, as measured by contemporary values. That is because it allows judges to be directly guided by those values, rather than by the founders’ intentions and purposes, which in some cases are now (rightly) condemned.

It would be unfair and evasive not to discuss the non-originalists’ counter-examples. But before doing so, it should be pointed out that interpreting the Constitution in the light of contemporary values and beliefs, rather than those of the founders, does not guarantee morally superior outcomes.[98] For example, in Cheatle v The Queen[99] a unanimous High Court clearly believed that the founders’ conception of the institution of trial by jury was preferable to one contemporary conception. Section 81 of the Constitution requires that the trial of anyone charged on indictment with an offence against Commonwealth law ‘shall be by jury’. The court had to decide whether juries in criminal cases should be able to convict defendants by a less than unanimous vote. A South Australian law permitted this, and the question was whether it could be applied to trials in State courts of persons charged with Commonwealth offences. The very existence of the law suggested that this new procedure reflected the contemporary conception of trial by jury, and contemporary values, in at least that State. But the court held that in this respect, ‘trial ... by jury’ continued to mean what it originally meant, and therefore required unanimity. The judges clearly believed that the contemporary conception and values expressed in the State law were inferior to those of the founders, in terms of the protection of human rights.

A The Right of Women to Vote

One example often used to demonstrate that the meaning of the Constitution should be held to change in response to changing community values concerns the right of women to vote. In 1900 the words in ss 7 and 24 requiring members of the Commonwealth Parliament to be ‘directly chosen by the people’ were understood not to guarantee that women would be able to vote. The right of women to vote was deliberately left to the Commonwealth Parliament to determine. If the meaning of those words has not changed since then, Parliament has constitutional power to repeal that right. But according to Justice Kirby, it is ‘unconvincing to suggest that [these words] would today, or ever again, be construed to exclude adult women from the suffrage.’[100]

I previously argued that an attempt to repeal the right of women to vote ‘is not even a remote possibility in Australia today’, that ‘[i]t would be foolish to abandon well-established principles of interpretation in order to enable the judges to meet a non-existent danger’, and that ‘[i]f it were genuinely feared that the right ... might one day be threatened, a formal constitutional amendment would be a far better pre-emptive strategy — much more reliable, as well as more legitimate.’[101] The point about reliability can be taken further. The right of women to vote could come to be threatened by repeal only if in some far distant future there were a massive change in community values. This is extremely unlikely. But if it were to happen, future non-originalist judges, applying Justice Kirby’s theory and interpreting the Constitution according to the values of their own generation, would probably be bound to uphold the validity of the law! So on this question, radical non-originalism would probably have exactly the same embarrassing consequence as originalism.

Non-originalists believe that it is folly for today’s judges to be bound by the values of the past. Yet they seem to expect future judges to be bound by the values of today. That is inconsistent with their own theory. As for legislative repeal of the right of women to vote, either it is so unlikely that we do not need to worry about it, or it is a real possibility, but only in a future context of community values so different from ours that non-originalists would probably be required by their own theory to uphold its validity. In either case, it is not a telling counter-example to originalism. It is a scarecrow that should no longer be permitted to haunt the debate.

B Same-Sex Marriages

Another question that raises concerns about the consequences of originalism is whether or not the Commonwealth Parliament has power to legislate for same-sex marriages.[102] The Parliament has power to make laws with respect to ‘marriage’. But in 1900 the word ‘marriage’ meant a union of a man and a woman — and this would almost certainly have been regarded as an essential part of the connotation, and not merely the denotation, of the word. If so, it might seem that an originalist must interpret the word today as excluding Commonwealth power to legislate for same-sex marriages. And that might seem to be an undesirable conclusion.

However, it is possible to make a respectable argument consistent with originalism that leads to the opposite conclusion. It requires adopting what I previously called a non-literal, purposive approach to interpretation.[103] If, because of developments unanticipated by the founders, the words of the Constitution fail to give effect to their intended purpose, and the words can be expanded or contracted in a simple and obvious way in order to remedy the failure, then a court might be justified in doing so. For example, the Constitution of the United States of America (‘US Constitution’) grants power to Congress to raise ‘Armies’ and ‘a Navy’, and to regulate ‘the land and naval Forces’,[104] but does not mention an air force, which was unknown when the US Constitution was created. Despite the (arguable) failure of the words to extend to an air force, the underlying purpose is surely quite clear: to give Congress power to raise and regulate all the military forces of the United States. And the only way to give effect to that clear purpose is to interpret the words non-literally, according to their purpose rather than their literal meaning. As I argued previously, this is not in itself inconsistent with originalism. Originalism holds that the Constitution means what it was originally intended to mean. But that meaning may not be the same as its original literal meaning. Any statute can be interpreted flexibly, departing from the literal meaning of its words in order to avoid absurdity and give effect to its obviously intended purposes. There is no reason to treat the Constitution any differently.[105]

The purpose of granting power to the Commonwealth Parliament to legislate with respect to ‘marriage’ was not to forbid or prevent same-sex marriages. After all, it could not have that effect on any interpretation. Even if the word ‘marriage’ in s 51 were confined to heterosexual marriage, same-sex marriages could be authorised by State legislation. Indeed, the Commonwealth Parliament would be unable to override or modify such legislation, since ex hypothesi its power would not extend to the subject-matter. The word ‘marriage’ would then have two different meanings in Australian law: its meaning in the Constitution, confined to heterosexual marriage, and a broader meaning defined partly by State legislation.

