A. Lack of detail

One source of alleged uncertainty was the lack of detail about the structure and operation of the Voice.Footnote ²⁵ The claim was that the proposed model for the Voice — how many members there would be, the appointment process for members, the resources that would be provided to it, the processes it would follow, etc — needed to be released to the public ahead of the referendum so that the people would know what they were voting on. The First Nations Referendum Working Group released a set of design principles prior to the referendum, which contained broad guidance on how the Voice would be structured, for example, stating that members would be selected by Aboriginal and Torres Strait Islander communities, not the executive, and would serve for a fixed period of time to ensure regular accountability to their communities.Footnote ²⁶ However, this document did not reduce the calls for more detail.Footnote ²⁷

The counter-claim was that leaving finalisation of the details of the Voice until after the referendum was the ordinary and logical way of proceeding. The proposed constitutional amendment provided for a two-stage process. The amendment provided for the establishment of the institution. It also provided that the details of the institution would subsequently be determined by the Parliament. That meant the amendment provided for two separate votes — a vote of the people to approve the establishment of the institution and a vote of the Parliament to approve the details of the institution. It also meant the two votes had to occur in that order — the Parliament would not be granted the power to determine the institution’s details until the people had voted to approve the amendment that gave the Parliament that power. This two-stage approach follows how other constitutional institutions operate. In the 1890s, the Australian people did not know the full details of the executive, the Parliament and the High Court when they were voting to approve the proposed constitution.Footnote ²⁸ They were voting on a text that established a set of new Commonwealth institutions and gave the Parliament the power to determine most details of these institutions, which were settled in subsequent votes in the Parliament. The fact that the approach adopted in relation to the Voice resembles what had occurred elsewhere in the Constitution and follows a logical order undercut the suggestion it was uncertain or risky.

However, there was a sharper rejoinder to the call for more detail — that the call itself was an attempt to create uncertainty. The two-stage process exists for a reason. Separating an institution’s establishment from its configuration allows the Parliament to reconfigure the institution as circumstances change. It is an important feature of Australia’s constitutional system of government that the Parliament has been able to alter, for example, the details of the High Court — the number of judges, its jurisdiction, etc — as the demands placed on it have changed.Footnote ²⁹ The same applies to the Voice. Leaving the details to the Parliament would have allowed its configuration and functions to be altered in order to, for instance, improve its effectiveness and accountability. Calling for details of the Voice prior to the referendum helped create uncertainty because any details that were released would not be binding on the Parliament. If, for example, the government had said the Voice would be comprised of 15 members, there was nothing in the constitutional text stopping the Parliament from establishing the Voice with 30 members. It would have created uncertainty by inviting questions about, for example, why were the details announced prior to the referendum if they were not going to be followed by the Parliament? Discussing details prior to the referendum would have risked confusion by talking about issues to be resolved in relation to the second vote — the Parliament’s vote to approve the Voice’s details — during the context of the first vote — the people’s vote to establish the Voice.

B. Justiciability

A second source of alleged uncertainty was the justiciability of the Voice. As mentioned above in Part I, a principle emphasised during the development and design of the proposed amendment was that the Voice be ‘non-justiciable’.Footnote ³⁰ In 2017, the Referendum Council recommended that the Voice ‘would not be justiciable’ in the sense that it ‘would not interfere with parliamentary supremacy’ and ‘the details of its structure and functions would be established by Parliament through legislation that could be altered by Parliament’.Footnote ³¹ In 2018, the Joint Select Committee on Constitutional Recognition relating to Aboriginal and Torres Strait Islander Peoples noted that one theme of the proposed drafts put to it was that the Voice’s ‘structure and functions [should be] non-justiciable’.Footnote ³² In 2021, the Indigenous Voice Co-Design Process ‘affirm[ed] that non-justiciability is an essential feature of the overall design’,Footnote ³³ by which it meant that ‘alignment with the [consultation] standards could not be challenged in court and could not legally affect the validity of legislation or policies.’Footnote ³⁴

