The Parliament as Partner: A Century of Constitutional Review

Research Paper 3 2000-01

Professor Cheryl Saunders, Consultant
Politics and Public Administration Group
15 August 2000

Vision in Hindsight

Vision in Hindsight is a Department of the Parliamentary Library (DPL) project for the Centenary of Federation.

The Vision in Hindsight: Parliament and the Constitution will be a collection of essays each of which tells the story of how Parliament has fashioned and reworked the intentions of those who crafted the Constitution. The unifying theme is the importance of identifying Parliament's central role in the development of the constitution. In the first stage, essays are being commissioned and will be published, as IRS Research Papers, of which this paper is the seventh.

Stage two will involve the selection of eight to ten of the papers for inclusion in the final volume, to be launched in conjunction with a seminar, in November 2001.

A Steering Committee comprising Professor Geoffrey Lindell (Chair), the Hon. Peter Durack, the Hon. John Bannon and Dr John Uhr assists DPL with the management of the project.

Centenary of Federation logo

Centenary of Federation 1901-2001



Major Issues
Part 1: The Vision
Part 2: Framework for the Referendum
Part 3: Initiation and Review

Part 4: Hindsight

Major Issues

This paper is one of a series written to mark the centenary of the Australian Constitution in 2001. From the outset, the prescribed procedure for altering the Constitution has required that a bill setting out the proposed changes first be passed by the Parliament, before being submitted to electors voting in a referendum. In 2001 Australia will have had 100 years experience of a still relatively unusual form of partnership between the institutions of representative and responsible government and the electors, voting directly. The paper examines this experience, in the light of the expectations of the framers of the Constitution. It suggests reasons for the obvious and growing gap between the views of Parliament and voters on constitutional change. It identifies ways in which the process of altering the Constitution might be assisted to operate in a way that is more satisfactory.

The paper is timely for reasons more pressing than the commemoration of the centenary. At the turn of the century, a stalemate has been reached in relation to changing the Australian Constitution:

  • of 44 proposals accepted by the Parliament and put to referendum, only eight have been approved, two of which were minor, and
  • there is some evidence that the difference of view between Parliament and people is becoming more marked. One of the two most recent referendums, proposing the establishment of a republic, failed despite apparently high levels of public support for a republic in principle (although the actual views of the Parliament were unclear in this case). The previous four proposals, put to referendum in 1988, failed with an historically high 'no' vote.

While people will have different opinions about the merits of particular referendum proposals, on any view the record of rejection suggests a waste of energy and money. It also contributes to a defeatist attitude towards the prospect of constitutional change, which prevents serious consideration of constitutional change as an option for dealing with significant national problems. By way of example, little serious consideration has been given to changing the Constitution to deal with the current uncertainty about the validity of the Corporations Law; and the Corporations Law itself is the product of an earlier failure to seek constitutional change.

Historical purpose

The paper shows that:

  • the framers of the Constitution adopted the referendum for progressive, democratic reasons, although some details of section 128 were influenced by familiar tensions between central and state power
  • the framers thought that the requirement for the Parliament to initiate proposals to be put to referendum would both provide an important filter and assist public understanding of the proposals, and
  • the framers expected that the Australian Constitution would be less rigid than the Constitution of the United States.

Reasons for the referendum record

The paper suggests that the general explanation for the high rate of rejection of referendum proposals is the failure of successive Parliaments and governments adequately to adapt practices developed in the context of representative government to the quite different demands of the referendum. This manifests itself in:

  • the highly adversarial character of most debate on constitutional change
  • the lack of importance that has been attached to an understanding of the Constitution on the part of people born in Australia or those migrating to the country
  • the lack of an accepted process for public consultation on constitutional issues, and
  • the inadequacy of the procedures for informing voters about particular proposals for change at the time of a referendum.

To overcome the perceived obstacles to constitutional change, often analysed by reference to 'bipartisanship', four general constitutional reviews have been conducted since federation, at intervals of roughly 25 years. The paper shows that each of these reviews, in its own way, also was subject to a degree of political intervention that seriously impeded its effectiveness. A fifth process, of a different kind, the Constitutional Convention of 1999, attracted widespread public interest and enthusiasm at the time. Even the Convention was flawed, however, for the purpose for which it was established. The ambiguity of its relationship with both the Parliament and the voters ultimately was reflected in the quality and acceptability of the proposal that was put to referendum.

They said it

Some lighter moments in the paper are supplied by quotes from the past. For example:

  • Samuel Griffith in 1891 feared that Australians would not be able to come to grips with the detail of a proposal to change the Constitution although 'if the question were to be simply a kingdom, or a republic, there might be a plebiscite on that'
  • Prime Minister Fisher, introducing the statutory requirement for distribution of the 'yes/no' cases in 1911, predicted 'no doubt at all that the cases will be put from both sides impersonally and free from any suggestion of bias', and
  • one problem with the 1999 Convention was anticipated by Playford in 1891: ' this mode of convention you can never ascertain correctly the views of the people. You only ascertain the views of the men who have been elected members of the convention.'

Future Directions

Broad options for the future, which are cumulative and not mutually exclusive, include the following:

  • recognise and accept that approval by referendum requires a different approach to government
  • take a longer-term view of constitutional issues within the Parliament and minimise unnecessary partisanship
  • find practical measures to make it clear that discussion of the Constitution is natural, important and need not be divisive. This might be done by, for example, establishing a joint standing committee of the Parliament, charged with making an annual or, at least, regular, report on the Constitution in accordance with agreed terms of reference
  • hold referendums in conjunction with elections unless there is a compelling reason not to do so, in the interests of minimising cost and controversy over cost.
  • Ascertain and take into account the views of voters at the time a proposal for change is being developed. The manner in which this is done is likely to depend on the nature of the proposal
  • provide more sophisticated and effective procedures for helping voters to understand proposals for change in the approach to a referendum. An independent body might be given the responsibility to design and execute an information program, with a right of review to ensure that the program is balanced and fair, and
  • give priority to enabling and encouraging Australians to understand and engage in discussion on the Constitution and system of government.

Some of these options may appear utopian, given Australia's robust political culture. In reality, however, there is little other choice. There is no prospect of removing the referendum requirement from the Constitution. Nor would it be desirable to do so, even if it were possible: the trend towards greater involvement of people in major public decisions, which the framers observed in the 1890s, is even more marked 100 years later, not only in Australia but elsewhere in the world. If the Parliament is to remain the sole filter for referendum proposals it is necessary to make the partnership work, in order to deal with national problems for which constitutional change is the best or only solution.


The Australian Constitution can be altered only by a procedure that involves both the Parliament and the people voting directly in a referendum, in accordance with section 128.(1)

In normal circumstances, a proposed law to alter the Constitution must be passed by both the House of Representatives and the Senate, with absolute majorities. There is also, however, a hard-won deadlock procedure, whereby a proposed law may be put to referendum if passed by one House alone. The procedure requires one House to pass the bill twice, with absolute majorities, and for an interval of three months to elapse between the second passage of the bill and the date on which the other House first indicated its disagreement.

The Constitution is not altered until a proposed law, passed by the Parliament, also is approved at referendum. All electors qualified to vote for the House of Representatives in each State and, since 1977, in most territories,(2) are entitled to vote in a referendum. Two majorities are required to signify approval: a majority of all electors voting and a majority of electors voting in a majority of States. Where both Houses pass a proposed law, section 128 specifies that the referendum must be held within two to six months. On the face of the section, a referendum apparently is mandatory once a proposed law has passed both Houses.(3) The section does not confer responsibility on anyone in particular, however. By contrast, where the deadlock procedure applies, the Constitution does not specify a time limit for the referendum. Instead, it confers on the Governor-General authority to submit the proposed law to referendum. .(4) This authority appears to be discretionary but in fact is consistent with polite drafting style used elsewhere in the Constitution in relation to the Crown.(5) Once approved at referendum, a bill must receive the Queen's assent from the Governor-General before the alteration takes effect.

Section 128 also requires a majority of electors in an affected State or States to approve a proposed law in certain cases. This additional requirement applies to alterations that would diminish the proportionate representation of a State in the Parliament or its minimum number of Members of the House of Representatives or would alter the limits of a State or affect the provisions of the Constitution relating to State limits.(6)

An odd caveat in the third paragraph, long since overtaken by events, enables only one half of the votes in a referendum to be counted in any State with adult suffrage, until the qualifications of electors for the House of Representatives are uniform. The explanation lies in the unusually early extension of the right to vote to women in South Australia, still the only Australian colony to have taken this step as the Constitution drafting process reached its final stages, in 1897-98. As an interim measure, the draft Constitution also provided that State laws on the qualification of voters would apply in Commonwealth elections until the Commonwealth Parliament passed its own electoral law.(7) The requirement to halve the votes in States with adult suffrage was included to prevent such States receiving what was perceived to be an undue advantage in a referendum.(8) It became redundant when the Commonwealth Franchise Act was passed in 1902, conferring the vote on both women and men, with the now notorious exception of 'Aboriginal natives'.(9)

The Australian Constitution was drafted at a time of speculation about the use of the referendum around the world and some experimentation with it. The form of the referendum adopted by Australia was passive, leaving the initiative with the Parliament. Nevertheless, the inclusion in the Australian Constitution of a requirement for change to be approved by the voters was unusual and, at first blush, surprising. The Australian colonies were familiar with Westminster-style representative government, and the new federal Constitution incorporated these principles as well. The addition of a referendum inevitably raised questions about the role of the Parliament, already complicated by the more familiar tension between Parliament and the executive, under a system of representative and responsible government.

Part 2 of this paper examines the reasons why the framers of the Australian Constitution chose the Constitution alteration process now prescribed in section 128, and their expectations of it. Some of their expectations related to particular aspects of the process and were specific. It is possible also to glean an impression of the prevailing expectations on more general matters. These include the role of the Commonwealth Parliament and of the Houses individually, the issues that might arise in relation to use of the referendum and the degree of difficulty of altering the Constitution by this means. The remainder of the paper focuses on the way in which the procedure has worked in practice, with particular reference to the Parliament. Section 128 confers two broad functions on the Parliament. First, Parliament provides the regulatory framework for a referendum vote.(10) This is the subject matter of the third part of this paper. Secondly, only the Parliament can initiate proposals to alter the Constitution. This is considered in the part that follows.

A Constitution alteration procedure does not operate in a vacuum. It is affected by political culture, the rest of the political process, the nature of the Constitution and the substantive constitutional issues that arise from time to time. In Australia the procedure has been further affected by continuing divisions between the major political parties in relation to the Constitution, some of which may be inherent,(11) albeit aggravated by the procedure itself. The paper cannot deal with this wider context in detail, but notes its relevance.

The record of 100 years reveals substantial activity by the Parliament in relation to alteration of the Constitution. It has led to relatively little change, however, and in that sense has been unproductive.. Since federation, 44 proposed changes have been put to referendum and eight have been approved. Two of these were minor.(12) Three others made limited but useful alterations.(13) Only three can be seen, with hindsight, to have been substantial.(14) In relation to the Aborigines referendum, at least, its significance may not have been foreseen either by the Parliament or by the voters at the time.(15) While views inevitably differ on the merits of the proposals that have not been approved, the referendum record represents a failure, if only in its waste of money and time. Even more seriously, the experience has created a defeatist national attitude towards constitutional review and the possibility of constitutional change.

