The research we do is inevitably affected by our life experience. I grew up in what was then Rhodesia, now Zimbabwe, and did my law degree in South Africa, where I began teaching constitutional law. Both countries were in the midst of conflict over their constitutional futures and debate on constitutional change was the norm, not only among politicians but in social life as well, at all levels of society.
I subsequently taught in New Zealand for five years. This was in the wake of the reformist tenure of Prime Minister Sir Geoffrey Palmer, which had seen the enactment of the Constitution Act 1986 and was followed by the enactment of a statutory bill of rights1 and the adoption of proportional representation.2
I moved to Australia in 1997. In contrast to Southern Africa and New Zealand, constitutional debate in Australia—and here I am referring to debate on systemic, fundamental change—has been striking by its absence. So what I am going to do today is to take the licence, which I hope I am allowed as someone who was originally an outsider but who has taught and researched constitutional law in Australia for 19 years, to cast a critical eye over our institutions from the perspective of pure theory, taking an a-historic, blank sheet approach, and asking: if we could re-design the Commonwealth Constitution, how would we do it, and what would we adopt from other jurisdictions? I also approach this task from the position of the academic who has the luxury—in fact I would say the duty—of discussing reforms without regard to how such reforms might be achieved, a question which lies in the province of political actors. I do, however, offer some thoughts on issues of political practicality at the end of this paper.
I am going to discuss reforms in five key areas: parliamentary representation; parliamentary control over the executive; rights protection; federalism; and the republic, including codification of the reserve powers. I conclude with a discussion of the practicalities of reform and of the pressing need to enhance civics education.
The quality of an electoral system must be measured by the extent to which it fulfils its purpose in a democracy, which is surely to produce results that accurately reflect the views of the voting population. One can thus say that an electoral system is democratic to a greater or lesser extent depending on how representative it is. Applying this criterion to the system for elections for the House of Representatives contained in the Commonwealth Electoral Act 1918 one can say that, while falling within the spectrum of democratic systems, it falls far short of giving equal effect to every citizen's vote. It is nothing novel to state that the single-member electorate system is the most distorting available when compared to the range of systems on offer. The key factor in determining how many seats a party obtains is not the number of votes it obtains nationwide, but rather the accident of where voters live relative to electoral boundaries. Furthermore, this arbitrary system (i) always leads to parties receiving a different percentage of seats to that which their percentage share of the nationwide vote entitles them to, (ii) frequently leads to a party winning government without obtaining a majority of votes and (iii) sometimes even leads to a government winning a majority of seats with fewer votes than the major opposition party, as happened in Australia in 1954, 1961, 1969, 1990 and 1998.
So, for example, while 12,930,814 votes were cast in the 2007 election, the outcome was effectively decided by 8,772 voters in 11 electorates3, who would have handed victory to the Coalition instead of Labor if they had given their first preferences to the former—and this in an election after which the allocation of seats in parliament (83 to Labor and 65 to the Coalition) gave the appearance of a Labor landslide. In 2010 the margin was even closer—13,131,667 votes were cast, but had just 2,175 voters in two electorates4 voted for the Coalition instead of Labor, the Coalition would have won power. How can an electoral system possibly be considered representative of voter sentiment when the winning of government depends upon the arbitrary fact of the geographical location of a tiny number of voters?
Another result of systems using single-member electorates is that they inevitably lead to a never-ending transfer of power between two parties, and thus the establishment of a duopoly rather than a democracy. A reflection of popular dissatisfaction with this state of affairs is the fact that an ever-increasing number of voters are expressing their frustration with the major parties by directing their first preference votes to parties other than Labor or the Coalition. In the 2007 election 14.5 per cent of first preference votes went to minor parties or independents5, but this increased to 18.2 per cent in 2010 and to 21 per cent in 2013—and this is despite the fact that a first preference vote cast other than for one of the major parties amounts, in most instances, to no more than a gesture before having to make a reluctant choice between parties that can actually win a seat but with which the voter may have no affinity whatsoever.