The uncertainty came from the fact that, in a strict sense, the amendment would not be — and, indeed, could not be — non-justiciable.Footnote ³⁵ Justiciability is a complex, multi-dimensional concept that implicates a number of constitutional relationships.Footnote ³⁶ During the development and design phases of the Voice, justiciability was primarily used to refer to the relationship between the Parliament and the judiciary. The desire to ensure the Voice would be non-justiciable was a desire to ensure that legislation, including legislation regulating the structure and functions of the Voice, could not be invalidated by a court for failing to heed the Voice’s representations. While the proposed amendment sought to ensure that was the case, it was still possible that this legislation would have been justiciable in some rare cases. For example, if the Parliament enacted legislation constituting the Voice in a way that could not possibly meet the description of an ‘Aboriginal and Torres Strait Islander Voice’ (eg it was comprised of a single non-Indigenous person), it was possible the courts may entertain a challenge to the legislation on this basis.Footnote ³⁷ There was, therefore, some uncertainty about this type of justiciability.

More significantly, this understanding of non-justiciability did not touch on the relationship between the executive and the judiciary. In respect of that relationship, it was possible to make a much stronger claim that the Voice would be justiciable. If the Parliament enacted legislation granting the executive the power to, for example, recognise and protect Indigenous heritage and indicated that the Voice’s representations on the subject were relevant to the executive’s powers on that subject, the Voice’s representations would be justiciable. It would be possible to bring an administrative law challenge in court if the executive failed to take the Voice’s representations into account on the basis that they had failed to take account of a relevant consideration. Some arguments went further, suggesting that, as executive decisions can be reviewed on administrative law grounds that are inferred rather than expressly set out in legislation, there would be decades of litigation to determine when and to what extent the Voice’s representations needed to be considered by the executive.Footnote ³⁸ Put alternatively, not only would the amendment be justiciable, it would create many justiciable controversies because there would be uncertainty over its administrative law implications.

Proponents of the Voice had to concede that, in relation to administrative law, the Voice would be justiciable.Footnote ³⁹ The counter-claim thus had to focus on the degree of uncertainty created by this type of justiciability.Footnote ⁴⁰ It stressed that the relationship between the Voice and executive decision-making would be determined in accordance with well-established principles of administrative law. Furthermore, the principles of administrative law state that, when determining when the executive would need to consider the Voice’s representations, the statute that conferred power on the executive to make the decision assumes prime importance.Footnote ⁴¹ Therefore, when assessing the degree of uncertainty that would be created, it needed to be remembered that any justiciability would be determined in accordance with longstanding legal principles and those principles give primacy to the statute that confers power on the executive. That meant the degree of uncertainty or risk that would be constitutionally entrenched would not be as significant as some supposed — even if an aspect of the Voice’s representations were held to be justiciable, it could be adjusted by the Parliament through the ordinary lawmaking process by adjusting the powers conferred on the executive.

C. The High Court

The High Court was considered by some people to be an important source of risk with the proposed amendment.Footnote ⁴² The argument was that the Court has been known to depart from the text of the Constitution in ways that expand limits on executive and legislative power. The Court’s implied freedom of political communication case law and its decision in Love; Thoms Footnote ⁴³ were cited as examples of this move.Footnote ⁴⁴ The risk, on this argument, was that the Court might construe the proposed amendment in ways that appear to be contrary to the intent or original understanding of the provision, for example, interpreting it to require the executive or even the Parliament to consult the Voice before taking action on Indigenous issues. As a result, the proposed amendment could bring about a change to Australia’s system of government that was more far reaching and more restrictive of executive and legislative power than was originally envisaged.