Many explanations are possible for the way in which section 128 has worked in practice. Those relating directly to the Parliament and to relations between the Parliament and the people, at referendum, are explored towards the end of the paper. This part suggests that the parliamentary system has not adapted itself sufficiently to the demands of partnership in a decision-making process that includes the referendum. The outcome of a parliamentary vote usually can be predicted and in any event is controlled by the respective parties. The dynamics of a referendum are completely different. Proposals are easily misrepresented, particularly when, as at present, the public is reputed to know relatively little about the Constitution. A referendum is a handy vehicle for a protest vote against the government or politicians generally. Leadership from the Parliament and a bipartisan approach are necessary but probably not sufficient for a referendum to succeed. Constructive use of the referendum in Australia is likely also to require the Parliament to respond better to the views of voters on constitutional issues, using improved mechanisms to explain the purpose and detail of proposals being put to the vote.

Australia is not alone in confronting these challenges. The referendum is now law or practice in many countries for constitutional change(16) Governments throughout the world are under pressure to give the public greater direct participation in decision-making,(17) particularly in the making and changing of constitutions.(18) Technically, moreover, direct public participation is feasible. In these circumstances, greater use of the referendum, in countries around the world, is likely in the first part of the 21st century. Public and political understanding of how and when referendums should be used gradually will increase as well. Australia's long experience of the constitutional referendum means that it has much to contribute to these issues. That same experience, however, suggests that it also has much to learn.

Part 1: The Vision

The detail of the procedure for altering the Australian Constitution was hammered out across the entire decade of the 1890s. Alfred Deakin raised the issue at the 1890 conference in Melbourne that initiated the later federation movement.(19) The draft of the clause that became section 128 was not finally settled until the Premiers' Conference in 1899. Section 128 is one of the relatively few sections of the Constitution to have been altered since federation by referendum, with the right to vote being extended to electors of the territories in 1977.

Debate on this part of the Constitution, as on so many others, was influenced by two immediate preoccupations of the delegates to the founding Conventions, preoccupations that created divisions between them. One was the appropriate balance between a national or truly 'federal' approach to decision-making and the interests of the individual States. The other was the potential for deadlocks between upper and lower Houses in a bicameral Parliament, of which most delegates had had practical experience in the Australian colonial Parliaments. In a sense, the version of the Constitution alteration procedure that ultimately prevailed represented a tactical victory for particular groups on each of these two issues. The tenor of much of the debate would justify claims that the outcome was progressive and innovative in terms of contemporary experience and democratic theory, however. There also were signs of the cynicism about the effectiveness of representation that has since become more overt and that led in part to the decision to adopt the referendum. Nevertheless, that cynicism did not appear to cause the framers seriously to doubt that the partnership between Parliament and people in this novel form of decision-making ultimately would work out.

It was assumed from the start that, unlike the position in Canada, the Australian Constitution would provide for its own alteration.(20) In fact, the resolutions introduced by Parkes to initiate debate on the Constitution in the National Australasian Convention in 1891 did not provide for alteration.(21) Subsequent debate, however, accepted that the reason for the omission was to leave the decision about the alteration mechanism in the relevant Convention committee,(22) rather than to avoid an alteration procedure altogether.(23) On that basis, Thynne withdrew his motion for a new resolution

that a system be established for submitting amendments of the Constitution for the approval of the electors of the several states, and for prescribing the necessary majorities'.(24)

The Convention accepted a less contentious resolution enabling each State to amend its Constitution to the extent necessary to achieve federation.(25)

There was also general agreement that the Australian Constitution would be entrenched through a special alteration procedure, reflecting the significance of the 'bargain' between the colonies to which the Constitution would give effect.(26) There was no consensus, however, about what that procedure should be. Despite Thynne's motion, the proposal the Constitutional Committee brought back to the Convention in 1891 adopted the mechanism of elected Conventions, drawing on the procedure for the adoption of the Constitution of the United States, rather than the referendum. The recommendation of the Committee would have required a proposed law to alter the Constitution to be approved by elected Conventions in a majority of States, after passage by an absolute majority in each House of the Commonwealth Parliament.(27) The Convention adopted the proposal, with one important qualification. The draft clause was amended to require that the people of the States whose Conventions approved an alteration must also be a majority of the people of the Commonwealth.(28) The rationale was to mirror the balance between national and State majorities, represented within the Parliament by the House of Representatives and the Senate respectively.(29)

The principle of conventions approving alterations to the Constitution was secured in 1891 over the dissent of a small but influential group of delegates who argued strongly for the referendum.(30) The objection to conventions, a mechanism that organised procedure exclusively around States, was only partly assuaged by the changes made to the Committee's recommendation. Deakin argued with force that, as conventions could do no more than approve or reject a proposal the Parliament had already passed, the question should be put directly to voters.(31) Cockburn, with an eye principally to the still unsettled issue of how the initial Constitution was to be approved, characterised conventions in the United States as 'a barrier against the popular will' that would never be accepted in Australia.(32)

In the different circumstances in which the Australasian Federal Convention met in 1897, supporters of the referendum easily prevailed. The draft clause proposed to the Adelaide session provided for approval by referendum. It retained elements of the 1891 provision, to the extent that it required a proposed law to be approved by electors in a majority of States as long as the people of those States also were a majority of the people of the Commonwealth.(33) The present requirement for the views of the national majority to be ascertained directly was substituted in the course of debate on the clause.(34) Thereafter, the referendum remained a constant feature of the procedure for alteration of the Constitution. During the compulsory adjournment in 1897 the Legislative Council of New South Wales proposed that alterations be approved instead by a 2/3 vote in both Houses of the legislatures of all the States. This proposal was negatived in Melbourne in 1898, without debate.(35)

Debate on the principle of the referendum arose in other contexts, however. It was prompted in Adelaide by Deakin's motion to remove the requirement for the Parliament to pass Constitution alteration bills with absolute majorities, given that 'the question of reform is practically remitted to the people'.(36) The motion was lost.(37) The concept of the referendum was canvassed again in Melbourne, in debate on whether it should be possible to submit to referendum bills on which the two Houses were deadlocked. In both of these debates, an underlying concern can be detected that the Senate would unduly inhibit alteration of the Constitution by preventing proposals from reaching the people.

Debate on the need for a provision to deal with deadlocks over Constitution alteration bills was prompted by a suggestion of the Legislative Assembly of Victoria that proposals should be submitted to the people 'in the case of difference between the two Houses'.(38) Delegates were divided at least in part by their attitudes to the referendum and their views about the ease with which a process of constitutional change should be set in train. There was some by-play that suggests the influence of other issues as well, however. Many of those espousing a procedure for the resolution of deadlocks, in giving example of the manner in which deadlocks might occur referred to a Senate in favour of a particular change and a House against.(39) By and large delegates from small States and upper Houses remained sceptical, however.(40) After a passionate debate, taking almost a whole day, the amendment was rejected 31-14.(41)

The issue was revived again by the Premiers' Conference in 1899, after the deemed failure of the referendum on the draft Constitution the year before. Amongst the changes adopted by the Premiers to increase support for the Constitution in New South Wales was the inclusion in section 128 of a procedure to deal with deadlocks.(42) The new paragraph was modelled on the deadlock procedure for ordinary bills, now in section 57 of the Constitution, but compatible in principle with the motion rejected in 1898. This drafting history helps to explain the differences between this paragraph and the rest of section 128.(43) It is possible that some or all of the Premiers also foresaw that the involvement of the Governor-General effectively would preclude use of the procedure for bills passed by the Senate, but whether they did is unknown.

Many years later, one of the advocates of the clause in 1898, H.B. Higgins, wrote to H.V. Dick:

I know that the thing which drew so many to vote for the 1899 bill who had voted against the 1898 bill was the striking change made by the Premiers in the amending power ... So important was this change in our eyes, that I did not open a campaign in Victoria against the 1899 bill ... [which] I foresaw ... would certainly be carried.(44)

One further aspect of the section was not finally settled until 1899. From 1891 there had been concern to protect the delicate compromise over the proportions of State representation in the Commonwealth Parliament from alteration by majorities of any kind, without the consent of an affected State.(45) Out of caution, this special protection was extended in 1891 to the minimum entitlement of each State to five seats in the House of Representatives as well.(46) Both qualifications on the general procedure for Constitution alteration survived the substitution of the referendum for State conventions in 1897. In 1899, in response to concerns about protection for State boundaries in New South Wales, alterations to the limits of a State 'or in any manner affecting the provisions of the Constitution in relation thereto' were added to the list of matters that required a majority in the affected State.(47)

The apparently uncharacteristic decision of Australians, at the end of the 19th century, to adopt the referendum for the purposes of constitutional change is readily explicable in context.

In the first place, the drafting of the Australian Constitution coincided with early use of the referendum elsewhere in the world, and burgeoning debate on its use. Approval at referendum had been required for alterations to the Constitution of Switzerland since 1848.(48) The popular initiative had been introduced for partial revision of the Swiss Constitution in 1891. Delegates to the Australian Conventions were familiar with the Swiss procedure(49)and, in general, favourably inclined towards it.(50) They were aware of the trend towards the adoption of the referendum in the American States(51) and concerned about the difficulty of amendment of the Constitution of the United States, where the referendum was not in use. Even in the United Kingdom, bastion of responsible government, a debate was under way on the merits of the referendum, with A.V. Dicey urging consideration of the 'possible expediency of introducing [it] into England' for constitutional purposes,(52) and James Bryce advising caution.(53)

Some delegates also argued specifically for the principle of the referendum, as the next logical step in the evolution of democracy. Isaacs put the point most forcefully in 1898. He said that since the passage of the Reform Act 1832:

... there has been a gradual but a sure shifting of power, from the Parliament to the people ... This is a tendency which we cannot resist. Every day we see the right of the people to be consulted upon their own affairs becoming more and more acknowledged. Who will stand here or anywhere and deny that right?(54)

There may be an echo here of contemporaneous arguments in England, although these tended to centre on the effect of party and faction on the ability of Parliament to make decisions on the merits.(55)

In any event, the adoption of the referendum for the purposes of ongoing constitutional change was a logical choice for a Constitution that would be approved by referendum in the first place. Even in 1891, there was some expectation that the Constitution would be approved by popular vote,(56) although clearly this was not shared by all.(57) Thynne's advocacy of the referendum as a mechanism likely to excite the enthusiasm of the people for the Constitution probably was regarded as misplaced by those who assumed that the colonial Parliaments would approve the draft.(58) Uncertainty on this point aided the decision in 1891 that Constitution alteration would require approval by State conventions. By 1897, however, the need for the draft Constitution to be approved by the electors was laid down in the enabling Acts of four of the five participating colonies. Ultimately, a similar requirement was prescribed for Queensland and Western Australia as well. After 1897 it was more readily assumed that the referendum also should be used for ongoing change although Glynn, a consistent critic, pointed out that this did not necessarily follow.(59)

