I. Introduction
The people are rarely asked to change the Australian Constitution, and when they are, they rarely say yes.Footnote 1 Following the defeat of the proposal to amend the Constitution to enshrine an indigenous voice to Parliament in 2023, some have questioned whether the amendment procedure provided by section 128 has effectively become a dead letter.Footnote 2 One constitutional expert asked: ‘What hope is there for future referendums in Australia at the federal level …?’ The answer, she suggested, is ‘very little.’Footnote 3
This article considers the implications of this constitutional inertia for the High Court and its approach to constitutional interpretation. The High Court favours an interpretive approach that I call legalism, a core tenet of which is that the Court cannot change the meaning of the Constitution. This approach is clearly informed by the presence of section 128, and the power it vests in the people and their representatives to change the Constitution. The desuetude of section 128 therefore raises the question of whether the Court’s approach is still sound.
In examining this question, I continue a conversation between two of Australia’s leading constitutional scholars, Jeffrey Goldsworthy and Lael K Weis. Goldsworthy presented the canonical defence of originalism — and more specifically, the principle that the High Court cannot change the Constitution under the guise of interpreting it. That would not merely be unlawful, Goldsworthy argued, but would ‘flout … the principle of democracy’.Footnote 4 Weis argued that this is unpersuasive: that legalism does not serve democracy in any clear or direct sense, and that we should reject fidelity to democracy as ‘a criterion of interpretive legitimacy’.Footnote 5
My position is closer to Goldsworthy’s, though I agree with Weis that the connection between interpretive method and democracy requires closer examination. I conclude that the interpretive method of the High Court remains normatively sound, and in that sense my ultimate aim is to defend the status quo. But the difficulty of amending the Constitution does require a sober reassessment and restatement of the precise values served by the Court’s approach. I argue that legalism still serves democracy, but in a thinner sense than has sometimes been claimed. It is more persuasively explained and justified as a way of sustaining law’s perceived autonomy from politics, and thus the legitimacy of the Court.
Before presenting this argument, it is important to clarify its core premise. That is, that section 128 has fallen into disuse. Some readers might think that the problem is more pronounced, and that section 128 has altogether failed — pointing for example to evidence of low constitutional literacy or other pathologies of the referendum process. Others might insist that section 128 is functioning just as it ought to. I think that it is difficult to say with certainty why the people and their representatives are reluctant to change the Constitution, or measure success or failure in this regard. But I also find it implausible that the lack of constitutional amendments reflects no more than the happy fact that the people and their representatives are happy with the Constitution in its present form. While Australians are right to be proud of certain aspects of their Constitution, and could point to many ways in which it works well, no one could sensibly say that the Constitution is perfect. It is also clear that Parliament is discouraged from putting proposals for constitutional change for partisan political reasons disconnected from the need for change or the merits of any particular proposal. But for the purposes of this article, desuetude is enough, and I do not seek to show that section 128 has failed. Successful referendums are rare, and experts doubt that any will occur in the foreseeable future. This is sufficient to prompt reconsideration of the High Court’s interpretive method because the High Court has referred to the possibility of constitutional amendment to justify that method in a way that does not squarely confront the reality that such amendments are unlikely to be made.
Three further qualifications must be made. First, while there have been few formal constitutional amendments, this does not mean that Australia has not seen constitutional change. Constitutional doctrine has changed significantly since 1901, and sometimes dramatically so, even though the constitutional text has all but stood still.Footnote 6 New rules and principles have been discovered (sometimes contentiously), and long-standing doctrines have evolved. Some might take that to mean that the orthodox position being defended in this article is not really orthodoxy at all, for the Court evidently changes the Constitution all the time. Moreover, this would mean that the lack of formal constitutional amendment is not really a problem, because the Court can simply amend the Constitution via “interpreting” it. I disagree. As I will explain, legalism can accommodate constitutional change, though only to a limited extent. For this reason, we can say that the High Court eschews any power to change the meaning of the Constitution, and acknowledge that constitutional doctrine has evolved, while also recognising that the lack of formal constitutional amendment significantly constrains the degree of evolution that can occur.
Secondly, this article considers constitutional amendment at the systemic level, and makes no comment on the reasons why any particular referendum failed, or whether any particular one ought to have succeeded. And finally, while my focus is Australia, the problems discussed here are not unique to that legal system. There are other legal systems with old Constitutions that have also proven to be very difficult to amend.Footnote 7 I focus on Australia because my aim is to make a sustained contribution to domestic thought at a time of reflection about how our constitutional system works and what it stands for. I begin by articulating the way the High Court interprets that Constitution in Part II, before considering whether that approach withstands normative scrutiny in Part III.