One counter-claim was to challenge the premise of the argument. The risk identified in the argument presumes the High Court has a history of departing from the text of the Constitution. In the abovementioned examples, it could be argued that the text is not definitive one way or the other. It is possible that ss 7 and 24 could be read as requiring freedom of political communication to ensure that the words ‘directly chosen by the people’ mean the choice is a genuine one and that certain prerequisites are required for it to be genuine. It is possible that s 51(xix) could be read as meaning that an ‘alien’ cannot mean someone who is indigenous to the country. As a result, the view that the High Court has a history of departing from the text relies on a particular interpretation of that text — that the text should be interpreted as not requiring a freedom of political communication or an understanding of ‘alien’ that excludes Indigenous persons. These interpretations are generally derived from an originalist approach to interpretation.Footnote ⁴⁵ Originalism is only one possible approach — it is not accepted by all lawyers, scholars or judges.Footnote ⁴⁶ As a result, the risk identified in relation to the High Court might not be considered a risk if one does not accept that originalism is the correct approach to constitutional interpretation.

A second counter-claim was to consider the likelihood of the risk eventuating. Even if one accepts arguendo the premise of the argument, there is a separate question of how likely the High Court would be to depart from the text of the proposed amendment in a way that expand limits on executive and legislative power. The Court’s case law provides ample evidence to mount the argument that its tendency has been not to expand limits on executive and legislative power.Footnote ⁴⁷ Even employing the strongest evidence to the contrary (the development of the implied freedom of political communication), the Court has only struck down legislation for incompatibility with the doctrine on a handful of occasions over the three decades it has been in existence.Footnote ⁴⁸ It has, on this view, not been a significant curtailment on the exercise of executive and legislative power.

D. Entrenchment

Finally, the fact that the Voice would be constitutionally entrenched was said to present a risk that, if anything went wrong with the Voice, it would be difficult to rectify. In this regard, entrenchment was an argument that built on and strengthened the preceding three arguments. If issues arose from the lack of detail, justiciability or the High Court interpreting the provision in a problematic manner, they would be more difficult to address because it might require further amendment, and amendment is no small task. This risk was commonly underscored by suggesting that a safer option existed — enacting the Voice as an ordinary statute — and that this option should be tried first before considering constitutional entrenchment.Footnote ⁴⁹

The first response to this argument was to emphasise how little of the Voice would be entrenched — only its establishment and its power to make representations to the executive and the Parliament. All other aspects of the Voice would be determined by ordinary statute. Indeed, it was not even clear if the existence of the Voice would be constitutionally entrenched. There was some precedent to suggest that the Parliament might be able to abolish the Voice at some point in the future. The Inter-State Commission was also established by the Constitution and later abolished by the Parliament.Footnote ⁵⁰ As no court challenge was brought following its abolition, there is no High Court decision on whether its abolition was constitutionally valid or not. At the very least it shows that it is possible to envisage a scenario whereby even the Voice’s existence is not entrenched — where it is abolished by the Parliament decades into the future and, due to a lack of constituency in support of its continuation, no challenge is brought to that act of abolition. Furthermore, even if formal abolition were not constitutionally permissible, the executive and the Parliament could significantly curtail its activities through, for example, altering the amount of funding it was granted or reducing the amount of administrative support it received. In other words, entrenchment would do very little to constrain the ability of the political process to control the Voice’s fate.

The second response to this argument was to emphasise that the vote of the Australian people should be respected. The Labor Party went to the 2022 federal election with a campaign pledge that, ‘[i]f elected, we will move quickly on a referendum to enshrine a Voice to Parliament in our Constitution.’Footnote ⁵¹ As the Labor Party was able to form government following the 2022 federal election, the pursuit of this change by means of constitutional amendment rather than statute was in line with the vote of the Australian people. Furthermore, the constitutional — as opposed to statutory — model also demonstrated respect for the Australian people by giving them the opportunity to have a direct say on their constitutional system of government.