Nevertheless, there was significant opposition to the mechanism of the referendum, both in 1891 and 1897-98. Some of the criticisms were practical. Describing the referendum as 'full of trouble and difficulty',(60) Gillies asked in 1891, for example, whether proposals would be put separately or combined in a single bill 'so that electors can say yes or no to each?' In 1897, McMillan referred to the potential of the referendum to 'dislocate the whole of the social and business life of the community'' as a reason for insisting on the requirement of an absolute majority in the Parliament before a proposal was submitted to the people.(61) In the next year Howe, another confirmed opponent, pointed to the probable expense of the referendum in a 'wide continent like Australia'.(62)

For the most part, however, critics pointed to two other, general issues, each of which is familiar today. The first concerned the capacity of the voters to contribute to decisions about the future of the Constitution by voting in a referendum. One delegate expressed a fear that voters 'may be swayed by a sudden gust of passion' to alter 'a Constitution which may have been framed with great care'.(63) Another expressed concern that voters would not be able to come to grips with the detail of a proposal to change the Constitution, although 'if the question were to be simply a kingdom, or a republic, there might be a plebiscite on that'.(64) Several delegates suggested that the people as a whole would not understand proposals for change,(65) and that in the absence of Convention delegates, there would be nobody to enlighten them.(66) Deakin suggested that Members of Parliament would do so. Baker was sceptical:

If they do then, I say their nature will be different from the nature of ordinary members of Parliament.(67)

The second general issue was the perceived inconsistency of the referendum with two of the principal pillars of the Constitution: representative government and responsible government. There is an echo here of the more familiar concern of the framers of the Constitution about the compatibility of responsible government and federalism. The potential for conflict between the referendum and representative government was obvious. The representative principle involved the delegation of power by the people 'to men selected by them as specialists'; the referendum involved repudiation of 'the wisdom of centuries'.(68) The objections to the referendum on grounds of responsible government were more diffuse. In part they were rhetorical, a convenient ground on which to distinguish Australia from Switzerland.(69)

The issue was discussed at length in the debate on deadlocks, however. The demands of responsible government were invoked often in opposition to an arrangement that would allow a referendum to be held following passage of a Constitution alteration bill by only one House. If the Senate was the House that passed the bill, the government would be obliged to submit the proposal to referendum, despite its own opposition to it.(70) Even where a bill was passed by the House and rejected by the Senate, Barton envisaged the potential of the referendum to 'eat away the foundations of responsible government' by giving the government a means of appealing directly to the people as a distraction from its difficulties in the Parliament.(71) Delegates did not perceive so clearly that the problem might be the reverse; that the influence of responsible government on the initiation of referendums might politicise the entire Constitution alteration process and make it more difficult to use.

Despite the recent Swiss adoption of the popular initiative, there was general agreement that the Constitution should provide only for the passive referendum, leaving the initiative with the Parliament. This meant that the referendum would be a 'veto, nothing more'(72) or 'simply a ratification'.(73) It would be for Members of Parliament to deliberate on proposals for change, refining them and taking responsibility for them.(74) Much of the debate on these issues in both 1891 and 1897-8 was directed to ensuring that the procedure for initiating proposals was sufficiently serious without unduly restricting 'the very narrow road, by which it will be possible for the electors to obtain an amendment of the Constitution'.(75)

The framers of the Constitution wanted a mechanism for Constitution alteration that would strike what they considered to be the appropriate balance between protection of the Constitution and flexibility to change it as need arose. Inevitably, there were different views about what this meant.(76) Nevertheless, delegates referred repeatedly to the problem of the rigidity of the Constitution of the United States, which they were anxious to avoid.(77) They seemed to think that they had done so. Their principal reservations about the mechanism they had chosen concerned the manner in which proposals for change would be initiated, rather than the referendum itself.

Glynn was an exception. Towards the end of the debate on the issue in Melbourne, he moved to add an alternative alteration procedure, involving passage by special majorities in the Parliament in two successive sessions.(78) His motion was negatived on the voices.

Part 2: Framework for the Referendum

Section 128 provides some constraints on the referendum process. The electors must vote on the proposed law itself. By contrast, in a plebiscite, voters may be presented with a range of options and invited to indicate their order of preference for them. The law must be put to referendum, if it is put at all, within two to six months after its passage by both Houses. Everyone who is qualified to vote for Members of the House of Representatives is entitled to vote in a referendum. A referendum is passed if majorities of those who vote approve the proposed law. An informal vote therefore effectively is a no vote. Otherwise, the section confers on the Parliament the responsibility of prescribing the referendum procedure and of dealing with any practical difficulties that arise.

Parliament exercised its power under section 128 for the first time in 1906. The Referendum (Constitution Alteration) Act 1906(79) was passed in preparation for the 1906 referendum.(80) It was replaced in 1984, by the Referendum (Machinery Provisions) Act,(81) as part of a major overhaul of the electoral legislation. The approach of both Acts was to model referendum procedure as nearly as possible on the familiar arrangements for the election of Members of Parliament or the passage of ordinary legislation. The referendum process begins with the issue of a writ by the Governor-General(82) acting, as usual, on government advice. A referendum must be held on a Saturday. There must not be a State or territory election on that day, although there may be a federal election.(83) Rules broadly comparable to those for elections are provided for modes of voting, scrutiny and disputed returns.(84) The Electoral Commission administers the ballot. The question for voters on the ballot paper refers to the long title of the proposed law.(85) In a necessary departure from ordinary voting procedure, voters must write 'yes' or 'no' in a space provided.(86)

This approach meant that, when compulsory voting was adopted for Commonwealth elections in 1924, it was extended to referendums as well.(87) Section 45 of the 1984 Act now imposes a 'duty' on every elector 'to record his vote at a referendum'. The change had no immediate effect on referendum results. By 1924, only two out of 13 proposals had been approved at referendum. The pattern that was to become the norm was well set, and three out of the next 13 proposals were approved. The effect of compulsory voting on the turnout for referendums was substantial, however, as was to be expected. Before 1924, turnout ranged between 50.2 per cent in 1906 and 73.7 per cent for the five referendums on Commonwealth powers in 1913. Since 1924, turnout consistently has been more than 90 per cent, although often not above 95 per cent.(88) There is an argument that compulsory voting is less appropriate in referendums than in ordinary elections because of the complexity of the issues and the degree of apparent voter apathy.(89) After 75 years, however, compulsory voting is well established. The voluntary vote for delegates to the 1998 Constitutional Convention may be one reason why the Convention failed to give voters a sense of ownership of its recommendations.

Some practical issues for referendum procedures, anticipated in the 1890s and enlivened by subsequent experience, are not resolved by legislation but are dealt with pragmatically, case by case.

The first is the question whether and in what circumstances proposed changes should be put to the voters in a single referendum bill.(90) While in some cases disparate proposals have been combined,(91) the objections to the practice are obvious, and a uniform lack of success does not encourage its indiscriminate use. An associated question, less easy to answer, is whether more than one bill should be put to referendum at the same time. It is sometimes argued that multiple questions are confusing to electors, who cannot discriminate between them. In practice, however, there have been only five occasions on which a single question has been put.(92) While two of the five proposals were approved, the sample is too small and the examples too old to allow useful conclusions to be drawn. Concern about multiple referendum proposals may be reinforced by the similarity of the results on all bills put to voters in this way in several referendums.(93) Significant differences in results in other cases, however, suggest that an influential proportion of the electorate, at least, is well able to distinguish between different proposals, if it chooses to do so.(94)

A second important procedural issue concerns the timing of referendums and, in particular, whether they should be held separately from, or in conjunction with, a federal election. The answer is significant for cost, which is substantially greater if a referendum is held alone. Writing during his term as Chief Electoral Commissioner, Colin Hughes estimated, for example, that the 1988 referendums would have cost $10 million rather than $34.4 million had they been held at the time of a general election.(95) There is, of course, a danger that a referendum held with a general election will become confused with 'personal issues, with political issues, with platform issues' as Deakin feared in connection with the approval of the draft Constitution.(96) The approval record of proposed laws submitted to referendum with a general election(97) is roughly the same as that for those submitted separately, however.(98) And the increasingly partisan nature of referendum campaigns, now generally conducted separately from elections, raises a question whether obfuscation could be any worse. Colin Hughes has speculated:

[While] a separate referendum might be less contaminated by partisanship ... hostility to the extra public expenditure and the likelihood of scare campaigns being heard in the quieter environment could cancel that advantage.(99)

Public understanding of referendum proposals and the manner in which the public is informed about them have been continuing practical problems which were anticipated by some of the framers.(100) The requirement for a referendum to be held within two to six months was included in the Constitution partly for this purpose, lest 'the question ... be allowed to get cold and the educative effect of the debates be lost'.(101) Higgins supported an amendment to the deadlock procedure to require one House to pass a proposed law twice to ensure that 'the people ... have the advantage of a long discussion before they commit themselves.' He continued: 'I want them to have the advantage of what appears in the newspapers and of what takes place in the Parliament, and on the platform.'(102)

The problem anticipated by the framers has been exacerbated since by several factors: the introduction of compulsory voting, the disappearance of systematic civics education from schools, the lack of information provided to immigrants about the Constitution and system of government. Whatever the cause, the result is notorious. Repeated inquiries have concluded that a surprisingly large proportion of Australians are unfamiliar with the fundamentals of the Constitution or, apparently, unaware that it exists at all.(103) To the extent that these results are accurate, they have significant implications for the referendum process and for the nature of the information the electorate needs before a referendum is held.

With a few minor variations, identified below, the response to this problem has been much the same for the past 90 years. The proposed law itself is publicly available.(104) The only other information made accessible through official channels sets out the arguments in favour of and against the proposed law, each limited to 2000 words. This material is posted to electors by the Electoral Commission at least 14 days before the referendum, with a statement of the textual changes.(105) The arguments are prepared by Members who voted respectively for and against the proposal in the Parliament. If there is no opposition to a proposal in the Parliament, as in the case of the Aborigines referendum in 1967, there is no provision for distribution of arguments against it. Distribution of the cases for and against is an extension of the expectations of many of the framers that the debate in the Parliament would be the principal vehicle for informing voters about referendum proposals. The highly adversarial and tactical nature of the cases, however, now raises a question about the extent to which they fulfil their intended purpose.(106)

The referendum legislation makes the usual provision against deception or undue influence in connection with referendums, including a prohibition against misleading or deceptive publications.(107) Dismissal of a challenge under this section during the 1999 referendum campaign points to the difficulty of invoking such a section where competing democratic values are at stake: in this case, the right of electors to understand a referendum campaign and the right to freedom of speech. The absence of an effective sanction against misleading statements would be less significant if there were a source from which reliable information could be sought.

There have been four distinct phases in the provision by the Parliament of information about referendum proposals in Australia:(108) 1901-1911; 1912-1983; 1894-1998; 1999.