The purpose of granting power to the Commonwealth Parliament to legislate with respect to marriage was to make possible uniform national regulation of a vitally important legal relationship that underpins family life, child rearing, and therefore social welfare throughout the nation. If current trends towards the recognition of same-sex marriages prove irresistible, confining ‘marriage’ to heterosexual marriage would result in a new legal relationship of a very similar kind having to be governed by disparate State laws rather than uniform national laws. Logically, the meaning of ‘divorce’ in s 51 would be governed by that of ‘marriage’, so that the dissolution of same-sex marriages would also have to be governed by State rather than Commonwealth law. And because it would be possible for same-sex couples to have children, by adoption or artificial fertilisation, their interests would also have to be governed by State rather than Commonwealth law. Family law would be even more fragmented than it is now.

The founders did not anticipate the possibility of same-sex marriage. In that respect, the example raises questions similar to those posed by the invention of air forces long after the creation of the US Constitution. In both cases, the important question is surely not whether the unanticipated phenomenon comes within the precise literal meaning of the word chosen by the founders to give effect to their purpose in allocating legislative power to the national legislature. It is whether the phenomenon comes within that purpose, and is so closely related to the word’s original meaning that it can be included by a simple and obvious expansion of that meaning consistent with contemporary conceptions. In both cases, there is a powerful argument that it does.

This purposive argument is admittedly dangerously slippery. Stricter originalists than I might object that it is the thin edge of such a broad wedge that it threatens to collapse originalism into non-originalism. As I previously argued:

But how far can this kind of purposive interpretation be taken? ... According to Lessig, the courts ought to interpret the American Constitution so that the founders’ purposes can be faithfully put into effect in today’s much changed world, even if that requires radical departures from the words of the Constitution, involving in effect substantial judicial redrafting.[106] Courts in the common law tradition have always been averse to engaging in such a process, regarding any necessary up-dating of statutes, for example, as the responsibility of the legislature. There is a difficult problem here. No-one would regard it as sensible for the courts to refuse to permit Congress to raise an air force until such time as the Constitution should have been formally amended. But if courts can engage in redrafting to that extent, how and where is the line to be drawn? It is clearly a question of degree. I would expect Australian judges to take a ‘flexible’ approach only when the words of the Constitution can be expanded or contracted in a simple and obvious way, and according to a very clear underlying purpose.[107]

On this approach, the question is: on what side of the line does extending the marriage power to same-sex marriages fall?

C The Race Power

A troubling example for originalism is the race power, s 51(xxvi), which grants to the Commonwealth Parliament power to make laws with respect to ‘the people of any race for whom it is deemed necessary to make special laws’. It is uncontroversial that in 1900 the word ‘for’ meant something like ‘with reference to’ rather than ‘for the benefit of’, because one of the main purposes of the power was to enable legislation to be enacted discriminating adversely against members of non-European races. The power expressly excluded ‘the aboriginal race of any state’ until 1967, when those words of exclusion were deleted by constitutional amendment. But the words defining the power were not otherwise changed, and therefore, although the purpose of the amendment was to enable the Commonwealth Parliament to legislate for the benefit of Aboriginal people, the power may support legislation discriminating adversely against them. That conclusion would be regrettable, because the inclusion in a constitution of a provision expressly authorising the enactment of laws discriminating against the members of a particular race is undeniably incompatible with contemporary values.

It is possible to argue that the power with respect to Aboriginal people is now confined to the enactment of laws for their benefit, on a number of different grounds, including: (1) that this was the purpose of the amendment in 1967; (2) that, for the same reason, the power as a whole was changed in 1967, so that any law with respect to any particular race must be for their benefit; and (3) that, irrespective of the 1967 referendum, the power should be interpreted in the light of contemporary values, rather than the purposes and intentions of the founders in 1900.[108]

It is important to note that the first two arguments, based on the purpose of the 1967 referendum, are originalist ones. Originalism holds that the meaning of a law is determined partly by the purposes or intentions of its makers. Where a constitutional amendment is concerned, the law-makers are the Parliament that passed, and the majority of electors who approved, the amendment. Only the third argument is non-originalist. In Kartinyeri[109] Justice Kirby relied on both the second and the third arguments, without clearly distinguishing between them. Indeed, at several points he said that the meaning of a constitutional amendment is determined partly by the reasons for which it was introduced. He said:

In such a case the Parliamentary Debates, and the referendum materials, may be used in the same way as the Court now uses the Convention Debates. This is to understand the cause which occasioned the amendment of the Constitution and to help resolve ambiguities in the resulting text. The search is not for the private intentions of the Members of Parliament ... [nor] the undiscoverable subjective intentions of the electors ... It is to help to derive the meaning of the Constitution, where amended, on the basis of a thorough understanding of the reasons for the amendment.[110]

Later he added that the involvement of the electors in the law-making process ‘requires that [the] Court, to understand the amendment, should appreciate, and give weight to, the purpose of the change.’[111] He concluded that in 1967 ‘[i]t was the will of the Australian Parliament and people that the race power should be significantly altered’, and that in giving meaning to the amended paragraph the court should not deny its ‘clear and unanimous object’.[112] These passages are entirely consistent with originalism, but not the radical non-originalism he subsequently advocated. Moreover, exactly the same reasoning should apply to unamended constitutional provisions: their meaning should be partly determined by the reasons for which they were created in 1900.

Of course, the main point is not that Justice Kirby relied on reasoning inconsistent with his own non-originalist theory. It is that in the case of the race power, as in that of same-sex marriages, originalism does not necessarily have the consequences that are often feared.

But it must be admitted that an originalist argument based on the purpose of the 1967 amendment may not, in the end, be persuasive. It is very difficult to see how the benevolent purpose of extending the power to Aboriginal people could have transformed the nature of the power in general, with respect to other races as well. If it did not, then the power would remain partly inconsistent with contemporary values. Even with respect to Aboriginal people, it is necessary to distinguish between the purpose of extending the power to them, and the nature of the power extended. It was undoubtedly intended and expected that the power would be used for their benefit. The question is whether it was also intended that the nature of the power be changed to prevent it being used for any other purpose.[113] The way in which the words of the power were altered indicates that it was not. Gaudron J reasoned that ‘the bare deletion of an exception or limitation on power is not, in my view, capable of effecting a curtailment of power. On the contrary, the consequence of an amendment of that kind is to augment power.’[114] An originalist may have to conclude that the legal effect of the 1967 referendum was to extend to Aboriginal people a power either to benefit or to disadvantage members of a particular race, even though the purpose of doing so was to enable laws to be enacted for their benefit.