II. Australian legalism
It is sometimes said that the High Court has not endorsed any particular method of constitutional interpretation.Footnote 8 Some judges have articulated their preferred method, at least extra-curially,Footnote 9 but many have not, and others have said that the theories of constitutional interpretation that abound in the pages of academic writing are unhelpful, ‘serv[ing] only to obscure much more than they illuminate.’Footnote 10 Yet, it is possible to articulate an approach to constitutional interpretation that has been endorsed and applied by a majority of judges for most of the High Court’s history, even if only at a high level of generality. I call that approach legalism, borrowing of course the language of Sir Owen Dixon.
Australian legalism gives primacy to the text of the Constitution. When I say ‘text’, I include the way in which that text is organised, that is its structure. But it would be remiss to call the Australian approach ‘textualism’. The text is always read in context, which is understood to include the history of the Constitution and its provisions and publicly available evidence of the purposes they were intended to serve. The Court also gives interpretive weight to the values and principles immanent in the constitutional text — an important point, that I explain further below. Legalism so defined makes three cascading commitments that are particularly relevant for my purposes. The first is that there is a distinction between constitutional interpretation and constitutional amendment.Footnote 11 The second is that courts should only do the former and not the latter. The third is that courts should hold this line, even if it appears that the Constitution so interpreted does not align with contemporary needs or values.
The label of legalism captures the Court’s concern to distinguish its method from moral or political reasoning. Accordingly, the Court ‘seeks the answers to legal questions in legal materials so far as possible and … avoid[s], so far as possible, reliance on extra-legal values or reference to consequences.’Footnote 12 Chief among these is, of course, the text of the Constitution itself. This does not ‘den[y] the existence of judicial choice’, but it does generally deny that courts ‘should act “pragmatically”, by balancing legal considerations against, and sometimes finding them outweighed by, considerations of justice or public policy’.Footnote 13 Where the Constitution appears to permit a choice — to run out, as the theorist might describe it — the Court evidently strives to fill the gap by reference to legal sources, in a way that is coherent with what the Constitution does provide. Typically, this leads to the conclusion that government power is (legally) unlimited, for that conclusion is most consistent with the principle of self-government that imbues the text and structure of the Constitution.Footnote 14
Thus the Court will generally ‘lean to the broader interpretation’ of a constitutional provision, particularly when the provision in question is a conferral of legislative power.Footnote 15 Secondly, the Court will not find that legislative power is subject to implied limits not expressed in the text of the Constitution or necessarily implied by its text or structure. This is exemplified by the cases concerning freedom of movement and the rule of law, discussed below. Thirdly, and relatedly, the Court will not read down provisions of the Constitution out of fear that that the broader construction will lead to inconvenient results, or even injustice. It says that ‘the extravagant use of the granted powers in the actual working of the Constitution is a matter to be guarded against by the constituencies and not by the Courts’.Footnote 16 Or in other words, the ‘possible abuse of power is rarely a convincing reason for denying its existence’.Footnote 17
Some have perceived an increasing tendency in the Court to look to its own precedent, or ordinary legislation, to resolve difficult questions of constitutional meaning.Footnote 18 That raises real questions, for it is difficult to reconcile with the accepted hierarchy between constitutional and non-constitutional norms.Footnote 19 But it seems consistent with the idea of legalism, that constitutional questions should be answered by reference to legal sources, rather than moral or political concerns.