Three referendums were held during the first 11 years of federation, involving five propositions. Two were approved. There was no additional provision for public funding of information of any kind during this period. The referendums of 1906 and 1910 were held in conjunction with a general election and information was provided in the normal course of those campaigns.

The rejection of the referendum bills in 1911 was attributed to misunderstanding of the proposals. The government moved to amend the referendum legislation to provide for the distribution of information to voters, in the form of arguments for and against referendum bills. According to Hughes, then Attorney-General, the arguments would be 'appeals to the reason rather than to the emotions and party sentiments'. Prime Minister Fisher expressed:

... no doubt at all that the case will be put from both sides impersonally and free from any suggestion of bias of misleading on the one side or the other.

He concluded with a rhetorical flourish:

Let it be a document that Parliament will be proud of and from which Australia will benefit.(109)

With the support of the Opposition, led by Deakin, the new requirement was inserted into the 1906 Act, as section 6A. The 1911 proposals were put to referendum again, in altered form, and rejected again, although on this occasion they attracted greater support.(110).

The operation of section 6A was suspended for the next three referendums, all of which were put by non-Labor governments. In 1919, the referendum was held in conjunction with an early election and it was argued that there was no time for the cases to be prepared and distributed. In 1926, despite the introduction of compulsory voting, the requirement was waived again, in the face of substantial differences of opinion between those supporting the proposals. In 1928, the referendum to authorise the Financial Agreement was held in conjunction with a general election and supported by both parties. Once more, the view was taken that no additional information was needed. The arguments were distributed again for the referendum of 1937, and there have been no waivers since.

Increasing frustration with the results of referendums and mounting criticism of the information provided under section 6A ushered in the next phase in the provision of public information. Five Constitution alteration bills passed the Parliament in 1983.(111) The then Labor government proposed to spend $1 million on a promotional campaign, in addition to $5 million on the usual referendum pamphlet.(112) In reaction, the Senate caused to be added to the referendum legislation a prohibition against the expenditure of public money other than as authorised by the Act.(113) The following year, in debate on the Referendum (Machinery Provisions) Bill, the government sought authority to fund promotion of both sides of the case, in proportion to support for, and opposition to, the referendum bill in the Parliament.(114) This attempt also failed. An amendment of another kind by Senator Macklin was successful,(115) however, and has potential significance. Its effect was to authorise the expenditure of public moneys on 'the provision by the Electoral Commission of other information relating to, or relating to the effect of, the proposed law'.(116)

Controversy in the Parliament during debate on the Macklin amendment may help to explain why the Commission so far has not provided information of this kind.

The final phase in the evolution of current Australian practice in relation to public information about referendum proposals occurred in 1999, in connection with the republican referendum. That referendum was unique. The proposal emanated from a half-elected Constitutional Convention. Members of the Commonwealth Parliament constituted a minority of the Convention delegates. The Prime Minister was opposed to the substance of the referendum. Many of the most powerful proponents and opponents of the referendum were outside, rather than inside, the Parliament. In these circumstances, section 11(4) was suspended to allow public expenditure on additional programs. One was a plain English public education program, to enable 'balanced, factual information to be made widely available'.(117) The second involved funding two external 'yes' and 'no' committees with $7.5 million each.(118)

The second program was a response to the particular circumstances of the 1999 referendum and is unlikely to be repeated. The first has more general potential, although its tentative nature in 1999 limited the impact it made. The role of the independent committee appointed for the purposes of the program(119) was confined to reviewing, often at short notice, a text produced by the government. The priority of the committee was to ensure that the two opposing sides would accept the material as fair, making the needs of the public a secondary consideration. The committee had no responsibility for the way in which the text would be published or used, or for the preparation of material for other media. The contrast with the relative sophistication of the public information provided by New Zealand in connection with its two plebiscites on the electoral system is marked.

As expected by many of the framers of the Constitution, the Australian Parliament has played a role in the provision of information to the electors about referendum proposals. In the early days, at least, its ideals were lofty, if rhetoric is any guide. In practice, however, party politics and partisan self-interest have dominated the process. There is little focus on the needs of voters. And despite the revolution in communication techniques over the course of the 20th century, the legislative framework for the provision of public information effectively has remained unchanged. The failure of the Parliament to develop a more effective regime may reflect a view that impartiality is impossible. This view has some merit, particularly in the atmosphere of suspicion that typically characterises Australian constitutional debate. Other countries have managed to overcome such difficulties, however, through a combination of goodwill, checks and balances and a commitment to the values of public participation. There is no reason why Australia cannot do so as well.

Part 3: Initiation and Review

Under section 128, the responsibility of initiating proposals to alter the Constitution is left to the Parliament.

A proposal is initiated once a Constitution alteration bill is passed by the two Houses, with absolute majorities, or by one House twice, in accordance with the deadlock procedure. Experience shows, however, that this is a necessary but not sufficient prerequisite to submit a bill to referendum. Constitution alteration bills are not put to the voters unless the government of the day so decides. The requirement for the Governor-General to act on advice in exercising the discretion to submit referendum bills that have passed one House twice has been assumed since 1914, effectively to nullify the Senate's power to initiate referendums.(120) Echoing the predictions of some of the framers of the Constitution, the Royal Commission on the Constitution described this aspect of the operation of the section as 'an instance of responsible government prevailing over a federal principle in a Constitution which seeks to combine them'.(121) Even where a Constitution alteration bill has passed both Houses, a referendum will not necessarily follow, as precedents from 1965 and 1983 show.(122)

Responsible government has wider ramifications as well for the initiation of referendums under section 128. Some Constitution alteration bills have been introduced as private members bills.(123) Most proposals are introduced with the support of the government, however, and are unlikely to be passed, at least through the House of Representatives, without such support. With the 1999 referendum on the republic as the sole-and not encouraging-exception, all proposals submitted to referendum in effect represent government policy. The dominance of the executive government over the process for the initiation of alterations to the Constitution has contributed to the partisan character of the debate within the Parliament and during the referendum campaign.

The party political character of the initiation process, in turn, plays a part in the high rate of referendum failure. In recognition of this difficulty, a succession of wide-ranging reviews of the Constitution has been established since federation, essentially as a source of policy advice to government about proposals for constitutional change that might have broader support. Parliament has played a role in most of these, either as active participant, as facilitator, or as both.

These reviews fall into three categories. The first category, that of expert committees, comprises the Royal Commission on the Constitution 1927-29 and the Constitutional Commission 1986-88. Reviews in the second category have involved Members of Parliament more actively. Again there are two: the Joint Committee on the Constitution of the Commonwealth Parliament, 1956-59; and the Australian Constitutional Convention, 1973-85. The hybrid character of the 1998 Constitutional Convention justifies a third category. All five review bodies are examined below. In each case, this paper considers the terms of reference of the body, its establishment and process, its principal recommendations and its outcome, in terms of proposals submitted to referendum and the results. While the ostensible purpose of each review was to provide a broader support base for proposals for change, each also was subject to a degree of executive government control that contributed to limited results.

Review by Experts

Royal Commission on the Constitution

The Royal Commission was appointed on 18 August 1927 by the Governor-General, on the advice of the Bruce government, under the Royal Commissions Act 1902-1912. Its terms of reference required the Commission to report on:

the powers of the Commonwealth under the Constitution and the working of the Constitution since Federation.

Ten subjects were drawn specifically to its attention, including the currently contentious issue of 'industrial powers'. Two of the seven commissioners were members of Parliament. Ominously, neither represented the federal Opposition, which had refused to participate.(124) The Commission took two years to complete its report. Its evidence covered 7000 pages and the report comprised 300 pages of 'detailed and exacting material'.(125)

The Royal Commission coincided with the end of the first quarter century of Australian federation and with the move to the new federal capital in Canberra. It had been preceded by a string of referendum failures. The most recent had occurred in 1926. There had been talk since 1920 of a Constitutional Convention to review the Constitution. A bill to establish a Convention had been introduced into the Parliament but had lapsed after the second reading stage.(126) Initially, the Bruce government had proposed that a select committee of the Parliament review the Constitution.(127) Negotiations with the Opposition broke down, however, over the proportionate representation of the parties on the committee(128) and the Commission was established instead.

Substantial divisions of view between the commissioners complicated their recommendations and their final report. In essence, these differences reflected different attitudes to federalism as a system of government for Australia. A majority supported continuation of the federal system, with some amendment. A minority advocated eventual transition to a unitary system through a procedure to enable alteration of the Constitution by an Act passed by absolute majorities of both Houses of the Commonwealth Parliament. Specific majority recommendations with contemporary political significance included deletion from the Constitution of Commonwealth power with respect to industrial disputes and relaxation of the procedure for creating new States. Other recommendations with continuing relevance over the decades that followed were extension of the maximum term of the House of Representatives to four years, imposition of a retiring age for Justices of the High Court, conferral on the High Court of power to give advisory opinions and power for the States to impose indirect taxes that do not interfere with customs duties.

The Letters Patent appointing the Commission required it to report 'with as little delay as possible'. In March 1928, Members of Parliament began asking in the Parliament when the report would become available.(129) They continued to ask, with little response, for the next 18 months. Eventually, the report was received on 20 September, one week after the Parliament had been dissolved for the general election. One issue in the campaign was the Government's proposal to return authority over industrial disputes to the States. Round Table commented that, since the views of the majority of Commissioners on industrial matters 'harmonised with those of Mr Bruce's policy speech, the heat of political controversy generated an unjustifiable suspicion that there was a certain timeliness in the appearance of the Report.(130)

In the event, Prime Minister Bruce lost both the election and his own seat. Response to the Royal Commission's report fell to the new Prime Minister, Scullin. He was hardly likely, in the circumstances, to be sympathetic to the majority view. A bill to implement the recommendation of the minority on the procedure for alteration of the Constitution was introduced into the Parliament, but failed at the second reading stage in the Senate.(131) No other action was taken in direct response to the report, although some of its recommendations reappeared in later reviews.

Constitutional Commission

The Constitutional Commission was established by the Hawke Labor Government on 19 December 1985 to carry out a 'fundamental review' of the Constitution.(132) There were six full commissioners during the life of the Commission, each with legal training. The Commission was assisted by five advisory committees, dealing respectively with the Australian judicial system, the distribution of legislative powers, executive government, individual and democratic rights, and trade and national economic management. Between them, the advisory committees had another 37 members, with a range of professional backgrounds.(133) The terms of reference of the Commission required it to consider whether the Constitution adequately reflected 'Australia's status as an independent nation', the economic structure of the federation, the division of powers and democratic rights.(134) At the request of the Attorney-General, the Commission presented a first report to the Government on 27 April 1988, on specified matters.(135) Its final report was presented on 30 June 1988.