If that is so, the question is whether a non-originalist interpretation should be accepted because it leads to a different conclusion. The most radical non-originalist argument would be that the word ‘for’, which originally meant ‘with reference to’, now means ‘for the benefit of’.[115] Its meaning has spontaneously changed along with changes in community values. The 1967 referendum is evidence that they have changed, but is not crucial to the argument.

To put the benefit of accepting this argument into perspective, it should be noted that, even if successful, it would not eliminate the danger of Commonwealth legislation discriminating adversely against members of the Aboriginal or any other race. In Kartinyeri Justice Kirby made much of that danger. He said that it was ‘totally alien to the character and meaning of our Constitution’ that ‘the Australian Parliament could ... enact racist laws such as those made in Germany during the Third Reich and in South Africa during apartheid’.[116] This is an odd statement to make about a constitution partly based on the British doctrine of parliamentary sovereignty.[117] That doctrine has traditionally been considered to be justified partly on the ground that Parliament may possess power to enact evil laws, but can be trusted not to do so.[118] Even if that idea is now regarded as naïve — indeed, even if it is naïve — that is not a good reason to think that the ‘character and meaning of our Constitution’ differs in this respect from its British parent. In any event, Justice Kirby’s arguments and conclusion in Kartinyeri would not prevent Parliament from enacting laws that discriminate adversely against the members of particular races. They would prevent it from using s 51(xxvi) to enact such laws, but not from using its other powers. There is no other constitutional obstacle to their enactment, in the absence of an implied right to equality, whose existence was rightly rejected by a majority of the court in Kruger v Commonwealth.[119] The only other way the court could prevent the Commonwealth from using most of its powers to enact racially discriminatory laws would be to adopt the radical reinterpretation of the words ‘peace, order, and good government’ discussed earlier in this article. But even that would not apply to the Territories power (s 122), which is not prefaced by those words. That plenary power enables the Commonwealth Parliament to enact laws operating within the Territories that discriminate on racial or any other grounds.

So the benefit of accepting this non-originalist argument is limited. And it must be weighed against the cost of doing so. The cost is that if the argument were accepted in the case of s 51(xxvi), it would logically be available in the case of every other provision in the Constitution. And that would place in the judges’ hands power to change the Constitution according to their own understandings of contemporary needs and values, without the approval of the electorate. It would make possible the radical reinterpretation of the words ‘peace, order, and good government’ that I discussed earlier. If the word ‘for’, in s 51(xxvi), can now be held to mean something quite different from its original meaning, simply because the original meaning is no longer consistent with community values, then in principle, so can the words ‘peace, order, and good government’ or any other words in the Constitution. My assessment is that the risks of embracing radical non-originalism — in terms of potential damage to democracy, the rule of law, and federalism — outweigh its limited benefit in the case of s 51(xxvi).

If the originalist argument based on the 1967 referendum is not persuasive, then s 51(xxvi) includes a power whose very existence is incompatible with contemporary values. But that is not surprising. An old constitution cannot be expected to be a perfect one, containing nothing that it should not contain and everything that it should. The point of including s 128 in the Constitution was to enable subsequent generations to change any aspects of it that might seem to them undesirable. What has happened to the spirit of democracy that we should hang back and expect judges to do our work for us?

IX ALTERNATIVE VERSIONS OF MODERATE ORIGINALISM

I have argued that radical non-originalism — the theory that the Constitution’s original, intended meaning is largely irrelevant to its interpretation today — is unacceptable. Even if that meaning can sometimes evolve to keep up with contemporary circumstances, the process of evolution must start from, and continue to be guided by, its original, intended meaning. But there are different ways of cashing out this vague proposition, and therefore different versions of moderate originalism.

A Concepts and Conceptions

In two recent judgments[120] McHugh J has expounded an approach to interpretation according to which the court’s principal duty ‘is to give effect to the intention of the makers of the Constitution as evinced by the terms in which they expressed that intention.’[121] On the surface, this appears to be a version of moderate originalism. Justice Kirby certainly believes that it is. But McHugh J’s statements and quotations from earlier judgments are not easy to reconcile with one another, and a close reading is required to determine whether his approach really is more originalist than non-originalist.

At several points McHugh J makes statements to the effect that ‘the meaning attributed to a particular provision now may not be the same as the meaning understood by the makers of the Constitution or their 1901 audience’.[122] But he also repeatedly insists that constitutional provisions must be interpreted to give effect to the intentions of those who made them.[123] He reconciles these propositions partly by arguing that the founders intended the Constitution to be capable of having different meanings for different generations.[124] Therefore, to attribute to a provision a meaning different from its originally intended one may (paradoxically) be faithful to the founders’ intentions. Seen in this light, McHugh J provides an originalist justification for a theory that is operationally non-originalist.

But his position is more complex than that, because he uses the word ‘meaning’ in different senses. He invokes Ronald Dworkin’s distinction between ‘concepts’ and ‘conceptions’ to justify changes in the meanings of constitutional words. He argues that

the words of the Constitution, for the most part, describe concepts and purposes that are stated at a sufficiently high level of abstraction to enable events and matters falling within the current understanding of those concepts and purposes to be taken into account ... That being so, once we have identified the concepts ... that the makers of our Constitution intended to apply, we can give effect to the present day conceptions of those concepts.[125]

McHugh J supports these propositions with a further argument to the effect that ‘the relevant intention of constitutional provisions is that expressed in the Constitution itself, not the subjective intentions of its framers or makers. It is an intention that is determined objectively.’[126] It is determined ‘not [by] a search for the mental states of those who made, or for that matter approved or enacted, the Constitution ... [but] from the words that they used in the historical context in which they used them.’[127] He seems to believe that usually the only intended meanings that can be ‘objectively’ attributed to constitutional provisions are abstract concepts. The founders’ conceptions of those concepts are subjective, and therefore not binding.