I would therefore argue that we should adopt a system of proportional representation and suggest the single transferrable vote (STV) system, with its multi-member electorates, best balances the requirements of overall proportionality and voter control over the identity of their representatives. This system has the advantage of already being used in the ACT and Tasmania.6 It is also used in countries such as Ireland and Malta. The key determinant of how representative the results produced by this system are is how many members are allocated to each electorate. A comparative analysis of election results from jurisdictions using STV indicates that one can state with a high degree of confidence that, if we had a system where each electorate returned at least seven members to parliament7, the possibility of a government coming to power with a minority of votes would be negligible.8 If this system were adopted, constitutional amendment would be required, as proportionality would be compromised unless the boundaries of the multi-member electorates could be drawn without regard to state boundaries, which would currently fall foul of s. 29 of the Constitution. I would also recommend that the size of the House of Representatives be increased, both in order to keep the new electorates to manageable size and in order to reduce the ratio between voters and their elected representatives, which is currently significantly higher in Australia than is the case in comparable democracies.9
Of course, any proportional representation system would almost inevitably lead to coalition government, but the argument that coalition governments are inherently unstable is not supported by research evaluating government stability under different electoral systems across a wide range of jurisdictions10 and is, in any event, a pragmatic argument, not a principled one, and should not trump the fundamental principle that each voter's views should, as far as is reasonably practicable, have an effect upon the composition of the legislature.
Parliamentary control over the executive
Although in theory the doctrine of responsible government applies in Australia, the system is barely functional in so far as the ability of the opposition to scrutinise the executive is concerned. This is because there is nothing that either house of parliament can do to force the executive to provide the information necessary for that scrutiny.
This was revealed most starkly in 2002, when former Minister for Defence Peter Reith refused to appear before the Senate committee investigating the Children Overboard affair, and the cabinet also ordered that his staffers not comply with the committee's requests to attend.11 At the time, the Coalition lacked a majority in the Senate, which meant that Labor, in conjunction with the minor parties, had sufficient numbers to compel Reith's attendance, and could have used their majority to initiate contempt proceedings against him. However, despite the fact that the Australian Democrats and Greens supported such steps, Labor refrained from using its Senate votes to exercise the contempt powers.12 The most that ever happens when ministers refuse to provide evidence to committees is that they are subject to a motion of censure, and both major political blocs are careful when in opposition not to initiate contempt proceedings leading to significant penalties, such as suspension from parliament, a fine or imprisonment, that could be used against them once they are back in power.13 This provides yet another example of the negative consequences for the Australian body politic of the Labor-Coalition duopoly.
The most striking recent example of ministerial defiance of legislative oversight occurred in 2013–14 when the then Minister for Immigration and Border Protection, Scott Morrison, refused to answer questions posed by a Senate committee on migration matters.14 Similarly, in February 2016 officials from the Department of Immigration and from Operation Sovereign Borders refused on public interest grounds to answer when a Senate committee asked whether the government had paid people smugglers to return asylum seekers to Indonesia.15 The fundamental problem with claims of public interest immunity is that there is no test—other than the government's own assertion—for determining whether the public interest indeed justifies non-disclosure of information to parliament.
How then is this to be remedied? Clearly constitutional conventions have lost their binding force in Australia and thus it is no longer satisfactory to leave the workings of responsible government to the goodwill of ministers. The answer is therefore to replace these conventional rules with statutory provisions, which would compel executive subordination to legislative oversight, with penalties for non-compliance.16
Obviously provision would have to be made for genuine cases where the national interest militated against public disclosure—but this would not mean allowing the executive to claim immunity from providing information merely on its own assertion that the public interest requires it. Rather what is required is a set of rules under which (i) the default position is that there is a legal, not just political, duty on ministers to answer questions and provide such other evidence as is required by parliamentary committees, (ii) proceedings can be taken in the courts in cases of non-compliance, with an appropriate regime of penalties and (iii) the onus of making out a defence of public interest at those court proceedings, in camera if necessary, is cast upon ministers. It would be critical to the success of such a system that the right to initiate proceedings for non-compliance should vest not only in a house and or its committees as a whole, but should also vest in individual committee members. This would be a radical change from the current position.
Putting executive accountability to the legislature on a legal, rather than a conventional, footing, and making the application of penalties no longer vulnerable to political majorities, would have dramatic consequences for the doctrine of responsible government. The experience of the United States, where the legislative branch has far stronger coercive measures at its disposal to ensure executive compliance with requests for information, is instructive. Long-standing precedent gives Congress the right to obtain information from the executive17, and to have recourse to the courts to enforce subpoenas against members of the administration. This was most famously demonstrated in cases which came before the Supreme Court during the Nixon era.18 More usually, however, the two branches reach a political compromise19, and it is a quite normal feature of the political process for members of the executive, including members of the cabinet, to appear voluntarily before public hearings of congressional committees20, or for information to be provided at a confidential committee hearing.21 The fact that the judicial branch is the ultimate determiner of the degree to which the executive is accountable has not led to the courts being confronted with policy questions that they are incapable of deciding—there is sufficient case law for the courts to engage with in determining whether a claim of executive privilege is valid. It is a matter of supreme irony that the legislative branch in the United States has far greater scrutiny power than is the case under the system of responsible government we have in Australia.