While the Commission did not recommend a new Constitution, its reports covered almost every aspect of the existing one. The Commission provided a sixfold classification of its own recommendations.(136)

  • provisions that adequately serve their intended purpose,
  • provisions that are outmoded or expended
  • provisions that have not met the purpose they are intended to serve (for example, in 1988, section 117)
  • aspects of the Constitution that are out of step with the economic, social and political needs of Australian life, or the role Australia plays in international affairs (for example, the absence of a federal power over defamation)
  • alterations to the structures of government, suggested as necessary by experience, and
  • new provisions, to repair omissions and to improve the ease with which the Constitution can be understood.

While there was dissent within the Commission on some individual recommendations, for the most part the reports were agreed to by all Commissioners.

The terms of reference of the Commission emphasised to a greater degree than earlier reviews the importance of public participation.(137) Both the Commission and its advisory committees solicited submissions and held public meetings around the country. Under the auspices of the Commission there was a survey of levels of popular understanding , which suggested that only 53.9 per cent of Australians knew that there was a written Constitution and that amongst young people the levels of knowledge were lower still.(138) The Commission distributed free copies of the Constitution, organised translations of some of its material into the principal ethnic and some Aboriginal languages and used a variety of other means to disseminate information to improve understanding about its work. These measures were important as symbolic recognition of the need to engage voters in the process of constitutional review. While undoubtedly they had some practical impact as well, subsequent events suggested that it was marginal and that more sophisticated strategies are needed to encourage more active public participation.

As with the Royal Commission, the circumstances surrounding the establishment of the Commission were inauspicious. The proposal to create a body of this kind was publicly announced by Attorney-General Lionel Bowen in the closing stages of the 1985 session of the Australian Constitutional Convention.(139) He urged delegates to accept that people other than parliamentarians might have 'other knowledge, other concepts and other qualities of leadership that might encourage this nation to talk about a new Constitution for Australia'.(140) His speech signalled the withdrawal of the Commonwealth Government from the Constitutional Convention process. His stand was opposed by non-labor delegates from the Commonwealth and all State Parliaments and by some State Labor delegates as well-those from Victoria, South Australia and Tasmania. Despite Commonwealth Government opposition, a motion was passed to continue the Convention, albeit in a modified form.(141) The Australian Constitutional Convention Council remained in existence until 1994, although crippled by the withdrawal of the Commonwealth and some State delegations.

As a result, despite Bowen's endorsement of the need for 'bipartisanship' in his Convention speech, the Commission had the support neither of the non-Labor parties nor of most States when it was established later that year. As far as the States were concerned, the situation was exacerbated by the terms of reference of the Commission and its advisory committees. These barely touched on the questions of reform of fiscal federalism, on which there had been a large degree of consensus at the Brisbane session of the Convention. As original partners to the federal 'bargain' the position of the States in an expert review of the Constitution is awkward in any event. The Commission reported that 'only the Tasmanian Government and the Queensland Government made comprehensive submissions'.(142)

The divisions between governments and between the parliamentary parties did not bode well for the referendum that followed in 1988. The resentment generated by the circumstances of the establishment of the Commission was further fuelled by signs that the referendum questions were a somewhat cynical exercise on the Government's part. The Commission reported that it had been asked by the Attorney-General in January 1988 to provide the Government with an early report on specified matters by the first week in May. These matters included four-year terms for the House of Representatives, constitutional recognition of local government, and selected civil and democratic rights.(143) The Commission's first report was submitted on 28 April. Four Constitution alteration bills were introduced into the House on 10 May, dealing with parliamentary terms, fair elections, local government, and rights and freedoms. They passed both Houses by 3 June. The Commission noted:

In some cases the Bills are similar in substance and in form to what was recommended in the First Report. In other cases they vary from what we recommended.(144)

The bills were put to referendum on 3 September, against vigorous opposition from the non-Labor parties, and were rejected by historically large majorities. The final report of the Commission was delivered to the Government before the referendum took place, but no subsequent action was taken in relation to it.

Review by Parliaments

Committees of the Commonwealth Parliament have been used from to time to advise on specific constitutional provisions and, occasionally, on the need for Constitution alteration. In particular, committees of both the Senate and the House(145) have recommended change to the provisions of section 44, dealing with the qualifications of Members of Parliament. So far no action has been taken. In addition to targeted inquiries of this kind, the Commonwealth Parliament has also participated directly in two general reviews of the Constitution. These are the subject of this part of the paper.

Joint Committee on Constitutional Review

The Joint Committee was established on a motion of Prime Minister Menzies on 24 May 1956.(146) The terms of reference for the Committee required it to:

review such aspects of the working of the Constitution as the committee considers it can most profitably consider, and to make recommendations for such amendments of the Constitution as the Committee thinks necessary in the light of experience.

Eight Members of the House of Representatives and four Senators were appointed to it. The Prime Minister and the Leader of the Opposition were ex officio members who did not, however, attend. The Prime Minister's motion was welcomed and supported by Dr Evatt, then Leader of the Opposition. The Committee presented its first report to the Parliament on 1 October 1958, before the Parliament was dissolved for the general election. It presented its final report on 26 November 1959.

With the exception of substantial reservations by Senator Wright, most of the recommendations of the Joint Committee were unanimous. They fell into three categories. The first dealt with the composition and function of the Parliament and the relations between the Houses. Issues included the nexus, numbers of Senators and Members, deadlocks, simultaneous elections, electoral distribution and discrimination against the Aboriginal people in section 127 of the Constitution. The second category dealt with extensions to Commonwealth legislative powers, including aviation, research, industrial relations, corporations, trade practices and economic powers. These recommendations also sought relaxation of section 92 of the Constitution for the marketing of primary products. The third, miscellaneous category included a companion recommendation about section 92 in relation to interstate road transport. It also dealt with the creation of new States and proposed a change to the Constitution alteration procedure itself, to require majorities only in half, rather than in a majority, of the States.

On the face of it, the circumstances in which the Committee was established and the cross-party acceptance of the recommendations within the Committee boded well. There were other less auspicious signs, however. The Committee was established following a general election in which the Government had increased its majority in the House of Representatives but had failed to win a majority in the new Senate, now elected on a basis of proportional representation. The Governor-General's speech at the opening of the new Parliament linked the proposal for the establishment of the Committee to this issue.(147) By the time the Committee delivered its final report, in 1959, another election had taken place and the Government had a majority in both the House of Representatives and the Senate. From the standpoint of the Government at least, the immediate urgency for reform had receded.

There were other portents as well. From the start, there was State dissatisfaction with the establishment of the Committee. After criticising the lack of State involvement,(148) the New South Wales Parliament established its own committee in June 1956, with the support of both sides of politics.(149) That committee, when it reported, called for repeal of the uniform income tax legislation, a matter on which the Commonwealth Committee made no recommendation. Other States also reacted negatively. Some State Premiers and Opposition leaders did not respond to the Committee's invitation to confer with it.(150) Premier Bolte of Victoria announced in March 1959 that his Government would not accept the changes proposed by the Committee.(151) In the following year, the Liberal Premiers of South Australia and Western Australia joined him in opposing the Committee's recommendations again.(152)

Nor was the cross-party agreement in the Committee representative of the rest of the Parliament. In his commentary on the Committee's work, Sawer noted:

Balance of opinion in the Committee was probably somewhat to the left of the average opinion of this Parliament, Labor Party views included, and certainly a good deal to the left of the average opinion of the Liberal and Country Party organisations in the country as a whole.(153)

Hostile comment on the Committee was made in the Senate as early as March 1957.(154) The opposition of the non-Labor Senators was reported in October 1958, in response to the Committee's first report.(155) During an Opposition attempt to force debate on the report in 1961, one non-Labor member, Drummond, suggested that he had been 'blocked from speaking'.(156) And Garfield Barwick, who was appointed as Attorney-General in the Menzies Government, on his election in 1958, clearly was not enamoured of the Committee's recommendations. In response to an Opposition motion in 1961, he referred pointedly to the lack of State involvement with the Committee(157) and suggested that the newly established Standing Committee of Attorneys-General would be a preferable mechanism for securing the desired uniformity of company law.(158) By 1962, he was openly critical of the Committee's approach, saying that the Committee should have presented priorities to the Parliament.(159)

Despite repeated attempts to raise the matter in the Parliament, on the part of the Opposition over a period of five years, there was no formal Government response to the Committee's report. Finally, in 1965, now frustrated by the constitutional nexus between the number of Members in the House of Representatives and the Senate, Prime Minister Menzies announced that he had two Constitution alteration proposals in mind: on Aborigines and on the nexus.(160) The bills were passed by the Parliament but not submitted to referendum on that occasion. They were passed in altered form in 1967 and a referendum vote was held. The Aborigines referendum was approved, with very high majorities, but the nexus proposal was defeated. There was no other direct outcome of the Committee's recommendations, although some of its recommendations were taken up in subsequent reviews.

Australian Constitutional Convention

The Australian Constitutional Convention (ACC) comprised delegations of members from the Commonwealth and all State Parliaments, territory Assemblies and local government representatives. The Commonwealth delegation had 16 members and the States had 12 each. Delegations were representative of the parties in the Parliament and of both upper and lower Houses. Each delegation was appointed by a resolution of the Houses from which it was drawn.(161)

The ACC was a Victorian initiative, supported by both sides of politics.(162) A Labor Member of the Legislative Council, Jack Galbally, first suggested a Convention in 1969. R.J. Hamer, later Premier of Victoria and leader of the early Victorian delegations, took up the suggestion in the following year. The immediate interest of both of them was the familiar State preoccupation with fiscal federalism.

The framework for the ACC was negotiated by the Attorneys-General of the participating jurisdictions, initially in the Standing Committee of the Attorneys-General and then as the Convention's Steering Committee. The terms of reference of the Convention were set out in a motion moved by Prime Minister Whitlam and seconded by Premier Hamer at the opening of the first plenary session, on 3 September 1973.(163) The work of the ACC was carried out through a series of Standing Committees and plenary sessions. Six such sessions were held between 1973 and 1985, in 1973, 1975, 1976, 1978, 1983 and 1985. The resolutions of all sessions now are collated in the Guide to the Archives.(164)

The strength of the ACC lay in its potential to develop necessary consensus on proposals for alteration of the Constitution, across all political groups with representation in Australian Parliaments. It involved the non-government parties as equal partners,(165) thereby avoiding the difficulties that had emerged in connection with the Royal Commission and that later would affect the work of the Constitutional Commission. Involvement of the States overcame one of the principal difficulties of the Joint Committee.