This approach is made palatable – or at least, arguably so – by the fact that the Constitution does limit the government’s ability to interfere with democracy itself. The implied freedom of political communication might be understood as the result of the Mason’s Court’s foray into more creative interpretive techniques, fuelled by some judges’ conceptions of the interpretive weight that could be afforded to ideas of popular sovereignty and the consequent need to judicially protect the rights and freedoms of the people.Footnote 20 That was always a shaky foundation, for it is not at all clear why respect for the power the Constitution vests in the people — particularly, the power to change the Constitution — should lead the Court to change the Constitution for them.Footnote 21 In hindsight, these interpretive techniques look like outliers. But the implied freedom has become a firmly entrenched part of the jurisprudence. A different kind of appeal to democracy has been instrumental in squaring this circle. The implied freedom of political communication is now described as a necessary outworking of ‘the great underlying principle of the Constitution … that the rights of individuals [are] sufficiently secured by ensuring each an equal share in political power.’Footnote 22 That principle is clearly reflected in the express provisions of the Constitution, but also in its silences: the legal limits placed on legislative power are relatively thin, for the very reason that the legislature is ‘directly chosen by the people’ and possesses the power to change the very document that gives the legislature its powers. This explains why the Court is willing to recognise and enforce this implied limitation on legislative power, but not others.Footnote 23 Indeed, it shores up the legitimacy of legalism itself: the Court is justified in taking a restrained approach to constitutional review in the majority of cases, entrusting a great deal to the political process, because the core of the political process is itself protected by law.Footnote 24
The way in which ideas of democracy inform legalism demonstrates that it is a richer mode of analysis than textualism — for it does look beyond the semantic meaning of the words used to the principles and values that the document serves. But it is the internal morality of the Constitution that the Court seeks to find: the principles and values that are immanent in that text, rather than ones that are imported from elsewhere. For example, the Court has refused to recognise ‘the rule of law’ as an implied limit on legislative power. No one can deny that the rule of law is one of the most important values that the Constitution was created to serve. But government is only subject to the Constitution, and not the free-floating concept of the rule of law. As the plurality explained it in Palmer v Western Australia (in which Clive Palmer challenged West Australian legislation that extinguished his right to damages against the government of that State):
Of the rule of law, no less than of “representative democracy”, it “is logically impermissible to treat [the term] as though it were contained in the Constitution, to attribute to the term a meaning or content derived from sources extrinsic to the Constitution and then to invalidate a law for inconsistency with the meaning or content so attributed”.Footnote 25
The plaintiff’s claim to a constitutionally implied freedom of movement in Gerner v Victoria Footnote 26 was rejected for similar reasons. The plaintiff argued that this freedom derived from the nature of Australian society; more specifically, by the fact that the Constitution created one Australian people whereas previously there had been several peoples and several colonies. ‘Freedom to move wherever one wishes for whatever reason was said to follow, it being the essence of being a community or society or nation that the people can know one another.’Footnote 27 This, the Court said, got things in entirely the wrong order:
the question is not, “what is required by federation”. It is now well settled that what the Constitution implies depends on “what … the terms and structure of the Constitution prohibit, authorise or require”. That is because federation is not a “one size fits all” proposition; the kind of federation that is created depends on the text and structure of its constitutive instrument. So, just as the Constitution “gives effect to the institution of ‘representative government’ only to the extent that the text and structure of the Constitution establish it”, the legal nature and effect of the federation established by the Constitution can be known only from the terms and structure of the Constitution itself.Footnote 28
The Court went on to conclude that there was insufficient basis in the text and structure of the Constitution for implying a constitutional freedom of movement.
People may reasonably disagree about whether the Court has always articulated the values immanent in the Constitution correctly. The robust conception of individual liberty from state punishment that underpins the Court’s interpretation of Chapter III is one contentious example.Footnote 29 Whether or not it is convincing, the Court has justified that approach in a distinctly legalistic fashion – pointing to limits on the government’s power to punish and detain that were well-accepted at Federation, and hence could be said to inform the meaning of the Constitution in an orthodox way.Footnote 30 For present purposes, it is important that this approach is backwards looking. The Court’s claim is that protection from extra-judicial punishment is part of the original and fixed meaning of the Constitution, though the implications of this took time to unearth. The claim is not that the Court ought now to protect people from punishment divorced from a judicial finding of guilt, because that is contrary to contemporary values. Of course, the realist might say that this is what the Court is doing, and merely dressing up its preferred reading in legalist garb. But as scholars like Goldsworthy and Roux have shown, realism does not adequately explain the Australian case law.Footnote 31 I also agree with Goldsworthy, that it matters how Courts explain their reasoning and articulate their approach.Footnote 32 This gives us good insight into what judges perceive their role to be, and shapes the way that legal arguments are presented — and in turn the law.Footnote 33
The political process placed at the centre of the Constitution most obviously comprises of the ordinary electoral process and workings of the State and federal Parliaments, but it includes section 128 too. And many members of the Court have explicitly referred to section 128 to justify their interpretive stance and the conclusions they have reached in individual cases, as the next section explains.
A. Legalism and section 128
Weis explains that:
the formal amendment procedure occupies an important place in Australian constitutionalism. It is critical to Australian constitutionalism that the formal amendment procedure be understood as the most important site —arguably, even the exclusive site — for the emergence of the popular will in its constitutional law-making capacity.Footnote 34
That is so, despite the fact it is so seldom used, and despite the fact it does not empower the people to change the Constitution directly. Rather, the popular will is channelled, as always, through their representatives – who must first decide whether or not to put a proposal for constitutional amendment to the people.