A. Form versus substance

Uncertainty and risk can be an argument against the substance of a proposed constitutional change as well as an argument against its form. For instance, if a system of government did not have a bill of rights, one might oppose the introduction of one on the basis of its substance, for example, that it would create greater uncertainty about the people’s rights and duties because bills of rights typically have language that is ambiguous and open to disagreement about its meaning in precise circumstances. However, one could also support the introduction of a bill of rights, but oppose the particular form that it would take. For example, one could argue against the introduction of a statutory bill of rights, but support a constitutional bill of rights on the basis that there is greater uncertainty as to whether a statutory model would be effective because it risks running up against existing constitutional doctrines that limit the ability to bring about constitutional change by means of statute.Footnote ⁵²

While the distinction between substance and form will not always be clear cut, it is important to make given that they touch on different concerns. Uncertainty and risk as an argument against the substance of a particular constitutional proposal relates to the merits of that proposal, can be offset by speaking to benefits of the proposal, and might be ameliorated through alterations to specific features of the proposal. By contrast, uncertainty and risk as an argument against the form of a proposed change relates to specific features of the constitution, cannot be offset by speaking to the merits of the proposal, and is less likely to be ameliorated by altering the proposal.

This distinction was an important feature of the debate about the Voice. Many people invoked uncertainty and risk as an argument against the decision to establish the Voice by means of constitutional amendment. They stated their concern was not with the proposal’s substance (ie a body comprised of Aboriginal and Torres Strait Islander peoples that made representations to the executive and the Parliament). Indeed, many of these people claimed to support this type of reform.Footnote ⁵³ Instead, their concern was with the form the proposal would take (ie that it would be established through an amendment to the Constitution rather than an ordinary statute). As a result, their concern could not be addressed by speaking to the merits of, or proposing to modify, the substance of the proposal — indeed, they purported to agree with the substance.

The use of uncertainty and risk as an argument against the form rather than substance of a proposal has a number of implications. First, it is a portable argument that can be deployed against a wide range of possible constitutional amendments.Footnote ⁵⁴ As it is not tied to the substance of the proposal, it has broad applicability. Even a proposed amendment that purports to have no substantive effect, such as a modification to the Preamble to the Constitution, could be challenged on this basis. The argument would be that the proposal’s form is objectionable because, if the High Court used the Preamble to interpret a constitutional provision in a problematic way, it would be difficult to reverse due to it being entrenched in the Constitution — it would be safer to place the preambular language in an ordinary statute so that it can be changed from time to time.

Second, it can be used to avoid engagement with the substance of the proposal. For someone who is opposed to the substance of a proposal but does not want to publicly debate the substance, they can frame their opposition in terms of its form. They can say, for example, that their concern with the proposal is merely that it will be constitutionally entrenched and, therefore, difficult to remedy if anything goes wrong. That can obfuscate debate — is opposition to the proposal in fact grounded in the form of the proposal, its substance or a combination of the two? The low rate of constitutional amendment in Australia means it is difficult to assess this matter by comparing the stance of people across multiple amendments (ie does someone raise the form of the proposal as a point of opposition in relation to some, but not other, proposals?).

Uncertainty and risk are a particularly attractive way of making this argument given they can be difficult to evaluate and even more difficult to resolve. In terms of evaluation, uncertainty and risk are almost impossible to estimate with precision in the constitutional arena. It is simply not possible to say that a particular constitutional amendment has, say, a 10% chance of producing a particular (or any) negative outcome that will be entrenched against subsequent revision. In terms of resolution, they are almost impossible to resolve except for abandonment of the amendment — if the problem with the proposal is that it will take the form of a constitutional amendment, the only solution is not to amend the constitution. These aspects of uncertainty and risk arguably help explain why opponents of the Voice stressed them even though the substance of the proposal did not, as discussed above in Part II, self-evidently invite criticism from the perspective of uncertainty or risk.

B. Two arguments rather than one

While the terms uncertainty and risk were used interchangeably and without much precision in the wider public debate about the Voice, it is more accurate to suggest that two different arguments were being put forward.Footnote ⁵⁵ In its most persuasive form, the argument about uncertainty was a concern about the unknown. It referred to the possibility that new circumstances and new developments would take the amendment in a direction that was not or could not have been foreseen by those who sought, drafted, and approved the amendment. As a result, one might end up with a state of affairs that, if shown to the people involved with securing the amendment, they might oppose. By contrast, in its most persuasive form, the argument about risk was a concern about the untrustworthy. It referred to the possibility that institutional actors would take the amendment in a direction that would be contrary to the will of those who sought, drafted, and approved the amendment. As a result, one might end up with a state of affairs that the people involved with securing the amendment would absolutely oppose.