On the other hand, there were signs that, in the eyes of the Commonwealth Government and Parliament, these strengths were weaknesses as well. State delegates were numerically superior in the Convention, despite the four additional members allocated to the Commonwealth. The outcome on important issues was unpredictable and required negotiation that was difficult and often frustrating. There were two particular indications of a lack of Commonwealth commitment to the ACC process. One was the number of proposals for change that were put to referendum during the early years of the ACC but did not emanate from it.(166) These complicated the task of developing a culture of co-operation within the Convention and added to the record of referendum failure. The second was some reluctance by the Commonwealth to put ACC recommendations to the vote. Over the 12 active years of the ACC, only five such recommendations were put to referendum, three of which were successful.(167) In 1983, five other Constitution alteration bills, which drew broadly on ACC recommendations, passed the initiation stage. In the event, however, they were not submitted to the people.(168)

The approval of three proposals in 1977 was evidence that the processes of the ACC could lead to alteration of the Constitution in ways that were useful, if not exciting. Ultimately, however, the ACC proved too vulnerable to party political manoeuvres. A successful 1976 session, during which both sides of politics showed remarkable goodwill in the aftermath of the constitutional events of 1975, overcame much of the damage of a non-Labor boycott of the October 1975 session in Melbourne. But a rowdy Adelaide session in 1983, during the height of the dispute over Commonwealth policy towards the Gordon below Franklin dam, assisted to prejudice the newly elected Hawke Government against the ACC and contributed to its withdrawal in 1985. The fracas over the funding of referendum campaigns that followed shortly after, which resulted in the decision not to put to referendum five bills that would have implemented ACC recommendations, soured relations further still. A final straw was the opposition of the non-Labor parties to the referendum on interchange of powers in 1984, which had been endorsed by the ACC in successive sessions, with the support of both sides of politics. The referendum failed, with consequences that are still being felt, in the reluctance of States to refer to the Commonwealth powers in relation to, for example, Corporations Law. The 1985 session of the ACC agreed to a range of measures to try to overcome the practical difficulties with its operation and to ensure that the Convention worked more effectively in future.(169) From the standpoint of the Commonwealth Government, however, it was too late.

The experience of the ACC raised some interesting questions about how a body comprising delegations of parliamentarians can and should operate. One was the extent to which such delegations can operate cohesively, given party divisions within them, in the interests of a constitutional process. While realpolitik eventually prevailed, some delegations displayed a remarkable commitment to the ACC during its active life, transcending party politics in difficult times.(170) A second question concerned the composition of delegations. The rationale for the ACC was to achieve a broad consensus on proposals for constitutional change. For this purpose, it was desirable for delegations to reflect a wide range of views as effectively as possible. It was obvious that motions passed narrowly were unlikely to succeed at referendum. From the outset, the principle that Government and Opposition should be equally represented in delegations was accepted. Towards the end, however, some jurisdictions adjusted their delegations to reflect the balance in their respective Parliaments. While this was insignificant for the purposes of substantive motions, it enabled obstructive procedural motions to be carried and contributed to the gradual deterioration of goodwill.

The effects of the ACC were not confined to constitutional change, but influenced some practical aspects of the Commonwealth Constitution and some changes to State Constitutions.(171) With the establishment of the Constitutional Commission, the ACC ceased to be an active force. The Constitutional Commission drew on some of the ACC's recommendations, however, apparently treating them as further submissions.(172)

The Constitutional Convention 1998

A Constitutional Convention was held in February 1998, over a two-week period, to consider questions connected with the proposed republic. The issues for the Convention, as specified by the Prime Minister, were:

  • whether or not Australia should become a republic
  • the republican model that should be put to the electorate to consider against the status quo, and
  • the time-frame and circumstances in which change might be considered.(173)

Half of the 152 delegates were elected, each State being a single electorate. The total number of elected delegates was divided between the States in proportion to the total representation of each State in the Commonwealth Parliament. Voting was optional, and by post. The other half of the Convention delegates were appointed. Twenty of these came from the Commonwealth Parliament and 20 from the State Parliaments and territory Assemblies.

By 1998, the idea of an elected Convention was familiar in Australia. The original Constitution was drafted in two Conventions. The second of these consisted largely of elected delegates, a feature that was credited with attracting greater public interest in the process. The Convention of 1891 resolved that the new Constitution should provide for State based elected Conventions to approve alterations of the Constitution. This proposal was abandoned in 1897, in favour of the referendum. Speculation about the use of Conventions continued after federation, however, albeit as a mechanism for initiating rather than approving proposals for change.(174) An elected Convention was proposed in 1919, as a mechanism for examining the scope of Commonwealth powers, although ultimately the proposal did not proceed.(175) As the limitations of the parliamentary Convention model came to be recognised in the early 1980s, a bipartisan team of Gareth Evans, Haddon Storey and John McMillan proposed an alternative Convention model involving a minority of non-parliamentary delegates.(176) In 1993, the non-partisan Constitutional Centenary Foundation released a discussion paper drawing attention to the questions that would need to be resolved if, as was likely, the approaching centenary of the Convention of 1897-98 prompted another Convention to be held.(177)

In the event, the idea of a partly elected Convention was taken up by the then federal Opposition, as a vehicle for making decisions about the proposed Australian republic.(178) The Convention would be an alterative to the Keating Government's preferred model of a plebiscite to ascertain public views on the principle of a republic, leaving the Parliament to develop the detail of the proposal for submission to referendum. Following the defeat of the Keating Government in the 1996 election, the incoming Howard Government reaffirmed its earlier commitment to a Convention.

The principal advantage of an elected Convention lay in its potential to attract public interest and to involve the voters to a degree in the development of a proposal before it reached the referendum stage. It also appeared to offer an opportunity to bypass or neutralise the party political division over constitutional issues in the Parliament, on which previous debates about the Constitution had foundered. But there were difficulties with the Convention as well. One concerned its relationship with the Parliament. This had been less of an issue in the 1890s, when there was no single, national Parliament to take the responsibility for constitutional change. The Constitution had established the Commonwealth Parliament, however, and given it the authority to initiate referendum proposals. There were questions about how this authority could be reconciled with proposals for change emanating from a Convention. This had been an issue even in relation to the ACC comprising delegations of parliamentarians. The inability of the ACC to contribute to the drafting of referendum or to influence the substance of the arguments that were distributed for and against its own proposals was an irritant. The greater democratic legitimacy of an elected or even partly elected Convention, however, highlighted the difficulty further.

These issues were resolved in relation to the 1998 Convention only on an ad hoc basis. The appointment as delegates of Members of the Commonwealth Parliament helped to build a bridge between the deliberations of the Convention and the ultimate responsibility of the Parliament to initiate proposals for constitutional change. The Parliament enacted enabling legislation for the Convention. The Constitutional Convention (Election) Act 1997 dealt largely with electoral matters, however, as its name implies. It did not attempt to prescribe a plan of action, as had the enabling Acts in the 1890s. On the last day of the Convention, the Prime Minister announced his decision that a particular model had attracted sufficient support to justify its submission to referendum. The criteria on the basis of which this decision was made were contested, however.(179) The Convention's communique was expressed in principle only and its model was not complete. The resolutions were translated into legal form and augmented to the extent necessary by the Government, without formal Convention involvement. The drafters adhered faithfully to the Convention's model, even where it was infelicitous. A Joint Select Committee of the Parliament(180) deliberated on details of the bill and caused some changes to be made without, however, interfering with its central tenets. In due course, both Houses passed the Bill, with absolute majorities. Members of Parliament prepared the cases for and against, in the usual way. The result tended to confirm existing doubts about the usefulness of this form of information in helping the public to understand the meaning of referendum proposals.

A second difficulty concerned the nature of the representation that could be expected from a Convention charged with negotiating a specific proposal of this kind. Playford had anticipated the problem in 1891, when he said:

... in this mode of convention you can never ascertain correctly the views of the people. You only ascertain the views of the men who have been elected members of the convention.(181)

The problem manifested itself clearly in 1998. Even where delegates stood on broad platforms, they needed room to manoeuvre in order to reach an agreed outcome. Other features of the Convention exacerbated the disjunction between voters and delegates as well. Only one half of the Convention was elected. Voting was voluntary rather than compulsory, and less than 50 per cent of enrolled voters participated.(182) Events moved quickly, from the passage of the enabling Act(183) to the election of delegates and thereafter to the meeting of the Convention itself.(184) In these circumstances, it was difficult for voters to comprehend either the nature of the process or the substance of the issues. The effect of the agenda, prescribed for the Convention in advance,(185) was to preclude serious consideration of the direct election model, already known to be favoured by the public at large. Delegates who were not connected with either of the two major groups-the Australian Republican Movement and Australians for a Constitutional Monarchy-effectively were excluded from the planning process.(186) The focus of much of the attention in the Convention was on developing a model sufficiently acceptable to Members of Parliament to ensure that it would be put to referendum, rather than on whether it was likely to succeed at that stage. There was no break in the Convention proceedings to gauge public and other reactions, as had been prescribed by the enabling Acts for the Convention of 1897-98.

Nevertheless, at the time, the Convention was deemed a great success. While the model it produced was novel and, in some respects, imperfect it was a genuine compromise, as all Constitutions must be. During the two weeks of its meeting, the Convention attracted great public interest and attention as delegates listened to proceedings, attended sessions at Old Parliament House, read newspaper reports and laughed at cartoons. It seemed a symbolic and intensely Australian moment. When the referendum eventually was held, however, on 6 November 1999, it was rejected in all States and by a national majority. Opinion polls suggested that the cause was not the principle of a republic, apparently favoured by almost 80 per cent of voters, but the particular republican model being proposed. If that is correct, the Convention failed in one of two ways. Either it developed an unacceptable model, or it developed the model in a way that did not give the public a sense of ownership and did not enable them to understand it.

These events leave Australia with the outstanding question about whether the Convention mechanism is appropriate for the development of proposals for constitutional change. Lessons can be learnt from the experience of 1998 and some of the flaws of the process eliminated for another occasion. Problems associated with the nature of a Convention and the effectiveness of representation in it are more difficult to eradicate entirely, however. In the wake of the referendum failure, there was a tendency to assume that the earlier proposal for a plebiscite would have been better after all, modified to enable some consideration of options, after the manner of the New Zealand electoral referendums.

Part 4: Hindsight

There were divisions between the framers of the Australian Constitution about the Constitution alteration procedure, as there were on most significant issues. The view that ultimately prevailed was, by the standards of the times, progressive and idealistic. Consistently with what was assumed to be the trend of democratic development, it gave the electors a voice in decisions about constitutional change. The voice was essentially passive: to approve or reject. Sole authority to initiate proposals for change was conferred on the Parliament. Despite strategic manoeuvring by supporters of the House of Representatives and the Senate respectively, and an occasional expression of concern that the path of initiation was unduly narrow, in general the framers accepted that the Parliament was best suited to this role. In general also, they assumed that deliberations in the Parliament would be an effective means of informing voters about what proposals meant.

The framers had different expectations of the extent to which a Constitution alteration procedure might need to be used. But they all assumed, explicitly or implicitly, that change would be needed at some stage. They wanted the mechanism to reflect the views of Australians organised both nationally and by States. They expected the mechanism that they chose to be relatively difficult. Nevertheless, they wanted and expected it to lead to change more readily than had proved possible in relation to the Constitution of the United States.

Their expectations have not been met, in several respects. The first and most obvious is the proportion of proposals rejected at referendum. Both the proportion and the level of opposition to individual proposals are increasing. In the absence of other explanation, this suggests a substantial and widening gap between the views of Members of the Commonwealth Parliament and the voters on constitutional questions. In a system of representative government, the existence of such a gap merits inquiry and appropriate remedial action.