In a later piece, Weis uses the disagreement between McHugh J and Kirby J in Al-Kateb v Godwin Footnote 35 to illustrate how appeals to democracy — and in particular, the somewhat democratic method of constitutional amendment provided for by section 128 — informs the dominant interpretive method of the Court. Here, Kirby J urged the Court to recognise that potentially indefinite detention of asylum seekers was unconstitutional, in part because it was incongruent with contemporary values and the current norms of international law; McHugh J put the orthodox position, that the Constitution could only be changed via section 128, even though he recognised that it produced ‘tragic’ outcomes for people like Al-Kateb.Footnote 36 As Weis explains it: ‘[t]he amendment rule [in section 128] itself embodies the constitutional commitment to democracy: constitutional change outside of that rule, however desirable, is an illegitimate usurpation of the popular will.Footnote 37 So stated, this is perfectly right, but for my purposes it is important to disentangle the various claims tied up in this statement: interwoven ideas of legitimacy, legality, and the harm that would be caused if the Court usurped a power vested in the people. That is important, because some of these claims are more implicated by the desuetude of section 128 than others.
The first is a legal proposition. Several judges have said that it would not be an exercise of judicial power to amend the Constitution. Some have indicated that this is something the Court inherently lacks, ‘owing its existence and its jurisdiction ultimately to the Constitution’, and there is some merit to that claim.Footnote 38 But the point is commonly bolstered by the reference to section 128, and the lines it draws around the possibility of constitutional amendment. McHugh J expressed this view in Re Wakim; Ex parte McNally:
the judiciary has no power to amend or modernise the Constitution to give effect to what the judges think is in the public interest. … Change to the terms and structure of the Constitution can be carried out only with the approval of the people in accordance with the procedures laid down in s 128 of the Constitution.Footnote 39
Similarly, in SGH Ltd v Federal Commissioner of Taxation, Gummow J said that:
It would be to pervert the purpose of the judicial power if, without recourse to the mechanism provided by s 128 and entrusted to the Parliament and the electors, the Constitution meant no more than what it appears to mean from time to time to successive judges.Footnote 40
In the recent decision of Vanderstock v Victoria, Justice Gordon rejected the approach taken by the majority ‘because it is not construing the text of the Constitution; instead it amends the Constitution’Footnote 41 — though her Honour was writing in dissent.
These passages treat section 128 as a textual justification for an interpretive approach that prioritises text. The fact that an express provision of the Constitution refers to its alteration, and stipulates that this thing called altering the Constitution ‘shall not’ occur except in the way prescribed is understood to commit the Court to drawing a distinction, however difficult that might be, between constitutional interpretation and constitutional amendment; moreover, it clearly indicates that only the former is within the power constitutionally vested in the Court. Judicial amendment of the Constitution is therefore said to be unlawful.
This assumes what theorists would call semantic fixation: that the meaning and thus the content of the Constitution was fixed at Federation. Writing extra-curially, Justice Edelman described this as a basic commitment to originalism, as distinct from ‘creationism’: a ‘form of interpretation’ in which judges ‘create new meaning’.Footnote 42 Edelman J described this as ‘unchallenged orthodoxy’: ‘[i]n this sense, there is no Australian judges of constitutional law who is not originalist.’Footnote 43 Yet, as Edelman has also emphasised in his extra-curial writing, a text of fixed meaning can still adapt to new circumstances. The orthodox distinction between connotation and denotation is one tool that allows this to occur.Footnote 44 Judicial understandings of the Constitution can change and indeed be perfected over time, as new laws and other events cast different and better light on what it means.Footnote 45 The High Court clearly endorses this, given it is willing to overturn constitutional precedent (albeit it cautiously, given amongst other things that government may have come to rely upon previous interpretations).Footnote 46
If we accept that amending the Constitution falls outside the concept of judicial power, then the lack of constitutional amendment under section 128 does not really matter. The desuetude of section 128 does not seem to enlarge the judicial power of the Commonwealth, and judicial amendment would remain unlawful no matter how practically or politically desirable it comes to seem. In other words, the core tenet of legalism — that courts cannot change the meaning of the Constitution, including to keep pace with contemporary needs or values — is still sound.
This assumes that there are very weighty reasons why the High Court should obey the law, which is obviously right. There may be a point at which the reasons why we want the Court to obey the law come to be outweighed by the reasons why they ought to break it — in short, where the Constitution they are charged with interpreting and applying becomes radically unworkable or unjust.Footnote 47 But the question of when a Court is morally obligated to break the law is more difficult to answer than that of when an ordinary person is justified in doing so. To give but one example of the kind of questions that would arise in this territory, would it not be better for the judge to simply resign?Footnote 48 Short of reaching that extreme position, the legal argument against judicial amendment holds. But as Weis observed, this legal proposition is not the only one the Court has offered to justify its interpretive stance. It — and its observers — have also justified that approach on richer normative justifications, which require closer reconsideration.