Once understood in this way, it is possible to see that the two arguments were pointing to different points of potential concern with the constitutional system of government. Uncertainty is a concern about entrenchment. The argument is that constitutional amendment introduces a form of uncertainty that is difficult to address. No approach to constitutional change exists that is entirely free from uncertainty. Constitutional change that occurs by means of statuteFootnote ⁵⁶ or judicial decision can also produce uncertainty — about the change’s interaction with other parts of the system of government, the consequences for different arms of government, the implications for the rights and liabilities of the people, and so on. Even a decision not to engage in constitutional change can produce uncertainty — about how the existing system of government will react to new developments such as an unforeseen economic or political shock or an unanswered demand for constitutional reform from the people.

In its most persuasive form, therefore, the argument against constitutional amendment is not that it produces uncertainty because every form of constitutional change produces uncertainty (including the decision to engage in no change), but instead that it produces a form of uncertainty that cannot easily be addressed. If, for example, an entrenched constitution is amended to add protection for a new right and subsequent events give rise to uncertainty about the scope of that right, the legislature’s ability to address that uncertainty through subsequent action is going to be more limited than if that right had been added by means of a statute or common law judicial decision.

By contrast, risk is a concern about the reliability of institutions. The argument is that constitutional amendment involves a change to the constitutional text and every time that occurs there is a risk that an institution will take the altered text in a direction that is different from what the people involved in the amendment sought to achieve. For example, if a constitution is amended to establish a new body, there is a risk that the executive or legislature will not take steps to provide it with adequate funding, preventing it from operating to any meaningful extent. In this scenario, the executive or legislature has proven to be unreliable or untrustworthy — they are frustrating what the people approved at a referendum. The ‘rogue’ institution might also be the judiciary. For example, if a constitution is amended to include protection for a new right subject to carefully negotiated limits, there is a risk that the judiciary will interpret it in a way that ignores those limits, taking it far beyond what the drafters agreed to.

Risk is also connected to constitutional entrenchment. An institution that enjoys constitutional entrenchment will pose a greater risk to a system of government if it goes ‘rogue’ than an institution that can be curtailed or abolished by statute. However, uncertainty and risk speak to very different concerns. Uncertainty speaks to the dangers of a design choice — to have an entrenched constitution — while risk speaks to the dangers of human agency — that institutional actors will not act in the public interest.

With the distinction between uncertainty and risk in mind, it becomes apparent that each argument is built on a central premise that will be the subject of disagreement within a jurisdiction and across jurisdictions. The next two sections explore these premises.

C. The importance of constitutional entrenchment

Uncertainty is a contingent argument against constitutional amendment. The argument’s relevance and strength will vary according to the extent to which a constitution is entrenched. One way of conceptualising entrenchment is to map a constitution along a spectrum from highly flexible to highly rigid.Footnote ⁵⁷ The former refers to a constitution where amendment is little to no more difficult than invocation of the regular lawmaking process. The latter refers to a constitution where amendment is or is almost impossible.

At the highly flexible end of the spectrum, uncertainty has less strength as an argument because constitutional amendment is not significantly more constraining on future generations than other forms of lawmaking. Revision of an amendment to deal with an unforeseen development is no more difficult than dealing with uncertainty arising from the common law or statute.

At the highly rigid end of the spectrum, uncertainty diminishes in relevance. If constitutional amendment is or is almost impossible, there will be little reason to consider the possible consequences of amending the constitution, both because serious proposals for amendment will rarely arise and because any serious proposal is likely to fail, rendering discussion of the consequences otiose.

Towards the middle of the spectrum is where the argument has most relevance and strength because constitutional amendment is sufficiently difficult such that the consequences are serious — it will prove challenging to revise the constitution again if significant problems arise from an amendment — yet constitutional amendment is sufficiently available such that it is a realistic possibility, even if it is far from easy to accomplish, and therefore the costs and benefits of amendment warrant careful consideration.