On this view, there are two different kinds of ‘meaning’ — one called concepts, and the other conceptions — of which only the second can legitimately change. The court can attribute to a provision a conception that is different from that of the founders, but only if both are conceptions of the abstract concept that the founders intended to enact. Constitutional conceptions can change, but constitutional concepts cannot. Seen in this light, McHugh J’s theory is a version of moderate originalism after all.

He claims that this is consistent with the founders’ intentions:

[W]hen we see meaning in a constitutional provision which our predecessors did not see, the search is always for the objective intention of the makers of the Constitution.[128]

Indeed, many words and phrases of the Constitution are expressed at such a level of generality that the most sensible conclusion to be drawn from their use in a Constitution is that the makers of the Constitution intended that they should apply to whatever facts and circumstances succeeding generations thought they covered.[129]

After mentioning some examples of such general concepts, McHugh J says: ‘In these and other cases, the test is simply: what do these words mean to us as late twentieth century Australians?’[130]

This test may be harmless in relation to the examples he mentions. To take an example he does not mention, consider Parliament’s power to make laws with respect to ‘copyrights, patents of inventions and designs, and trade marks’ conferred by s 51(xviii) of the Constitution. These legal concepts are partly defined by the laws that constitute and protect the things they denote. As the laws that create patents change, so does the nature of patents. And the purpose of conferring the power on Parliament was to enable it to change those laws. The founders must have had a concept of ‘patents’ more abstract than its embodiment in the laws of their own day.[131] As the majority put the point in Grain Pool, there is a ‘dynamism which, even in 1900, was inherent in any understanding of the terms used in s 51(xviii).[132] Of course, there must be limits to that dynamism: Parliament cannot acquire power over anything at all by calling it a ‘patent’ or a ‘trade mark’.[133] But if there is no reason to think that our conceptions of these concepts are so different from the founders’ that they would regard them as exceeding the limits of their inherent dynamism, then McHugh J’s suggested test is harmless. To ask whether they would have regarded some new development as a suitable candidate for patent protection, it is sufficient to ask whether we so regard it.

But the test is too loose to be suitable for general application. To ask in every case no more than ‘what do the Constitution’s words mean to us as late twentieth century Australians?’ would be to adopt radical non-originalism. It would be to ignore the distinction between different concepts, on the one hand, and different conceptions of the same concept, on the other. A contemporary meaning can be a conception of an original concept, rather than a new concept, only if it is consistent with whatever criteria or standards govern the use of the original concept. So the test should be: does the current popular meaning of the words satisfy the criteria or standards that govern the use of the original concept intended by the founders?

Even if the distinction between concepts and conceptions is a valid one (and I suspect that it is much more philosophically problematic than the distinction between connotations and denotations), I have two reservations about its utility. The first reservation concerns the extent to which it is applicable. Dworkin used it to explain the nature of disagreements about abstract moral concepts such as ‘justice’ and ‘equality’. He used it to distinguish genuine disagreements, which are in principle resolvable, from interminable talk at cross-purposes about different concepts. If two parties to an apparent disagreement are really using the word ‘justice’ to express quite different concepts, their disagreement is illusory. They are really talking at cross-purposes, and once they realise this, they could agree that they are both right, because they are applying different concepts. Their disagreement can be a genuine one only if it involves competing conceptions of a single, more abstract concept to which they both subscribe. It makes sense to distinguish between concepts and conceptions where such elusive and controversial moral norms are concerned. Most of us are aware that our current conceptions of them are fallible, and may have to be revised. In other words, we are aware that there is a difference between the concepts themselves, and our current imperfect conceptions of them. But in the case of many other kinds of concepts, it is less clear that there is such a difference. In the cases of ‘table’ and ‘chair’, for example, it seems doubtful that we can draw the same distinction between elusive and controversial abstract concepts, and fallible conceptions of them.

My second reservation concerns the extent to which the distinction, even if it is sound and properly applied, adds much in practice to distinctions drawn in my earlier article, especially the one between enactment and application intentions.[134] According to Dworkin, when there is a disagreement between competing conceptions of the same concept, one of them must be a more accurate conception of it than the other.[135] If so, then adopting a contemporary conception of a constitutional concept that differs from the founders’ conception of it can be justified only if the latter is less accurate. McHugh J often writes as if that is true of examples he discusses.[136] For instance, he says that ‘[w]ith experience, we may see that the meaning of particular expressions in the Constitution is different from or more expansive than that which people understood in 1900.’[137] ‘[T]he events following federation and the experiences of the nation can be used to see more than the Constitutional Convention participants or the 1901 audience saw in particular words and combinations of words.’[138] His suggestion is that our conceptions of the founders’ abstract concepts are sometimes superior to their own conceptions. If so, to prefer our conceptions to theirs is to be faithful to their intentions in enacting those concepts.[139]

In such cases, my own theory of moderate originalism would also justify adopting the contemporary conception, because (ex hypothesi) the founders’ enactment intention was to constitutionalise the abstract concept, rather than an erroneous conception of it. Erroneous conceptions must be disregarded along with, and for the same reason as, erroneous application intentions. As I argued earlier, ‘[t]he law consists of the provision which the law-makers actually enacted, and not their possibly mistaken beliefs about its meaning or proper application’.[140]