The first is the idea of judicial legitimacy, advanced by Justice Dixon. The claim – famously made and much critiqued – was that by ‘close adherence to legal reasoning’, courts could demonstrate that they were immune from politics, and thus their judgments could be trusted as an impartial application of the law.Footnote 49 Gleeson J emphasised this idea, writing extra-curially:
The power to declare invalid an expression of the will of a democratically elected legislature involves a responsibility of a special kind. The existence of an “unelected body with a capacity to decide that an enactment of an elected Parliament is ‘without effect will only be accepted if the community is confident that the power will be exercised for the purpose for which, and in accordance with the conditions upon which, it was given. This was the reason behind Sir Owen Dixon’s famous observation concerning the need for strict and complete legalism in the resolution of federal issues.
It is sometimes overlooked that what he said, in its context, contained a challenge. He asked, in effect: what is the competing view? If the High Court is not to resolve federal conflicts by a legalistic method, what other method is it to employ? Different lawyers have different ideas as to the techniques that are appropriate to strict and complete legalism, but who would care to suggest an alternative to legalism? A complaint that a judgment is literalistic is one that I can understand, and with which, on occasions I may agree. But what exactly is the meaning of a complaint that a judgment is legalistic? Judges are appointed to interpret and apply the values inherent in the law. Within the limits of the legal method, they may disagree about those values. But they have no light to throw off the constraints of legal methodology. In particular - they have no right to base their decisions as to the validity of legislation upon their personal approval or disapproval of the policy of the legislation. When they do so, they forfeit their legitimacy.Footnote 50
Gleeson J seems aware here of the potential objection — that while it is very difficult to deny that judges should only apply legal method, this does not tell us what legal method is. The same could be said of Dixon J, who did not clearly articulate what he understood legalism to mean. Why is a method that prioritises legal text and structure the only legitimate one?
It could be argued that limits on legitimate method flow from the very nature of adjudication. Fuller famously presented an argument along those lines: he argued that attempting to resolve polycentric disputes or create rules of an essentially legislative nature was incompatible with the nature and purpose of adjudication, and moreover something that courts were universally ill-equipped to do.Footnote 51 I say more about the issue of judicial competence below. But more recent scholarship provides us with other and potentially better ways of understanding the link that Australian judges claim between legalism and legitimacy. Theunis Roux’s thesis, broadly put, is that judicial legitimacy is a social construct. There is no one right or wrong way for judge to perform their function. Rather, judges are understood to act in an appropriately judge-like way when they act in the way that has come to be understood as appropriately judge-like in the legal system in question; more particularly, when they act in accordance with ‘prevailing conceptions of the appropriate relationship between law and politics’.Footnote 52 Australia is presented as an exemplar of a legal system where judicial legitimacy depends upon courts applying a body of law that is autonomous from politics. The more specific rules and principles of constitutional law that I described above could all be fairly described as outworkings of that conception of the law/politics divide; the result of the Court’s sustained attempts to articulate a methodology that looks distinctly legal. Indeed I have stressed that seeing the various interpretive rules and principles in this light allows us to make sense of the case law far better than labels like ‘textualism’ or ‘formalism’.
The apparent fact that the Constitution cannot be amended without great difficulty does not seem to alter any of this. Roux explains that ‘absent an exogenous shock, the continued political acceptance of judicial decisions premised on the relevant reasoning methods sustains the public’s faith in law’s autonomy.’Footnote 53 Widespread realisation that section 128 does not provide a meaningful avenue for constitutional change might cause judicial methods that were previously seen as illegitimate to be recognised as such. But that is just conjecture. It might instead galvanise the people and their representatives to make better use of section 128. If the Court were to leave legalism behind it would then have to acquire legitimacy on some other basis — for example, by showing that it can make interpretive decisions that amend the Constitution well. It is not clear whether the Court could do so, or whether a legal system in which judges have to win and hold the respect of the people in this way is at all preferable to the status quo.Footnote 54
A third way in which legalism has been justified relates to the institutional capacities of the Court. It is sometimes said that courts are ill-equipped to decide contentious political issues. But as Dixon J warned long ago, claims like this must be taken with a grain of salt in the constitutional context, for the Constitution is ‘a political instrument’, and nearly every decision made about its meaning and effect will have profound political implications.Footnote 55 But there remains a distinction to be drawn between deciding what the Constitution means, and deciding that its meaning ought to change. The Court is duty bound to interpret and give effect to the Constitution, no matter how politically contentious its provisions may be, but it has no clear duty to update the Constitution in order to improve it — and it would be ill-equipped to assess whether it would be desirable to do so, let alone whether that is “what the people want”.