It is, therefore, possible to see why the argument resonated so strongly in Australia. The country is arguably located somewhere towards the middle of the spectrum.Footnote ⁵⁸ No one would contend that the Constitution sits at the highly flexible end of the spectrum, having been amended only eight times since 1901. While some might contend that it sits at the highly rigid end of the spectrum,Footnote ⁵⁹ the Voice itself brings that classification into doubt. The Voice was a serious proposal for amendment that had a realistic prospect for success. Indeed, the debate surrounding the Voice reflected the Constitution’s position towards the middle of the spectrum and the importance of entrenchment to arguments about uncertainty.

The ‘no’ side emphasised the high degree of constitutional entrenchment to make the point that any uncertainty would be unaddressable. The ‘no’ pamphlet, for example, said ‘Putting a Voice in the Constitution means it’s permanent. We will be stuck with negative consequences.’Footnote ⁶⁰ By contrast, the ‘yes’ side emphasised the fact that most aspects of the Voice would be alterable by ordinary legislation and therefore capable of adaptation to new circumstances.Footnote ⁶¹ It also sought to highlight that the Constitution is not as entrenched as the ‘no’ side suggested. As Megan Davis said, ‘[t]he Constitution is meant to change. It is built to change.’Footnote ⁶²

D. The importance of institutional trust

Risk is also a contingent argument against constitutional amendment, but contingent on a different aspect of the system of government. The argument’s strength and relevance will vary according to the extent to which the institutions of government can be trusted to implement the constitution in a responsible manner. If one considers the three arms of government generally to be trustworthy implementers of the constitution, there is less risk that a constitutional amendment will generate problematic results. Conversely, the risk grows the more untrustworthy the arms of government are or are perceived to be.

What amounts to a breach of trust is a large question. There are some actions that most people agree would amount to a breach of trust. Few would dispute that, if the proposed amendment on the Voice had passed and the Parliament refused to enact legislation enabling the Voice to operate, the Parliament would have betrayed the trust of the Australian people who had voted in favour of the amendment.

However, as the degree of divergence between the constitutional text and the constitutional practice diminishes and the time between the enactment of the constitutional amendment and the practice in question increases, more scope for reasonable disagreement emerges. For instance, if the proposed amendment on the Voice had passed and, twenty years later, the Parliament enacted legislation significantly reducing the amount of funding made available to the Voice because the government was facing an economic crisis that required a general reduction in public spending, there would be more room for reasonable disagreement as to whether that amounted to a breach of trust.

The question of trust arises frequently in constitutional theory. It is tied to one’s preferred approach to constitutional interpretation.Footnote ⁶³ For example, an originalist will argue that a court that departs from the original meaning of the text has exceeded its constitutional role and thus breached the trust of those who conferred power on the court, even if the departure is small and even if the text was written centuries ago. By contrast, a functionalist will argue that a court that gives effect to the functions or objectives of the constitution has fulfilled its constitutional role and demonstrated it is a trustworthy steward, even if that involves some departure from the original meaning of the text and even if that departure is from a recently-enacted text.

The question of institutional trust also arises in the debate that occurs in relation to constitutional rights protection. Political constitutionalists and legal constitutionalists disagree about which institution can be most trusted with the protection of rights.Footnote ⁶⁴ The former trust the legislature, while the latter trust the judiciary. As this longstanding debate has shown, it is a question that is not capable of objective answer. Legal constitutionalists and political constitutionalists can look at the same history of rights protection over the same period in the same jurisdiction yet reach very different conclusions about which institution has proven to be more trustworthy in the performance of that task.Footnote ⁶⁵ They can also look at the very same institution and arrive at very different conclusions about the characteristics that indicate whether it can be trusted with difficult moral and political questions.Footnote ⁶⁶ For example, in relation to the judiciary, legal constitutionalists focus on the judiciary’s capacity for reason and ability to add a unique perspective to complex rights questions while political constitutionalists focus on the judiciary’s lack of electoral credentials and the narrow range of people it consults prior to decision.Footnote ⁶⁷