I previously discussed at some length the possibility that in providing that members of Parliament be ‘directly chosen by the people’, the founders intended to enact an abstract principle of political morality, which should now be held to guarantee the right of women to vote, regardless of the founders’ contrary application intention.[141] The entire discussion could easily be recast in terms of a distinction between an abstract concept of political morality, and the founders’ erroneous conception of it as excluding women. I concluded that it was implausible to think that in using those words the founders did intend to enact an abstract moral concept. And that is likely to be the major question whenever the concept/conception distinction is invoked to justify the conclusion that the meaning of a constitutional provision has evolved. The question will be: is it really plausible to think that the founders’ enactment intention was to constitutionalise an abstract concept, rather than a more concrete one? It is all too easy to attribute such an intention to them in order to reach a desired conclusion. One of the most common complaints about Dworkin’s interpretations of the American Constitution is that he tends automatically to attribute such intentions to its founders, without showing the slightest interest in relevant historical evidence.[142]

B Evolutionary Originalism

Jeremy Kirk recently proposed a theory of ‘evolutionary’ originalism, which holds that

the meaning of the text is to be understood as originally intended, with some potential for evolution from this, subject to the outer constraint of being within the realm of what can reasonably be characterised as the original idea or concept.[143]

This is clearly a version of moderate originalism, since he insists that ‘substantial deviation from the original understanding of these ideas is not permitted.’[144]

Kirk describes the Constitution as often ‘communicating broad ideas and concepts, the width of which allow evolutionary changes in how particular subjects have been, can be or should be understood.’[145] This sounds much like McHugh J’s use of Dworkin’s distinction between concepts and conceptions.[146] But Kirk is more of a non-originalist than McHugh J. Kirk says that McHugh J requires ‘the essential meaning of a provision [to remain] constant’, whereas ‘[w]ithin evolutionary originalism, the accepted meaning itself may change, within limits’.[147] It is not clear how this claim can be reconciled with his insistence that ‘[j]udges should be limited to giving effect to original ideas’,[148] but presumably the answer lies in his statement that ‘substantial deviation from the original understanding of these ideas is not permitted.’[149] ‘[A]ny evolved meaning must remain within the realm of the original general idea, broadly understood.’[150] In other words, small changes to the original ideas or concepts are permissible, but not large ones. It is a question of degree.

Whether this differs significantly from my own version of moderate originalism is unclear. I previously argued that, provided it is clearly essential to achieve a provision’s original purpose, its original literal meaning can sometimes be adjusted or stretched — but not by too much.[151] So I, too, have conceded that it is sometimes a question of degree. But I would argue that my discussion of ‘purposive’ interpretation more clearly justifies, and more clearly limits, the court’s ability occasionally to stray beyond original literal meanings. I therefore remain of the opinion that the only ways in which our constitutional law can legitimately change, without the Constitution being formally amended, are those described in my previous article.[152]

X CONCLUSION

I have argued that Justice Kirby’s radical non-originalism has unacceptable consequences, in terms of democracy, the rule of law, and federalism. The original, intended meaning of the Constitution cannot be, as he claims, largely irrelevant to its current interpretation. An attempt to ascertain that meaning must be, at the very least, the starting point for interpretation. It is uncontroversial that the application of the Constitution can change as circumstances change. The question is how, and to what extent, its meaning can legitimately be supplemented or adjusted to meet contemporary needs, without a formal amendment. Orthodox principles of interpretation already provide some scope for that to happen. Even if they should be relaxed, the High Court should, and surely will, continue to be guided by interpretive principles consistent with moderate originalism. If, as a result, we find that the expense and difficulty of a referendum is sometimes required for desirable change, we should be neither surprised nor disappointed. That is a small price to pay for one of the most democratic constitutional amendment procedures in the world.[153]


[*] LLB (Hons) (Adelaide), LLM (Illinois), MA, PhD (UC Berkeley); Professor, Faculty of Law, Monash University.

[1] To establish these propositions would require an article in itself. But see, eg, J Story, Commentaries on the Constitution of the United States (first published 1833, reprinted 1987)

678–82, 693–8; Samuel Miller, Lectures on the Constitution of the United States (1893) 100–3; Henry Black, Handbook of American Constitutional Law (3rd ed, 1910) 75–81; William Casto, The Supreme Court in the Early Republic: The Chief Justiceships of John Jay and Oliver Ellsworth (1995) 234–6.

[2] The early history of the idea is described in Thomas Peebles, ‘A Call to High Debate: The Organic Constitution in Its Formative Era, 1890–1920’ (1980) 52 University of Colorado Law Review 49.

[3] See below Part VII.

[4] Jeffrey Goldsworthy, ‘Originalism in Constitutional Interpretation’ [1997] FedLawRw 1; (1997) 25 Federal Law Review 1, 8–16.

[5] Ibid.

[6] Jeremy Kirk, ‘Constitutional Interpretation and a Theory of Evolutionary Originalism’ (1999) 27 Federal Law Review 323. For further discussion, see below Part IX(B).

[7] Justice Michael Kirby, ‘Constitutional Interpretation and Original Intent: A Form of Ancestor Worship?’ [2000] MelbULawRw 1; (2000) 24 Melbourne University Law Review 1, 1, 6.

[8] Ibid 2.

[9] Ibid 8.

[10] Ibid 4.

[11] He says, for example: ‘So there are the competing views. The one that a constitution is anti-evolutionary. The other that it must be evolutionary’: ibid 6. No third alternative is mentioned in this paragraph.

[12] Ibid 10.

[13] Ibid 14.

[14] Ibid (emphasis added).

[15] Grain Pool of WA v Commonwealth (2000) 170 ALR 111, 134 (emphasis added) (‘Grain Pool’). Similar words are used in the abstract to Kirby, above n 7, 1.