Justice Gageler emphasised this point — albeit it in dissent — in Love v Commonwealth.Footnote 56 His Honour held that the power to make laws with respect to ‘aliens’ could not be read to implicitly exclude indigenous Australians, describing that as a ‘supra-constitutional innovation’.Footnote 57 One reason offered for this stance was the inability of the Court to properly assess the implications of such a constitutional change:
The limits of judicial competence are reinforced by the limits of judicial process. The hearing of the special cases in these proceedings has been conducted at a time when a national conversation is occurring about the appropriateness of amending the Constitution to include an Aboriginal and Torres Strait Islander “Voice” to the Commonwealth Parliament. Noticeably absent from the viewpoints represented at the hearing has been the viewpoint of any Aboriginal or Torres Strait Islander body representing any of the more than 700,000 citizens of Australia who identify as Aboriginal or Torres Strait Islander. On the basis of the case as presented, I cannot presume that the political and societal ramifications of translating a communal, spiritual connection with the land and waters within the territorial limits of the Commonwealth of Australia into a legislatively ineradicable individual connection with the polity of the Commonwealth of Australia are able to be judicially appreciated.Footnote 58
This is a claim about good decision-making and the relative strengths of the various branches of government. But it is obviously intertwined with ideas of judicial legitimacy and the broader health of the legal system. Theorists sometimes link this to the idea of ‘democratic debilitation’:Footnote 59 that is, that if important political questions come to be decided by the courts rather than the people and their representatives, the people will be less engaged in the political process, to the detriment of democracy as a whole.Footnote 60 But it may also produce the opposite result, sometimes called ‘democratic backlash’, in which a people enraged by the fact that important political questions have been decided without them rails against the Court,Footnote 61 while those questions become all the more polarising. The question of whether abortion should be constitutionally protected in the United States seems a clear example of the latter.Footnote 62 In this way, the competency claim can be understood as a claim about democracy, and the ways that legalism serves it, though only indirectly so.
The fourth way in which legalism has been justified by the Court, and its observers, appeals directly to democracy and the power of the people to change the Constitution — and for this reason, it is most implicated by the fact that this power is not being used. The idea is that judges do not merely lack the power to amend the Constitution, but that that power has been deliberately entrusted to someone else: the people, and their elected representatives. As Callinan J put it in Singh v Commonwealth:
Judges should in my opinion be especially vigilant to recognise and eschew what is in substance a constitutional change under a false rubric of a perceived change in the meaning of a word, or an expression used in the Constitution. That power, to effect a Constitutional change, resides exclusively in the Australian people pursuant to s 128 of the Constitution and is not to be usurped by either the courts or the Parliament.Footnote 63
His Honour reiterated that point in New South Wales v Commonwealth, stating that ‘[i]f Parliament cannot persuade the people to change, it is not for this Court to treat the people’s will as irrelevant by making the change for them. … The Constitution should not be construed to enable the Court to supplant the people’s voice under s[ection] 128 of it.’Footnote 64
This idea, that judicial amendment would ‘flout … the principle of democracy’,Footnote 65 has been more powerfully articulated outside the Court, and particularly in the work of Jeffrey Goldsworthy. His is presented as a defence of ‘originalism’, particularly against the charge that it reflects some strange form of ‘ancestor worship’. But the key point being defended is that judges cannot and should not change the Constitution, including to update its meaning to keep pace with contemporary needs and values. That approach, Goldsworthy argued, 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.Footnote 66
These ideas have real salience in Australia, given the broad structure of our Constitution: the fact that most questions of substance are left to the political process to decide. They speak to the dignity of people to shape their society and in turn, their lives. One cannot deny the power of these ideas, or their influence on Australian jurisprudence. One also cannot deny that they seem to be diminished, or at least cast in a different light, by the fact that section 128 is so seldom used.
III. Testing the case for legalism
Weis questions the link that Goldsworthy claims, between originalism and democracy. While these scholars talk about originalism, rather than the method of the High Court that I have labelled ‘legalism’, the points they debate are still salient here – because their focus is whether an interpretive method that insists the meaning of the Constitution is fixed and cannot be updated by the courts to keep pace with contemporary needs and values serves democracy, or not.