It is a truism to say that the purpose of a constitution is to allocate powers between institutions of the state and to define the powers of the state vis-à-vis the individual. Although our Constitution does the first, it does the second hardly at all, as it grants protection to only five express rights. Yet of course a constitution is the only document capable of protecting the individual from legislative power.
The usual justification advanced for the absence of a bill of rights from the Australian Constitution is that Australians prefer to put their trust in democratically elected representatives rather than in the courts. The classic enunciation of this by Robert Menzies was as follows:
There is a basic difference between the American system of government and the system of 'responsible government' which exists both in Great Britain and Australia ... With us, a Minister is not just a nominee of the head of the Government. He is and must be a Member of Parliament, elected as such, and answerable to Members of Parliament at every sitting ... Should a Minister do something that is thought to violate fundamental human freedom he can be promptly brought to account in Parliament.22
Menzies' comments reflect a glib fantasy: as already discussed, the executive is not subject to control by parliament—the strength of the party system and the way the rules of parliamentary privilege operate serve to make the government a virtual elective dictatorship. Furthermore, Menzies' argument, which is still re-stated in various forms by opponents of a bill of rights, ignores the fact that it is parliament itself that poses the principal threat to rights. As Geoffrey Robertson states, a bill of rights:
... means justice for people whose particular plight would never be noticed by parliament, or prove interesting enough to be raised by newspapers or by a constituency MP. Far from undermining democracy by shifting power to unelected judges, it shifts power back to unelected citizens: democracy from its inception has relied on judges ('unelected' precisely so they can be independent of party politics) to protect the rights of citizens against governments that abuse power.23
Robertson's point is important. It is precisely because judges are unelected that the protection of rights should lie in their hands, as the issues they would be charged to determine, which in aggregate boil down to the protection of human dignity, are not ones which should be decided through the interplay of party political forces. Furthermore, few seem to have grasped the inconsistency inherent in the argument based upon democracy: democracy, in the sense of an entitlement to political participation can itself be justified only by reference to an external norm, namely the political equality of individuals and the corollary that each person has a right to participate in the law-making process. In other words, democracy is itself logically subordinate to, and dependant on, the concept of rights.
The absence of comprehensive rights protection from the Australian Constitution is all the more cynical, given that Australia is signatory to all the major human rights conventions—and you will search these documents in vain to find an asterisk directing the reader to a footnote which says, 'These rights do not apply to democracies.' There seems to be an attitude of exceptionalism at play in relation to fundamental rights that puts us at odds with the post-World War II international consensus that emerged in the wake of the Nuremberg trials, which rejected positivism and called for the universal recognition of fundamental rights by all legal systems. Given that our Constitution already grants express protection to five rights and that legislation inconsistent with those rights can be invalidated by the High Court, it cannot be said that the constitutionalisation of the full range of rights we have pledged to uphold internationally would be alien to Australian constitutionalism. Such a step, while expanding the range of rights protected, would certainly not confer any new function on the courts. However, if the existence of justiciable rights is offensive to constitutional principle, then surely opponents of a full bill of rights should be calling for the Constitution to be amended so as to remove such rights as it does protect. Yet calls to remove provisions such as s. 116, which protects freedom of religion, have been conspicuous by their absence, so the question needs to be asked: if freedom of religion is protected, why should that not be so in the case of other fundamental rights?
The absence of a full bill of rights leaves the individual vulnerable in the face of legislation which infringes fundamental freedoms. Let me give just a few examples. It puts Australia in the position where there is no express constitutional right to due process—it being a terrible irony that, in the very week of the 800th anniversary of Magna Carta last year, the principal concern of the government was the drafting of legislation to allow deprivation of citizenship without the need to go to court, the very antithesis of due process promised by article 39 of Magna Carta.24 The absence of constitutional protection of the right to privacy, in the sense of personal autonomy, means that there is no recognition that in relation to intimate personal choices—and here I am thinking specifically of same-sex marriage—the individual should be shielded from the prejudices of parliamentary majorities. Similarly, the fact that there is no constitutional prohibition of cruel and unusual punishment means that there is no limit to the harshness to which asylum seekers may be subjected, either on or offshore.