[16] Grain Pool (2000) 170 ALR 111, 140 (emphasis added).

[17] Ibid 142 (emphasis added).

[18] Ibid 139 (emphasis added).

[19] Ibid 141–2.

[20] Kirby, above n 7, 11.

[21] Re Wakim; Ex parte McNally (1999) 198 CLR 511, 600 (‘Re Wakim’).

[22] [1998] HCA 22; (1998) 195 CLR 337, 417 (‘Kartinyeri’). See also Newcrest Mining (WA) Ltd v Commonwealth [1997] HCA 38; (1997) 190 CLR 513, 657.

[23] Kirk, above n 6, 364.

[24] I have previously identified these words as a problem for non-originalists: see Goldsworthy, ‘Originalism’, above n 4, 39; Jeffrey Goldsworthy, ‘The High Court, Implied Rights and Constitutional Change’ (1995) 39(3) Quadrant 46. Non-originalists have failed to confront the problem.

[25] R v Burah (1878) 3 App Cas 889; Hodge v The Queen (1883) 9 App Cas 117; Powell v Apollo Candle Co Ltd (1885) 10 App Cas 282; Riel v The Queen; Ex parte Riel (1885) 10 App Cas 675.

[26] See D’Emden v Pedder [1904] HCA 1; (1904) 1 CLR 91, 110–11 (Griffith CJ). These Privy Council decisions are discussed in Union Steamship Co of Australia Pty Ltd v King [1988] HCA 55; (1988) 166 CLR 1, 9–10.

[27] Quoted in the text accompanying n 22, above.

[28] As Justice Kirby rightly says, ‘even an assertion that a particular construction of the text is “settled” by many past decisions does not necessarily bolt the door against re-examination if new scrutiny is considered necessary by the majority of the Justices of the court’: above n 7, 2.

[29] In Re Governor, Goulburn Correctional Centre, Goulburn; Ex parte Eastman, Kirby J said that the court was ‘as free today, as the court felt itself free in the Engineers’ case, to throw off the holdings of the past, not because they were wrong when made but because, read in a new light, the Constitution can now be seen to bear a different meaning’: (1999) 165 ALR 171, 193–4.

[30] The same point applies to other recent commentators. For example, George Williams has suggested that on questions of human rights, the founders’ attitudes and intentions should be ignored because they were racist and are therefore inconsistent with contemporary values: George Williams, Human Rights under the Australian Constitution (1999) 45. On the other hand, he later argues that the ‘text and structure’ of the Constitution provide only limited scope for expanding the protection of human rights through constitutional interpretation: at 249. But, given the presence of the words ‘peace, order, and good government’, if the founders’ attitudes and intentions are ignored, the text of the Constitution is entirely compatible with judicial protection of a comprehensive package of human rights. There is no logical reason for adhering to the original, intended meaning of these general words, while rejecting that of other, more particular words.

[31] Kirby, above n 7, 1, 6.

[32] Occasional suggestions that this aspect of s 128 is unacceptably undemocratic are puzzling. A national majority has no more inherent right to prevail on every question that arises in a federation, which is an organisation of partly independent states, than an international majority has an inherent right to prevail on every question that arises in the world of fully independent states.

[33] Kartinyeri [1998] HCA 22; (1998) 195 CLR 337, 399.

[34] Goldsworthy, ‘Originalism’, above n 4, 10.

[35] Arthur Machen Jr, ‘The Elasticity of the Constitution(1900) 14 Harvard Law Review 200,

203–4:

Changes in the meaning of words are entirely fortuitous. A catching comic song, the whim of a popular writer, or a thousand other purely accidental circumstances may affect the signification of words far older than the Constitution. It cannot be that the fundamental law of the land is subject to variation with every such passing fashion. Otherwise, the interpretation of the Constitution would be the sport of chance and as variable as fortune herself.

[36] Goldsworthy, ‘Originalism’, above n 4, 38.

[37] See text following n 29, above.

[38] Re Wakim (1999) 198 CLR 511, 600.

[39] I will not repeat here all the arguments against the ‘dead hand of the past’ complaint: see Goldsworthy, ‘Originalism’, above n 4, 27.

[40] (1981) 146 CLR 559 (‘DOGS Case’).

[41] Goldsworthy, ‘Originalism’, above n 4, 36–9.

[42] Ibid 36–7 (citations omitted). For more detailed explanation of this point, see Jeffrey Goldsworthy, ‘Marmor on Meaning, Interpretation, and Legislative Intention’ (1995) 1 Legal Theory 439, 452–8.

[43] Goldsworthy, ‘Originalism’, above n 4, 37–9.

[44] Kirby, above n 7, 1.

[45] Ibid 7.

[46] Ibid 9.

[47] [1988] HCA 18; (1988) 165 CLR 360.

[48] Kirby, above n 7, 10.

[49] Goldsworthy, ‘Originalism’, above n 4, 12–15.

[50] Henry Burmester, ‘The Convention Debates and the Interpretation of the Constitution’ in Gregory Craven (ed), The Convention Debates 1891–1898: Commentaries, Indices and Guide (1986) 25, 30.

[51] Kirk, above n 6, 324–5. See also Carl McCamish, ‘The Use of Historical Materials in Interpreting the Commonwealth Constitution(1996) 70 Australian Law Journal 638.