Weis’ primary aim is to show that it is simply not clear whether an interpretive method which forbids judicial updating serves democracy in any meaningful way. For example, it is possible — and several scholars have attempted to show how — that courts could be said to speak for the people, too.Footnote 67 For another, the case is put entirely in the negative: it purports to show ‘why allowing the judiciary to amend the constitution offends popular rule’, but does not demonstrate that ‘permitting constitutional change exclusively through formal amendment procedures can be thought to give effect to popular rule’; indeed, scholars in this camp typically have ‘very little to say about’ this point.Footnote 68
The question of whether and how courts could be made (more) responsive to popular opinion is a thesis in itself. There is a rich body of scholarship on this topic, including nuanced accounts of how a responsive judiciary would approach constitutional review, and how the appointment and education of judges could enhance their responsiveness. Much of this aims to counter or indeed rectify the weakness identified by Gageler J in Love: the fact that courts, as a matter of institutional design, are poorly placed to ascertain the needs and wants of contemporary Australians. But it is very telling that members of the High Court openly state that they do not feel well-placed to make such assessments; moreover, that they do not think that it is their role to do so. That indicates just how foreign the idea of a Court that ‘speaks for the people’ is in Australian legal culture. Our courts are understood to act for the people in faithfully interpreting and applying the Constitution and the laws made by their elected representatives — including, in a way that protects individual liberty against arbitrary or unlawful government action. But the far more ambitious idea of popular representation advanced in some academic literature is an uneasy fit.
What of the contention, that the democratic case against judicial updating is only presented in the negative? Weis is critical here of Goldsworthy’s claim, that it would be undemocratic for the High Court to update the Constitution because the Constitution gives this power to the people. Goldsworthy makes clear that if the Constitution provides a reasonably practicable method of constitutional amendment, his argument might not hold (or at least, might differ).Footnote 69 But Weis argues that practicability is not enough: indeed, it is the wrong criterion, for the fact an amendment procedure is practicable does not mean that it is democratic. ‘[I]f we are to view the requirement that constitutional amendment take place only through a constitution’s formal amendment rule as reflecting a commitment to democracy’, Weis argues, then
what must be shown is how amendment procedures allow the ‘popular will’ (however defined) to emerge both when and how it ought to. Put differently, it must be shown that amendment procedures do not produce too many ‘false negatives’ (producing no amendment when the popular will is in favour of change) or too many ‘false positives’ (producing amendments that are inconsistent with the popular will).Footnote 70
While she does not aim to prove the point, Weis clearly doubts that section 128 passes this test. For one, the referendum mechanism suffers from the same well-known flaws as any ‘pure majoritarian decision rule’:
permanent majorities may not adequately protect the interests of minorities; and temporary majorities may act for short-term self-interested goals rather than the long-term public benefit, particularly when they know that they are unlikely to maintain majority control.Footnote 71
For another, the Australian Constitution is poorly understood by the public, which hinders their ability to assess the need for constitutional change or the merits of any particular proposal.Footnote 72
Let us assume that section 128 is insufficiently democratic for these reasons. Even if this is so, it does not clearly undermine the case against judicial updating. More specifically, it does not clearly undermine the reasons for that stance that have been offered by the High Court. The Court has not made claims about the efficacy of section 128 — merely, emphasised that it is there. This might sound glib. Can the simple fact that the Constitution empowers the people and their representatives to change the Constitution dictate the interpretive method of the High Court, if we also accept that the people are failing to exercise that power in a sufficiently active or informed way? I think that it can, for what the Court seems to be talking about is not democracy in the fulsome sense discussed in academic literature, but the mere giving of space to make choices.Footnote 73 The Constitution ensures that the people can exercise the powers of choice given to them by the Constitution, but does not take positive steps to ensure that they make those choices well (whatever that might mean), or alter its approach if it thinks that they are failing. Take, for example, the case law on the implied freedom of political communication. The Court will not allow for the flow of political information and ideas to be unjustifiably stifled, because if this occurs the people are deprived of the opportunity to choose, in a meaningful sense of that term. But the Court claims no power to ensure that the people make use of that information, and so actually make reasonable and informed choices. The powers vested in the people and their representatives by the Constitution are not conditioned in this way. In these respects, the Constitution takes Dahl’s gamble: that a people left to act autonomously ‘will learn how to act rightly’.Footnote 74
For these reasons, it is still possible to say that judicial updating would be undemocratic, even if we think section 128 is dysfunctional. The claim is not that the interpretive method of the Court, coupled with the amendment procedure in section 128, ensures that the Constitution reflects the will of the people. We cannot even go so far as to say that judicial updating would thwart the popular will, in the sense that the Constitution would end up with content that the people did not want. The point is simply that the Court should not exercise a power that has been given to the people, and make choices that have been left to the people on their behalf. To my mind, there is power in that idea. It respects the agency of the people and assumes that they have the capacity to do right. But I am sure that for some, this will seem too anaemic a conception of democracy, and incapable of justifying the core tenets of constitutional interpretation.