One possible, partial explanation for the referendum record is the level of public understanding of proposals that are put. Opinion is divided about the nature and significance of this phenomenon. There is an argument that Australians have a certain innate level of understanding of constitutional principles, whatever their knowledge of the Constitution itself may be. Whether that view is correct or not, however, principles become technical when translated into constitutional terms, and technical detail can easily be misrepresented and misunderstood. In this respect also, the expectations of the framers have not materialised. Parliament has not proved an effective mechanism for helping voters to understand proposals for change that Parliament itself, at least formally, has initiated. The current procedure has been in place for the last 90 years and does not aid public understanding, whatever other function it may perform.

Governments and Parliaments are well aware of the escalation in the degree of difficulty of changing the Constitution. Bipartisanship generally is assumed to be the answer and ostensibly has been sought through a variety of different means. On average, there has been one comprehensive review of the Australian Constitution every 25 years since federation. Each has identified a large number of proposals for change. Each also has been largely ineffective in securing sufficient consensus on change, within either Parliament or the electorate. Very little has followed from any of them, as a result.

One possible explanation for this peculiarly Australian experience lies in the practical challenge of reconciling the referendum with two more pervasive features of the Australian constitutional system: representative government and responsible government. Responsible government has caused the initiation process to be narrowed further still, by effectively excluding the possibility that referendums might emanate from the Senate alone. Habits of mind encouraged by responsible government may lie behind the manipulation of every attempt at constitutional review by the Government of the day and, in most cases, its disruption by the Opposition. The adversarial style of politics that tends to accompany responsible government has left its mark on constitutional debate as well, ranging parties and jurisdictions against each other on issues over which agreement and consensus logically are possible. Familiarity with decision-making through the parliamentary process may have obscured the need for representatives to develop techniques to explain to voters the substance of one special category of laws: those to alter the Constitution.

These tensions between the referendum and other principles of government were foreshadowed by some of the framers of the Constitution, although not in detail. Some of the framers would have abandoned the referendum altogether, in favour of other procedures. This has been suggested since federation as well, most notably by a minority report of the Royal Commission on the Constitution. It is no longer an available option, even if it were desirable. There is no realistic prospect of removing the referendum requirement from the Constitution. If anything, the trend towards its greater use, which some delegates perceived, is becoming more of a reality as the world moves into a new century.(187) If the constitutional referendum is here to stay, with the Parliament as the sole avenue for initiating change, clearly it is important that the two work effectively together.

Part of the answer may lie with the Parliament itself. If its Members wanted to do so, Parliament could develop a more bipartisan, more inclusive approach to matters such as the Constitution that require a longer-term view. A joint standing committee charged with making a regular report on the Constitution in accordance with agreed terms of reference might be one mechanism for this purpose. Agreement on proposals for change that were not issues between the parties would enable them to be put to referendum with elections. Cost would be less of a factor, even if a proposal were rejected. Similarly, if it wanted to do so, Parliament could develop more effective means of informing the public about referendum proposals. The cases for and against might be co-ordinated with each other and made less strident. An independent body might be authorised to design and execute an information program, with a right of challenge to another body if the program was perceived not to be fair. Finally, if it wished to do so, the Parliament could involve a wider group or the people in the development of particular proposals, drawing upon the now well-stocked Australian tool-box of procedures for constitutional review.

The Australian Constitution will continue to change whether the formal alteration process is invoked successfully or not. The nature and extent of this process can be gauged by the present irrelevance of many of the proposals that have been rejected at referendum and some of the recommendations of bodies of review. In some cases this is because the proposals themselves are now outmoded.(188) In many others, however, it is because the objects of the proposals have been met in other ways: through judicial decision,(189) inter-governmental co-operation,(190) or other changes in government practice. This is not intended to suggest that all these early proposals were unnecessary; the course of Australian history may have been different had particular referendums been approved at the time. The point is made to illustrate that there are alternatives to formal constitutional change.

Sometimes evolution is preferable to formal change. There is no case for no change, however. There are limits to what can and should be achieved through either judicial review or political practice. Practical limits lie in the text of the Constitution. Normative limits are imposed by considerations of democracy, Australian-style. It cannot be assumed that national life must adapt itself to the same constitutional provisions for all time. The challenge for the next century is to find ways of enabling the Parliament and the electorate to work in a more productive partnership to achieve constitutional change, when it is appropriate to do so.


  1. Literally, the procedure applies only to 'This Constitution' and not to the rest of the Commonwealth of Australia Constitution Act (UK), of which the Constitution is still part.

  2. Constitution Alteration (Referendums) 1977. The territories to which the section refers are those allowed representation in the House of Representatives.

  3. '... the proposed law shall be submitted ... to the electors ...'.

  4. '... the Governor-General may submit the proposed law ... to the electors ...'.

  5. For example, section 64: 'The Governor-General may appoint officers to administer ... departments of State ...'. See also section 12.

  6. There is some controversy about whether the reference to affecting the provisions of the Constitution refers only to provisions dealing with State limits or to those dealing with State representation as well. While the contemporary commentators Quick and Garran, who ought to know, take the broader view, both history and logic appear to support the narrower interpretation here: J. Quick and R. R. Garran, Annotated Constitution of the Australian Commonwealth, (1901, Legal Books repr. 1976), p. 991.

  7. Section 30.

  8. See, for example, Alfred Deakin: '... the double voting power in that colony and in any which follow its example would be certainly unfair to the remaining States'. Australasian Federal Convention, Adelaide, 1897, Debates, p. 1025. (hereafter, 'Adelaide session').

  9. Section 4.

  10. 'When a proposed law is submitted to the electors the vote shall be taken in such manner as the Parliament prescribes'.

  11. Scott Bennett and Sean Brennan, 'Constitutional Referenda in Australia', Department of the Parliamentary Library, Research Paper no. 2, 1999-2000, pp. 20, 21.

  12. Senate Elections (1906); State Debts (1910).

  13. Casual Vacancies, Referendums and Retirement of Judges, all in 1977.

  14. State Debts (1928); Social Services (1946); Aborigines (1967).

  15. John Chesterman and Brian Galligan, Citizens Without Rights, Cambridge University Press, 1997, p. 186.

  16. David Butler and Austin Ranney, eds, Referendums around the World, American Enterprise Institute, Washington, 1994, p. ix.

  17. David Butler and Austin Ranney, ''Practice' in Butler and Ranney, eds, op. cit., pp. 1, 5, Table 1-1.

  18. See, for example, the 'Principles of Constitution-Making' in the Commonwealth Human Rights Initiative, Recommendations to Commonwealth Heads of Government, Promoting a Culture of Constitutionalism and Democracy in Commonwealth Africa, 1999.

  19. Australasian Federation Conference, 1890, Record of Proceedings and Debates, Melbourne, p. 55. (hereafter, Melbourne Conference 1890).

  20. Deakin, Melbourne Conference, 1890, p. 123.

  21. National Australasian Convention, Sydney, 1891, Debates, p. 23 (hereafter, Sydney, 1891).

  22. Committee on Constitutional Machinery, Functions and Powers. The members were Parkes, Barton, Gillies, Deakin, Griffith, Thynne, Playford, Downer, Clark, Douglas, Grey, Russell, Forrest and Lee-Steere; Sydney, 1891, pp. 509-510.

  23. Cockburn, 'They omitted to make such a provision in the Canadian act', Playford, 'They made a mistake there; but we are not likely to do so', Sydney, 1891, p. 497.

  24. ibid., p. 495.

  25. ibid., pp. 499-500.

  26. Gillies, ibid., p. 884.

  27. ibid.

  28. ibid., p. 897.

  29. ibid., pp. 887-891.

  30. Bray, Cockburn, Deakin, Dibbs, Grey, Kingston, Playford, Smith and Suttor; ibid., p. 897.

  31. ibid., p. 895.

  32. ibid., pp. 892-3, 'I am inclined to think that, just as they will insist on this right in the establishment of the constitution, so they will in regard to any alteration of the constitution'.

  33. Adelaide, 1897, p. 1020.

  34. ibid., p. 1027.

  35. Australasian Federal Convention, Melbourne, p. 1898, Debates, p. 766, (hereafter Melbourne, 1898).

  36. Adelaide, 1897, p. 1021.

  37. ibid., p. 1023. A proposed amendment by the Legislative Assembly of Victoria, to similar effect, was negatived at the Melbourne session, Melbourne, 1898, p. 716.

  38. ibid.

  39. Reid at 736, Higgins at 740, Cockburn at 742, Kingston at 748.

  40. Forrest, 'Oh, you have that in your mind, do you?', Kingston, 'I have both positions in my mind ...', ibid., p. 748.

  41. ibid. p. 765. Those in favour of a deadlock provision were Berry, Cockburn, Deakin, Gordon, Higgins, Holder, Isaacs, Kingston, Lyne, Peacock, Quick, Reid, Trenwith and Turner.

  42. Quick and Garran, op. cit., p. 220.

  43. Above, text associated with fn. 4.

  44. Letter dated 4 December 1928, copy in the possession of the author.

  45. Sydney, 1891, p. 884.

  46. ibid., p. 898. The five seat minimum is prescribed in section 24.

  47. Quick and Garran, op. cit., p. 220.

  48. Kris W. Kobach, 'Switzerland' in Butler and Ranney, op. cit., pp. 98, 101.

  49. Sydney, 1891, p. 885 (Munro), p. 891 (Playford). See also Melbourne, 1898, pp. 740-1, where Higgins cites sections of the Swiss Constitution from 'the book which the Attorney-General of Victoria (Isaacs) has handed to me'.

  50. For example, Sydney, 1891, p. 888 (Playford). But see Melbourne, 1898, p. 755 (Howe).

  51. Melbourne Conference 1890, p. 121 (Deakin). On the use of the referendum in the US see David Magleby, 'Direct Legislation in the American States' in Butler and Ranney, op. cit., pp. 218, 219.

  52. A. V. Dicey, 'Ought the Referendum to be introduced into England?' The Contemporary Review, 1890, pp. 489, 511.

  53. Edward A. Freeman, 'The Referendum' The Universal Review, 1890, p. 331.

  54. Melbourne, 1898, pp. 758-9.

  55. Dicey, op. cit., p. 511; Freeman, quoting Bryce, op. cit., p. 347.

  56. Sydney, 1891, p. 495 (Thynne), p. 893 (Cockburn).

  57. ibid., Cockburn: 'I should like to know if any Hon. Member of the Convention holds a different opinion?'. Gillies: 'Certainly, dozens do!'. Griffith: 'I do, for one; I think it is absolutely impossible!'. Cockburn: 'I am surprised ...'.

  58. Sydney, 1891, p. 495. Gillies observation that 'You will never carry a Constitution with that proposal in it' may have assumed parliamentary ratification: at p. 892.

  59. Melbourne, 1898, p. 738.

  60. Sydney, 1891, p. 498.

  61. Adelaide, 1897, p. 1022.

  62. Melbourne, 1898, p. 755.

  63. Sydney, 1891, p. 497 (Russell).

  64. ibid., p. 894 (Griffith).