[52] Hawkins v Gathercole [1855] EngR 119; (1855) 6 De G M & G 1, 22; [1855] EngR 119; 43 ER 1129, 1136 (Turner LJ). This principle was affirmed in many cases, including A-G (UK) v Sillem [1863] EngR 989; (1863) 2 H & C 431, 531; [1863] EngR 989; 159 ER 178, 223 (Bramwell B); Holme v Guy [1877] UKLawRpCh 129; (1877) 5 Ch D 901, 905 (Jessel MR); River Wear Commissioners v Adamson [1877] 2 AC 743, 763–4 (Lord Blackburn); Read v Bishop of Lincoln [1892] UKLawRpAC 40; [1892] AC 644, 652–3 (Lord Halsbury LC); Herron v Rathmines and Rathgar Improvement Commissioners [1892] UKLawRpAC 25; [1892] AC 498, 502 (Lord Halsbury LC); Eastman Photographic Materials Co, Ltd v Comptroller-General of Patents, Designs, and Trademarks [1898] UKLawRpAC 44; [1898] AC 571, 573–6 (Lord Halsbury LC).

[53] Kirk, above n 6, 344.

[54] Ibid 329 (citations omitted).

[55] A Inglis Clark, Studies in Australian Constitutional Law (1st ed, 1901).

[56] Kirby, above n 7, 10–11.

[57] Ibid 11, quoting Clark, above n 55, 21.

[58] Clark, above n 55, 21–2. I previously pointed this out in Goldsworthy, ‘Originalism’, above n 4, 17.

[59] Kirk, above n 6, 333, fn 76.

[60] Clark, above n 55, 19–20. The longer passages quoted then follow, in order to explain and justify this method of interpretation, although how they do so is not entirely clear.

[61] R v Commonwealth Conciliation and Arbitration Commission; Ex parte Professional Engineers’ Association [1959] HCA 47; (1959) 107 CLR 208.

[62] Victoria v Commonwealth [1971] HCA 16; (1971) 122 CLR 353, 396–7 (‘Payroll Tax Case’), discussed in Goldsworthy, ‘Originalism’, above n 4, 16.

[63] [1969] HCA 31; (1969) 122 CLR 177, 230.

[64] Engineers’ Case [1959] HCA 47; (1959) 107 CLR 208, 267.

[65] Goldsworthy, ‘Originalism’, above n 4, especially 28–35, 50.

[66] [1999] HCA 30; (1999) 163 ALR 648.

[67] Kirby, above n 7, 13.

[68] Goldsworthy, ‘Originalism’, above n 4, 31.

[69] Sue v Hill [1999] HCA 30; (1999) 163 ALR 648, 692–4.

[70] Kirby, above n 7, 12.

[71] Ibid fn 64.

[72] Ibid 13. See the similar comment in Kartinyeri [1998] HCA 22; (1998) 195 CLR 337, 400.

[73] See the excellent discussion in Leslie Zines, ‘Characterisation of Commonwealth Laws’ in H P Lee and George Winterton (eds), Australian Constitutional Perspectives (1992) 33, 34–9, and my comment in Goldsworthy, ‘Originalism’, above n 4, 33, text accompanying fn 179.

[74] [1935] HCA 78; (1935) 54 CLR 262.

[75] Kirby, above n 7, 7.

[76] Ibid.

[77] Randy Barnett, ‘An Originalism for Nonoriginalists’ (1999) 45 Loyola Law Review 611, 614. I am grateful to Nick Aroney for drawing this article to my attention.

[78] Ibid 613.

[79] Ibid 623.

[80] Ibid 617.

[81] Ibid 619.

[82] Ibid 619–20.

[83] Ibid 615–20.

[84] Richard Saphire, ‘Originalism and the Importance of Constitutional Aspirations’ (1997) 24 Hastings Constitutional Law Quarterly 599, 609.

[85] Michael Perry, The Constitution in the Courts: Law or Politics? (1994) 9.

[86] Ibid ch 2, especially 47–50 (defence of originalism), and chh 8–9 (application to activist decisions).

[87] For full details, see Jeffrey Goldsworthy, ‘Dworkin as an Originalist’ (2000) 17 Constitutional Commentary 49.

[88] Ronald Dworkin, ‘Comment’ (on Antonin Scalia, ‘Common-Law Courts in a Civil-Law System: The Role of United States Federal Courts in Interpreting the Constitution and Laws’) in Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law (1997) 122, 122.

[89] Ibid.

[90] Ronald Dworkin, Freedom’s Law: The Moral Reading of the American Constitution (1996) 13. For many other similar statements, and a discussion of Dworkin’s current views, see Goldsworthy, ‘Dworkin as an Originalist’, above n 87.

[91] Dworkin, ‘Comment’, above n 88, 126 (emphasis in original).

[92] Ronald Dworkin, A Matter of Principle (1985) 49.

[93] James Fleming, ‘Fidelity to Our Imperfect Constitution(1997) 65 Fordham Law Review 1335, 1344, quoted in Barnett, above n 77, fn 25.

[94] Jack Rakove, ‘Fidelity through History (or to It)’ (1997) 65 Fordham Law Review 1587, fn 14, cited in Barnett, above n 77, 617, fn 24.

[95] Paul Horwitz, ‘The Past, Tense: The History of Crisis — and the Crisis of History — in Constitutional Theory’ (1997) 61 Albany Law Review 459, 472. See also Jonathan Macey, ‘Originalism as an “ism” ’ (1996) 9 Harvard Journal of Law and Public Policy 301, 306.

[96] Michael Dorf, ‘Integrating Normative and Descriptive Constitutional Theory: The Case of Original Meaning’ (1997) 85 Georgetown Law Journal 1765, 1766 (citations omitted).

[97] Cass Sunstein, ‘Justice Scalia’s Democratic Formalism’ (1997) 107 Yale Law Journal 529, 558 (citations omitted).

[98] Goldsworthy, ‘Originalism’, above n 4, 37–8.

[99] [1993] HCA 44; (1993) 177 CLR 541, 562.

[100] Kartinyeri v Commonwealth [1998] HCA 22; (1998) 195 CLR 337, fn 271.

[101] Goldsworthy, ‘Originalism’, above n 4, 49.

[102] Discussed in Re Wakim (1999) 198 CLR 511, 553 (McHugh J) and Grain Pool (2000) 170 ALR 111, 145 (Kirby J).