Perhaps then Weis is right, and we would do better to jettison democracy as a criterion for assessing interpretive method altogether. But it would not follow, that the interpretive method of the High Court is normatively unsound. As explained above, the High Court has offered numerous other reasons why it ought not to amend the Constitution: that this would be unlawful; that it is ill-equipped to do so (well); that it would compromise the institutional legitimacy of the Court.Footnote 75 If we accept that there is at least some merit to these claims, as I think we ought to, then it is not enough to simply show that the Constitution has fallen out of step with the needs or values of contemporary Australia. We would need to have reached a point where the Constitution was so out of step with contemporary needs and values, that the harm caused by judicial amendment outweighs the harm of all those ‘false negatives’ (of ‘false positives’, though they do not seem to be the problem in Australia). If the problem of false positives and negatives can be ameliorated without changing the judicial method — through better public education, and better legislative and political processes to support the referendum process — then that would seem to be a preferable solution. That is especially so, given even a Court willing to read the text of the Constitution in light of contemporary needs and values would still (assumedly) be constrained by the possibilities of the text before them. In other words, judicial amendment would seem to be at best a partial solution to the problem of constitutional inertia, and one which could cause significant problems of its own.Footnote 76
Similar challenges confront those theoretical approaches which — if applied in this context — would seem to call for the court to step in and remedy the constitutional inertia. This the central tenet of a new and important school of scholarship – built upon the work of John Hart Ely and now led by Rosalind Dixon – which argues that courts should perform their review function in a ‘responsive’ fashion which aims to rectify, within the limits of the judicial role, the dysfunction of the democratic process.Footnote 77 To be clear, I am not aware of any scholar in this camp who has directly considered the problem of constitutional amendment (or lack thereof) — though Weis has noted that many of the well-known problems that afflict democratic decision-making are particularly pronounced in the context of section 128.Footnote 78 The lack of constitutional literacy makes constitutional change a political hard sell, because the people simply may not care about it. It makes it easy for opponents of constitutional change to persuade the people to vote no — including, by making false or misleading claims about a proposal for change would work. And many of the changes that could be proposed to the Constitution could also be presented as benefiting only certain groups within society, which again makes opposition easy.
Dixon’s theory of responsive judicial review is a sophisticated one, which lends support to many of the concerns that the High Court has raised about attempting to update and improve the Constitution — including the risk of democratic backlash and debilitation. The core idea that courts should strive to preserve democracy is broadly consistent with the jurisprudence of the Court, which as explained above, recognises that legal protection of the democratic minimum core is a legitimate – even vital – part of its role. Within this, the Court clearly recognises the danger that the people will lack the knowledge that they need to properly exercise their democratic rights, and guards against this by striking down laws that disproportionately restrict the flow of political information. But that is very different from suggesting that the Court should shift its method out of concern that the people are nonetheless too ignorant to choose. It also cannot be ignored, that the reason why the Court says it is able to preserve democracy in this (limited, but important) way is textual: the fact that sections 7, 24 and 128 of the Constitution, amongst others, clearly establish a system of representative and responsible government. It is very doubtful there would be any constitutional constraints on Parliament’s ability to restrict or distort the electoral process, if these provisions were not there.
The broader ideas that Dixon raises — including that Courts should strive to not merely preserve but enhance democracy, and that democracy requires the constitutional protection of far more than free and informed elections — are more alien to Australian constitutional law. But for present purposes, it suffices to say that applying this type of approach in order to rectify the inertia of the people, rather than the Parliament, would require a substantial leap. The reasons why legislatures “malfunction” are relatively well understood. The reasons why the people are reluctant to change the Constitution are not. They are difficult to articulate without resorting to paternalism, or compromising the commitment to the autonomy of the people that sits at the heart of our constitutional structure.
IV. Conclusion
It is right to reconsider how the Australian Constitution should be interpreted, given that document has proven all but impossible to amend. That is especially so, given one of the core tenets of Australian legalism — that the Constitution dictates a distinction between constitutional interpretation and constitutional amendment, and denies the High Court power to do the latter — is often justified by reference to the referendum procedure provided for in section 128. I have argued that this interpretive method withstands normative scrutiny. It serves democracy, though only in the thin sense, of leaving certain constitutional choices to the people and their representatives. If we accept that the Court can update the Constitution because the people and their representatives are not exercising these powers in a sufficiently active or informed way, we fundamentally alter the nature of those choices, and our constitutional order: we make them conditional on a judicial assessment of what the people need. This is something that the High Court has made clear it should not do, and I agree, not merely because of the damage it would do to democracy so defined, but because of the damage it would do to the legitimacy of the Court — an institution that has earned an enormous ‘reservoir of goodwill’ by constructing and applying an interpretive method that makes it plausible to say that in Australia, constitutional law is something different from politics.Footnote 79
Acknowledgements
I thank Elisa Arcioni, Lynsey Blayden, Lael Weis, Rosalind Dixon, and Jeffrey Goldsworthy for helpful discussions or feedback on drafts of this article.
Open access