The point of connection to this article is that the arguments developed by originalists and functionalists as well as legal constitutionalists and political constitutionalists can be deployed in the realm of constitutional amendment. If one adopts the view that the judiciary demonstrates a lack of trustworthiness every time it departs from the original meaning of the constitutional text, that will form an argument against further amendment to the constitution to add more text. It only increases the degree of risk inherent in the system of government by giving the judiciary more opportunities to develop the constitution beyond the confines of the text. If one adopts the view that the judiciary cannot be trusted with rights because judges develop them in expansive and unpredictable ways that lead to ‘government by judges’,Footnote ⁶⁸ the same concern would apply to other constitutional amendments — if they can use a rights provision to expand their power at the expense of the other arms of government, they may also be able to do it with, for example, structural provisions. The converse views lead to the converse conclusion.

The connection between risk and trust was prominently on display during the debate on the Voice. Ian Callinan’s view about the risk posed by the Voice was based on his view that the judiciary was not a trustworthy institution: ‘In modern times, it is an open question whether the US constitutionalist Alexander Hamilton’s assertion that the courts are the least dangerous branch of government holds true.’Footnote ⁶⁹ He used the High Court’s development of the doctrine of legitimate expectations in Teoh Footnote ⁷⁰ as evidence of its propensity to engage in ‘heresy’ in the interpretation of the Constitution. This view was echoed by other people who thought that there was no possible way of drafting the amendment that would ameliorate this risk. Greg Sheridan wrote ‘even if [the proposed amendment] contains words to the effect that it must have no influence on how the Constitution is interpreted, it is virtually impossible to stop activist judges reading things into any part of the Constitution’.Footnote ⁷¹ As mentioned earlier, the Court’s decision in Love; Thoms was commonly invoked to show that the Court could not be trusted with the interpretation of constitutional provisions related to Indigenous affairs.Footnote ⁷² On the other side of the debate, former High Court Justice Kenneth Hayne emphasised the Court’s trustworthiness on these matters.Footnote ⁷³ In response to the suggestion that a litigant might try to invalidate a statute if the Parliament failed to consider the Voice’s representations, he said ‘[t]he courts have always shown great reluctance to interfere in the internal affairs of parliament.’Footnote ⁷⁴ He also emphasised the limited power of the judiciary to interfere with executive decision-making: ‘the most that could happen is that the decision-maker would be told to remake the decision.’Footnote ⁷⁵

A. Constitutional stasis

Uncertainty as an argument against constitutional amendment has a self-reinforcing character, which means that it can engender constitutional stasis. As mentioned above, uncertainty relies on a degree of constitutional entrenchment — the more entrenched a constitution is, the more problematic any uncertainty is because it will be more difficult to address. If the argument succeeds and a proposed amendment fails, the constitution becomes slightly more entrenched, as an empirical matter, and thus uncertainty strengthens as an argument in relation to future proposed amendments. The successful invocation of concern about uncertainty begets more concern about uncertainty.

We can observe this dynamic in relation to the Voice. Uncertainty will have even greater force when the next proposed amendment comes up for debate because the Voice has pushed the Constitution further along the spectrum towards the highly rigid category. Put in numerical terms, the amendment success rate is now 8 out of 45 rather than 8 out of 44. Even more importantly, it means the last successful amendment was over 45 years ago — if the Voice had succeeded, that counter would have been reset to 0 and thus dramatically reduced perceptions of the Constitution’s level of entrenchment. This defeat means that the claim ‘any change to the Constitution will be permanent’ will have even more plausibility next time.

At a certain level of entrenchment, uncertainty will diminish in relevance, but that will only be because the Constitution has come to be considered so difficult to amend that amendment is no longer considered a viable means of achieving constitutional change. Importantly, however, the demand for constitutional change never disappears. A constitutional system of government always faces pressure to change from new circumstances and from new demands made by the people. An ossified constitutional text instead means those demands are channelled towards other avenues of constitutional change, such as judicial development of the constitution and the enactment of constitutional statutes, which bring their own challenges.Footnote ⁷⁶

B. Institutional distrust

Risk as an argument against constitutional amendment feeds off, and exacerbates, distrust of government. It taps into the idea that the institutions of government serve themselves rather than the people, especially in the context of the people having a direct say on what is and is not included in the constitutional text by means of a vote at a referendum. The institutions of government and the people are put in opposition to each other, with the people ratifying the constitutional text and the institutions of government acting inconsistently with that text.