[103] Goldsworthy, ‘Originalism’, above n 4, 33–5.

[104] US Constitution art I § 8.

[105] Goldsworthy, ‘Originalism’, above n 4, 33.

[106] Lawrence Lessig, ‘Fidelity in Translation’ (1993) 71 Texas Law Review 1165, 1185.

[107] Goldsworthy, ‘Originalism’, above n 4, 33–4.

[108] A fourth reason concerns the phrase ‘for whom it is deemed necessary’, which Gaudron J relied on to limit the scope of the power in Kartinyeri [1998] HCA 22; (1998) 195 CLR 337. I put this aside for the purposes of discussion. This is not the occasion for a full treatment of the race power.

[109] [1998] HCA 22; (1998) 195 CLR 337.

[110] Ibid 401.

[111] Ibid 413.

[112] Ibid. See also at 417, where he takes account of what ‘the electors of this country were generally aware of’ when they voted in the referendum.

[113] On the distinction between application and enactment intentions, see Goldsworthy, ‘Originalism’, above n 4, 30–1.

[114] Kartinyeri [1998] HCA 22; (1998) 195 CLR 337, 363.

[115] But see above n 108.

[116] [1998] HCA 22; (1998) 195 CLR 337, 414.

[117] The doctrine is partly enshrined in the words ‘peace, order, and good government’, as explained above at n 26.

[118] Jeffrey Goldsworthy, The Sovereignty of Parliament: History and Philosophy (1999) especially 261–71.

[119] [1997] HCA 27; (1997) 190 CLR 1 (Dawson, Gaudron, McHugh and Gummow JJ).

[120] Re Wakim (1999) 198 CLR 511; Eastman v The Queen [2000] HCA 29; (2000) 172 ALR 39 (‘Eastman’).

[121] Re Wakim (1999) 198 CLR 511, 549.

[122] Eastman [2000] HCA 29; (2000) 172 ALR 39, 73. See also at 66, 69, 70; Re Wakim (1999) 198 CLR 511, 533.

[123] Re Wakim (1999) 198 CLR 511, 549–52; Eastman [2000] HCA 29; (2000) 172 ALR 39, 69, 72, 73–4. To this could be added many quotations to the same effect that he includes from earlier judgments, but it is often unclear whether or not (or to what extent) he approves of them. For example, it is not clear to what extent he approves of what he calls ‘the traditional view of constitutional interpretation’, which he sets out at length in Eastman [2000] HCA 29; (2000) 172 ALR 39, 66–8.

[124] Re Wakim (1999) 198 CLR 511, 551–2; Eastman [2000] HCA 29; (2000) 172 ALR 39, 66 and 73.

[125] Re Wakim (1999) 198 CLR 511, 552. See also Eastman [2000] HCA 29; (2000) 172 ALR 39, 72–3.

[126] Eastman [2000] HCA 29; (2000) 172 ALR 39, 69.

[127] Re Wakim (1999) 198 CLR 511, 551.

[128] Eastman [2000] HCA 29; (2000) 172 ALR 39, 72.

[129] Re Wakim (1999) 198 CLR 511, 552.

[130] Ibid 553.

[131] This was the point made, but with specific reference to trade marks, by Higgins J in A-G (NSW) v Brewery Employés Union of NSW (‘Union Label Case’) [1908] HCA 94; (1908) 6 CLR 469, 610–12.

[132] (2000) 170 ALR 111, 118 (Gleeson CJ, Gaudron, McHugh, Gummow, Hayne and Callinan JJ).

[133] Ibid 145 (Kirby J).

[134] See Goldsworthy, ‘Originalism’, above n 4, 30–1.

[135] This follows from his well-known ‘right answer’ thesis.

[136] Note that the founders’ conception and the contemporary one could be equally correct or valid only if the abstract concept were a ‘relative’ one, or they were not competing conceptions of the same abstract concept. I have previously discussed possibilities of the former kind (Goldsworthy, ‘Originalism’, above n 4, 42–3). In the latter case, the ‘conceptions’ would really be different concepts. In that case, abandoning the founders’ concept and adopting a different one would not be consistent with any version of originalism: it would not be faithful to ‘the intention of the makers of the Constitution as evinced by the terms in which they expressed that intention’: Re Wakim (1999) 198 CLR 511, 549 (McHugh J).

[137] Eastman [2000] HCA 29; (2000) 172 ALR 39, 66. See also at 70.

[138] Ibid 73; see also Re Wakim (1999) 198 CLR 511, 553–4.

[139] Re Wakim (1999) 198 CLR 511, 551; Eastman [2000] HCA 29; (2000) 172 ALR 39, 72–3.

[140] Goldsworthy, ‘Originalism’, above n 4, 30.

[141] Ibid 47–9.

[142] See, eg, Keith Whittington, ‘Dworkin’s “Originalism”: The Role of Intentions in Constitutional Interpretation’ (2000) 62 The Review of Politics 197.

[143] Kirk, above n 6, 359.

[144] Ibid 361.

[145] Ibid.

[146] Kirk adverts to Dworkin’s distinction: ibid.

[147] Ibid 365. See also at 358.

[148] Ibid 359.

[149] Ibid 361 (emphasis added).

[150] Ibid 362 (emphasis added).

[151] See above n 107 and accompanying text.

[152] Goldsworthy, ‘Originalism’, above n 4.

[153] It is by no means ideally democratic, mainly because the initiation of amendments is tightly controlled by the executive government. Section 128 should itself be changed to enable amendment proposals to be put to the people by the States. See Jeffrey Goldsworthy, ‘A Role for the States in Initiating Referendums’ in Upholding the Australian Constitution: Proceedings of the Eighth Conference of the Samuel Griffith Society (1997) 39.