  65. Melbourne, 1898, p. 746 (Dobson).

  66. Sydney, 1891, p. 896.

  67. ibid.

  68. Melbourne, 1898, p. 738 (Glynn).

  69. ibid., p. 725 (Wise, Downer).

  70. ibid., p. 764 (Wise). This difficulty has not eventuated, under the procedure added to the Constitution by the Premiers.

  71. ibid., pp. 750-752.

  72. ibid., pp. 724, 727 (Downer).

  73. ibid., p. 758 (Isaacs).

  74. Adelaide, 1897, p. 1022 (McMillan), Melbourne, 1898, p. 741 (Higgins).

  75. ibid., p. 727 (Deakin).

  76. Compare Braddon, Adelaide, 1897, p. 1021 with Isaacs, Melbourne, 1898, p. 723.

  77. Adelaide, 1897, pp. 1021-2 (Isaacs), Melbourne, 1898, pp. 719-20 (Isaacs), p. 737 (Glynn), p. 750 (Barton).

  78. ibid., p. 771.

  79. Hereafter the 1906 Act.

  80. Senate elections.

  81. Hereafter the 1984 Act.

  82. 1906 Act section 5, 1984 Act section 7.

  83. 1984 Act sections 143, 23.

  84. 1984 Act parts III, IV, V, VI, VIII.

  85. 1906 Act section 14A, 1984 Act section 25.

  86. 1984 Act section 24.

  87. Commonwealth Electoral Act 1918 section 128A.

  88. Colin Hughes, 'Australia and New Zealand' in Butler and Ranney, op. cit., pp. 154, 155.

  89. F. Crisp, Australian National Government, (Longmans 5th edn., 1983), p. 51.

  90. Sydney, 1891, p. 498.

  91. 1911, 1919, 1926, 1946, arguably 1988.

  92. 1906, 1928, 1944, 1948, 1951.

  93. 1974, 1988.

  94. 1937, 1946, 1967, 1977 (simultaneous elections), 1999.

  95. Hughes, op. cit., p. 161.

  96. Melbourne Conference 1890, p. 121.

  97. 1906, 1910, 1913, 1919, 1928, 1946, 1973, 1974.

  98. Four of the 21 submitted at the time of a general election have been approved, as against four of the 23 submitted separately.

  99. Colin Hughes, op. cit., p. 161.

  100. Sydney, 1891, p. 894 (Griffith), p. 896 (Baker and Deakin), Melbourne, 1898, p. 747 (Dobson).

  101. Adelaide, 1897, p. 1023 (Cockburn).

  102. Melbourne, 1898, p. 741.

  103. Civics Expert Group, Report, AGPS, Canberra, 1994, pp. 80-106. For a discussion of the meaning of these findings, see Australian Quarterly, 1995, vol. 67, no. 3.

  104. 1984 Act sections 8, 11, 14.

  105. 1906 Act section 6A, 1984 Act section 11.

  106. For a summary of each of the cases that has been used, see House of Representatives Standing Committee on Legal and Constitutional Affairs Constitutional Change, February 1997, pp. 62-114. (hereafter Constitutional Change). Some of the issues were canvassed before and during the Brisbane session of the Australian Constitutional Convention, Official Record of Debates 1985, vol. 1, xci, pp. 288, 424.

  107. 1906 Act section 48, 1984 Act section 122.

  108. The information below is taken principally from two sources: Lynette Lenaz-Hoare, 'The History of the "Yes/No" case in Federal Referendums and a Suggestion for the Future' and Cheryl Saunders, 'Referendum Procedures' in Australian Constitutional Convention, Proceedings, Brisbane, 1985, vol. II, Constitutional Amendment Sub-Committee Report, pp. 85, 111.

  109. Lenaz-Hoare, op. cit., pp. 87-88.

  110. Bennett and Brennan, op. cit., p. 5.

  111. ibid., 11, fn.20. The bills dealt with advisory opinions, interchange of powers, simultaneous elections, four-year terms and outmoded and expended provisions.

  112. Commonwealth Parliamentary Debates, Senate, 7 June 1984, p. 2731 (Evans).

  113. 1906 Act section 6A(4), 1984 Act section 11(4).

  114. Commonwealth Parliamentary Debates, Senate, 7 June 1984, 2731 ff.

  115. ibid., p. 2762.

  116. 1984 Act section 11(4)(b).

  117. Commonwealth Parliamentary Debates, House of Representatives, 11 March 1999, p. 3762 (Williams).

  118. Damien Freeman 'Public Information Machinery and the 1999 Referenda' (1999)10, Public Law Review, p. 243.

  119. Of which the author was a member.

  120. Commonwealth Parliamentary Debates, House of Representatives, vol. 74, 1914, p. 2420.

  121. Royal Commission on the Constitution, Report 1929, p. 50.

  122. Bennett and Brennan, op. cit., p. 11.

  123. For a list to 1988, see Brian Galligan and J. R. Nethercote, eds, The Constitutional Commission and the 1988 Referendums, CRFFR, 1989, pp. 138-146.

  124. 'Australia', in XX Roundtable 396, p. 409.

  125. K. Bailey, 'Report of the Royal Commission on the Constitution of the Commonwealth', reprinted in W. Prest and R. Mathews, The Development of Australian Fiscal Federalism, ANU Press, 1980, p. 83.

  126. Colin Hughes, 'Commonwealth Constitution: Methods of Initiating Amendments' in Constitutional Amendment Sub-Committee, Report to the Standing Committee of the Australian Constitutional Convention, op. cit., pp. 34, 36-38. (hereafter Hughes, Initiating Amendments).

  127. Commonwealth Parliamentary Debates, House of Representatives, vol. 113, 2 June 1926, p. 2537.

  128. G. Sawer, Australian Federal Politics and Law 1901-1929, Melbourne University Press, p. 292.

  129. Commonwealth Parliamentary Debates, House of Representatives, 9 March 1928, p. 3768.

  130. 'Australia', in XX Roundtable op. cit., p. 415.

  131. Constitution Alteration (Power of Amendment) 1930.

  132. Constitutional Commission, Report 1988, p. 33.

  133. ibid., Appendix A, p. 895.

  134. ibid., p. 33.

  135. ibid., p. 46.

  136. ibid., pp. 39-41.

  137. ibid., p. 42.

  138. Estimated at c. 70 per cent: Constitutional Commission, op. cit., para 1.56.

  139. Australian Constitutional Convention, Debates, Brisbane, 1985, p. 297.

  140. ibid., p. 301.

  141. ibid., p. 366.

  142. ibid., p. 42.

  143. ibid., p. 46.

  144. ibid., p. 48-49.

  145. Senate Standing Committee on Legal and Constitutional Affairs, 'The Constitutional Qualifications of Members of Parliament', Parliamentary Paper no. 131/1981; House of Representatives Standing Committee on Legal and Constitutional Affairs, 'Aspects of Section 44 of the Australian Constitution', July 1997.

  146. Commonwealth Parliamentary Debates, House of Representatives, 24 May 1956, p. 2453.

  147. Commonwealth Parliamentary Debates, House of Representatives, 15 February 1956, p. 18.

  148. Sydney Morning Herald, 25 May 1956, p. 5.

  149. Sydney Morning Herald, 21 June 1956.

  150. J.E. Richardson, ''The Parliamentary Joint Committee on Constitutional Review'', Canberra Bulletin of Public Administration, pp. 154, 155.

  151. Victorian Parliamentary Debates, Legislative Assembly, 1958-59, vol. 257, p. 3009.

  152. Sydney Morning Herald, 23 June 1960.

  153. Geoffrey Sawer, ''Reforming the Australian Constitution'', XXXII, Australian Quarterly, 1960, p. 29.

  154. Commonwealth Parliamentary Debates, Senate, 27 March 1957, p. 189 (Seward).

  155. Canberra Times, 3 October 1958.

  156. Commonwealth Parliamentary Debates, House of Representatives, 13 April 1961, p. 820.

  157. ibid., p. 809.

  158. ibid., p. 811.

  159. Commonwealth Parliamentary Debates, House of Representatives, 12 April 1962, p. 1635.

  160. Commonwealth Parliamentary Debates, House of Representatives, 11 April 1965, p. 533.

  161. For the 1973 session see Australian Constitutional Convention, Proceedings, pp. lxxix-xcvi.

  162. Heather McRae and Anne Mullins, The Australian Constitutional Convention 1973-1985: A Guide to the Archives, Centre for Comparative Constitutional Studies, 1998, p. 3.

  163. ibid., p. 183.

  164. ibid. pp. 183-226.

  165. The normal rule was that Government and Opposition had equal representation. Departures from it occurred, but were always controversial.

  166. In 1973 and 1974.

  167. Casual Vacancies, Referendums and Retirement of Judges in 1977. Simultaneous Elections was unsuccessful in 1977, as was Interchange of Powers in 1984.

  168. Bennett and Brennan, op. cit. p. 11.

  169. McRae and Mullins, op. cit., pp. 222-225.

  170. Particularly the Tasmanian and, to a lesser extent, the Victorian and South Australian delegations.

  171. Cheryl Saunders, 'Historical Overview', in McRae and Mullins, op. cit., pp. 4-5.

  172. Constitutional Commission, Report, op. cit., p. 39.

  173. Commonwealth Parliamentary Debates, House of Representatives, 26 March 1996, pp. 3061-3. (Howard)

  174. Colin Hughes, 'Commonwealth Constitution: Methods of Initiating Amendments', op. cit., p. 36.

  175. ibid., p. 37.

  176. John McMillan, Gareth Evans, Haddon Storey, Australia's Constitution: Time for Change?, Allen and Unwin, 1983, pp. 364-370.

  177. Constitutional Centenary Foundation 'If we wanted to review the Constitution, how would we do it?'

  178. Alexander Downer MP, Press Conference, 10 November 1994.

  179. Constitutional Convention 1999, Report, pp. 986, 991.

  180. Joint Select Committee on the Republic Referendum.

  181. Sydney, 1891, p. 891. See also pp. 892-3 (Cockburn); cf. pp. 894-5 (Griffith).

  182. 5.4 million voted in the referendum, from an electorate of nearly 12 million.

  183. Constitutional Convention (Election) Act 1997.

  184. The legislation passed the Senate on 28 August 1997. The formal notice of the election was issued on 29 September and nominations closed ten days later. Voting papers were posted between 3 and 14 November and the poll closed on 9 December. The Convention met on 2February 1998.

  185. Moira Rayner, 'A view from the fringe', (1998) 21 University of New South Wales Law Journal, pp. 908, 911.

  186. ibid.

  187. Butler and Ranney, op. cit., p. 5. See also The Economist 'Happy 21st century, voters', 21 December 1996.

  188. For example, early proposals seeking authority to nationalise industries (1911, 1913, 1919) or to outlaw the Communist Party (1951).

  189. For example, proposals dealing with trade practices (1911, 1913, 1919,1926) or seeking to free regulation from the strictures of section 92 (1937, 1946).

  190. For example, proposals dealing with corporations regulation (1911, 1913, 1919, 1926).