I cannot leave the issue of human rights without discussing the constitutional recognition of Indigenous people. It is scarcely credible that there are mainstream voices in 21st century Australia who are either openly antagonistic towards the inclusion in the Constitution of a right prohibiting racial discrimination25 or who, while they may support such a right in theory, argue that its incorporation would frighten the conservative horses and thus lead to defeat at a referendum.26 We are left in the truly bizarre position that the Constitution protects the right not to be discriminated against on the grounds of which state one resides in, yet does not offer protection against racist legislation. This is not the time to propitiate conservatives. What is needed is the same moral leadership as was in evidence during the 1967 referendum, which confronts the constitutional conservatives on this issue and overcomes their arguments. We must reject any approach which makes concessions bargaining away the rights of Indigenous people—and even before battle has been properly joined—in order to win conservative support for watered-down reform. Above all, we need to move away from the idea that consensus is the only basis for constitutional change. Sometimes change requires that its opponents be confronted head-on, and their arguments refuted in the public arena. A commitment to non-discrimination is certainly such an occasion.
Seen at its best, the adoption of federalism in preference to unitary government was the necessary price of creating Australia as a nation. At its worst it can be seen as a base compromise pandering to colonial jealousies, which saddled the country with an unnecessarily complex and expensive form of government and, although I hesitate to say it given where I am speaking today, a second chamber which has never performed its designated function as a states' house.
If the federal system is looked at with cold, a-historical objectivity one must conclude that it is difficult to believe that a country with a population the equivalent of a major city in many other countries should have nine governments. The economic cost of federalism is enormous: as long ago as 2002 it was estimated that, at an absolute minimum, the existence of the federal system drained the economy of $40 billion per year27, a figure which would now be much higher. This covers obvious costs such as running state and territory governments, costs to the Commonwealth of interacting with the states and compliance costs to business. It excludes intangible costs in terms of time and inconvenience—think of simple matters such as car registration or trades licensing—experienced by anyone who has moved interstate.
Furthermore, this cost is not balanced by any benefit. It would be idle to pretend that US Supreme Court Justice Louis Brandeis's famous statement that federalism creates circumstances where a 'state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country' operates in any real sense in Australia.28 It cannot be said that Australia presents a vibrant diversity of social dioramas. The other supposed major benefit of federalism, is that it provides protection against tyranny by diffusing power.29 But federalism does not affect what things government may do to individuals, only which government may do them. As I have argued above, only a bill of rights can do that.
De-federalisation would obviously remove a key rationale for the existence of a second chamber. Yet this would not mean a diminution of legislative scrutiny over the executive, because the enhancement of the powers of committee members recommended earlier in this paper would enable members of committees of the House of Representatives to subject the government to more scrutiny than even the Senate can today. Furthermore, the adoption of proportional representation for the House of Representatives would make anomalous the continued existence of an upper chamber elected under an inherently disproportionate allocation of an equal number of senators to each state irrespective of population.
Finally on this topic, there is already a degree of public appetite for the abandonment of federalism. A 2014 survey on public attitudes by the Griffith University Centre for Governance and Public Policy found that 71 per cent of respondents favoured changing the current system (among whom there were differing preferences for the allocation of functions to national, regional and local governments).30 This is consistent with a survey commissioned by the public lobbying group Beyond Federation that same year, which found that 78 per cent of respondents favoured having a single set of laws for the country.31 It therefore seems that de-federalisation is a reform proposal which would be well-received by voters. I leave consideration of this issue by posing the following question: if we were writing the Constitution de novo, would we really create this nine-government system again? And if the answer to that is 'no', then why would we now not abandon it?
The republic and codification of the reserve powers
I have left the issue of a republic until last because, although it is the most frequently discussed constitutional reform, it is in my view the least important. This is not to say that issues of symbolism are without any importance. I remain committed to the view that a severing of the constitutional link between the monarchy and Australia would serve to signal Australia's separate identity on the world stage, and would ensure that there is no office under the Constitution to which Australians may not aspire.