We can once again observe this dynamic in relation to the Voice. Among opponents of the Voice, it was common to argue that the High Court could not be trusted with the Voice because it is populated with ‘activist’ judges.Footnote ⁷⁷ The language of activism was carefully chosen to draw from the campaign against the Court that occurred in the mid-1990s in relation to its decisions on Indigenous land claims, but had largely fallen into abeyance since that period.Footnote ⁷⁸ The debate about the Voice provided an opportunity for it to rise to prominence again. The Voice served as a platform for those who think segments of the judiciary are untrustworthy to spread that message throughout the community.

The question of how to speak about public institutions is a complex one. It is unquestionably the case that public institutions should be subject to criticism, and it is possible to argue that even unfair criticism is justifiable in some circumstances.Footnote ⁷⁹ Yet it is also the case that, as the extensive literature on democratic decline indicates, some level of public trust in the institutions of government is necessary for the health of a democratic polity.Footnote ⁸⁰ If the people believe institutions are not acting in the public interest, even if this belief is not well founded in fact, a number of dangers arise in the system of government, ranging from public support for anti-democratic figures through to threats of violence against officials.

It is worth noting that, in a survey taken after the referendum, the judiciary was the only institution that experienced a substantial decline in public trust over the course of the referendum campaign.Footnote ⁸¹ Although it cannot be proven that the arguments about activism from the judiciary were the cause for this decline in public trust, it does indicate that Australia is not immune from declines in public trust of institutions and, therefore, the dangers associated with it.

C. The positive dimensions of uncertainty and risk

Finally, framing uncertainty and risk as an argument against constitutional amendment diminishes if not outright denies their positive dimensions. Uncertainty and risk are what make constitutional law possible. A constitution cannot answer every question about its operation ahead of time. And every constitution requires implementation by a variety of institutional actors. That means it is not possible to have a constitution free from uncertainty and risk. Importantly, these aspects of constitutional law are a feature, not a bug. The silences and ambiguities in the text give the constitution the flexibility to respond to unforeseen circumstances and alleviate the burden of needing to resolve every issue at the point of drafting. In relation to the latter feature, Ros Dixon and Tom Ginsburg have noted that deferral is a crucial element of the constitutional lawmaking process that makes agreement possible.Footnote ⁸²

Furthermore, leaving the constitution in the hands of institutional actors to implement not only risks abuse of power, but also opens the possibility of improvement. Their implementation of the constitution might improve the system of government by reducing the impact of errors made by the drafters and correcting mistakes made by previous generations of institutional actors. The risk that comes with entrusting the constitution to institutional actors must, therefore, also be seen as an opportunity for adaptation, the resolution of ambiguity, the addition of detail, and the addressment of problems. There is value in the flexibility that constitutions, including entrenched ones, offer.

Again, it was possible to see this dynamic in the debate around the Voice. The uncertainty and risk associated with the proposed amendment was framed purely in negative terms. It was an argument for voting no, but not for voting yes. However, it is possible to view the points of uncertainty and risk with the Voice in positive terms. For example, the uncertainty over the relationship between administrative law and the Voice (eg could the Voice’s recommendations become a mandatory consideration in government decision-making?) was a form of flexibility. It was leaving the question open to be resolved by the Parliament in the future if it formed the view that consideration of the Voice’s recommendations would improve government decision-making on Indigenous issues. Similarly, the risk that the High Court might interpret the amendment in unexpected ways was also an avenue for adaptation. If a government sought to develop an innovative way of disabling the Voice from performing its functions, the flexibility offered to the Court would give it some scope to develop new doctrines to ensure the will of the people — to establish a Voice that would make recommendations — was respected.