However, of far greater importance than this, in my view, is codification of the conventions regulating the use of the reserve powers, a step which should be taken irrespective of whether we retain the link with the crown or abandon it. This issue is of course linked to that of a republic in so far as significant political capital is made by monarchists out of the supposed risk that an Australian president would abuse the reserve powers by departing from the conventions which govern their use. This problem must therefore be addressed if there is to be any chance of a republic, particularly one involving a popularly elected president, which opinion polls indicate is the preferred model. Yet, to repeat what I said at the outset, codification is necessary even in the absence of a move to a republic. It remains a puzzle as to why, in the wake of the 1975 constitutional crisis, no attempt was made to do this in order to remove uncertainty in relation to the circumstances in which the powers should be exercised.
There is no shortage of examples from the international Commonwealth which could be drawn upon. Several Commonwealth countries have maintained the office of governor-general but have codified the conventions.32 The same is true of others that have become republics with a figurehead president exercising the powers formerly exercised by a governor-general.33 Finally, one can point to Germany and Ireland, republics whose constitutions are based on parliamentary government and contain codified rules almost identical to those which operate by convention in Australia.34 I would therefore argue that codification of the conventions would be beneficial in itself as well as being a necessary corollary of a move to a republic.
Prospects for reform and the need to enhance civics education
Turning finally to the question which I deferred at the start of this paper: what are the prospects for constitutional reform? In answer to this I would make three key points.
First, public opinion in Australia reveals a paradox of extreme conservatism in relation to constitutional change, coupled with disenchantment with, and disengagement from, the political process. Yet there seems to be a failure to recognise that, unless people become accepting of constitutional reform, none of the shortcomings which are the source of disillusionment with the political process can be addressed.
Second, history supposedly shows that successful constitutional amendment requires bipartisan endorsement by Labor and the Coalition. This has a number of invidious consequences: only the most uncontentious amendments—which in reality means those which have the least impact—have a chance of passing at referendum. The perceived need for bipartisan support means that the major political parties enjoy a de facto stranglehold over reform. Furthermore, since the major parties are unlikely to endorse changes that alter the balance of power in the Constitution in a direction that is adverse to their own interests, the capacity they have to derail constitutional reform perpetuates the political status quo. Why do the public allow this to continue, given their disillusionment with the political process in general and the major political parties in particular? Much of the answer to this lies, in my opinion, in the fact that a lack of civics education puts voters at a significant disadvantage when evaluating constitutional reform proposals, making them easy prey for politicians who exploit ignorance about constitutional matters and stoke groundless fears about the effect that constitutional change would have. In my view this means since most of the necessary reforms are antithetical to the interests of the major parties, true reform will happen in spite of them, not because of them, and that the only hope of achieving real reform lies in mass mobilisation of public opinion to an extent which puts the major parties under irresistible pressure to put reform to the people.
Third, it follows from the first two points that the key to constitutional reform lies in harnessing prevailing public disenchantment with the political order to whichever constitutional reform measure has sufficient populist appeal to overcome the voters' notorious suspicion of constitutional change. In my view, a campaign advocating the adoption of proportional representation might have the greatest chance of success. It has the advantage that its case can be based squarely on the concept of fairness and would be able to draw upon rising levels of dissatisfaction with the major parties, who are so obviously and unfairly advantaged by the current electoral system.
Leaving aside this immediate strategy, it is clear that, in the long term, constitutional reform depends upon having a citizenry sufficiently knowledgeable about the current Constitution and its shortcomings to be able to critique it. Here the deficiencies in civics education need to be considered. The Commonwealth syllabus Discovering Democracy35, made available in 1997, and the Civics and Citizenship subject contained in the new Australian Curriculum, published during the period 2011–1336, do a good job at explaining the Constitution as it is, but fail to critique the existing constitutional order. We desperately need a new model of civics education, which enables students to become both informed and critical.
Finally, academic lawyers, who one would normally expect to be bold in their critique of public institutions and innovative in suggesting alternatives but who have in general not done so, also need to discuss broad constitutional reform from the perspective of principle, leaving aside, at least initially, consideration of the politics involved in changing the Constitution. Public resistance to constitutional change is seen as being so ingrained that academic writers rarely venture into this area, presumably believing that anything that is truly significant is doomed to failure. This approach sacrifices principle for pragmatics and ignores the fact that meaningful reform rarely occurs by following public opinion. Radical reform is, by its nature, controversial, and so the role of the advocate must of necessity be that of leading, rather than following. We ought not to be daunted by the apparent difficulty of the task confronting those of us who seek progressive constitutional change in Australia today.