• Specific Year
    Any

Kildea, Paul; Smith, Rodney --- "The Challenge of Informed Voting at Constitutional Referendums" [2016] UNSWLawJl 14; (2016) 39(1) UNSW Law Journal 368

THE CHALLENGE OF INFORMED VOTING AT CONSTITUTIONAL REFERENDUMS

PAUL KILDEA[*] AND RODNEY SMITH[**]

I INTRODUCTION

As the push for a referendum on constitutional recognition of Aboriginal and Torres Strait Islander peoples has gathered momentum, so has discussion of the importance of ensuring that Australians are well-prepared to cast a vote on the topic when the moment comes. There is broad acceptance that one of the highest priorities in this respect is building citizens’ capacities to cast informed votes. Both official reports and expert commentary have expressed concerns about Australians’ existing levels of knowledge and awareness of constitutional recognition, and the Constitution more broadly, and have proposed measures to address them. In September 2014, for instance, a government-appointed advisory panel cited survey data showing poor community awareness and understanding of Indigenous recognition and made the blunt assessment that ‘[t]he wider Australian public is not yet ready for a referendum’.[1] George Williams has written about a ‘disturbing lack of knowledge about the Constitution and Australian government’ and the need for public education on Indigenous recognition ‘so that voters can feel confident in embracing the change’.[2] An Expert Panel on Indigenous recognition recommended in 2012 that ‘[b]efore the referendum is held, there should be a properly resourced public education and awareness program’.[3]

This is only the most recent public conversation about the challenge of promoting informed voting at referendums in Australia. In fact, this conversation has been going on for more than a century. As reflected in recent debates about constitutional recognition, the dominant strains are those of anxiety and pessimism – whether these feelings are directed at the innate limitations of ordinary citizens, or the improbable logistical challenge of getting an entire electorate up to speed on technical legal issues. As early as the Convention Debates, Sir Samuel Griffith worried that it was impractical to expect that millions of voters could ‘exercise an intelligent vote’ on complicated amendment proposals, for to do so they ‘must be thoroughly familiar with every detail’.[4] Since then, numerous observers have suggested that citizens are not competent to decide referendum questions – L F Crisp, for example, wrote that people are ‘usually ill-equipped and disinclined’ to inform themselves about constitutional matters[5] – and this view has recently received support from public opinion surveys conducted in conjunction with the 1999 referendum.[6] This apparent ignorance has prompted concerns that public understanding is a barrier to referendum success,[7] a problem that compulsory voting is thought to compound.[8]

In opposition to this strain of commentary is one which is far more sanguine about the ability of Australians to cast informed votes at referendums. Adherents to this view argue that people possess an instinctive understanding of general constitutional principles and are capable of making prudent decisions about constitutional change.[9] As evidence of this, they point to examples of Australians voting consistently across different referendums (for example, on State debts in 1910 and 1928), or times when Australians have distinguished between two referendums held on the same day (as happened in 1967, when voters overwhelmingly supported amendments regarding Aboriginal and Torres Strait Islander peoples, but defeated a proposal to alter the parliamentary ‘nexus’).[10] Looked at in this way, the referendum record is evidence that voters possess common sense and sound judgment, and are not easily swayed by a government push for change.[11] If poor judgment has indeed occurred, it may have been perpetrated by elites, who have miscalculated both the need for change and the public mood.[12] Despite the sharp differences between this sanguine view and the pessimistic one described above, both strains of commentary see it as desirable for voters to make considered, as opposed to arbitrary, choices when voting at referendums.

This article does not take sides in this ongoing public conversation about informed referendum voting. Instead, it seeks to provide a perspective on the subject that is currently missing. There is scope for more reflection on the larger questions that sit in the background of this conversation, such as why informed voting might be considered valuable in the first place, and what it means to cast an informed vote in the context of referendums. Following on from this, there is also a need for more detailed empirical analysis of the extent to which Australians are already meeting the challenge of informed voting, and the ways in which they fall short. Further, there is room for greater appreciation of the ways in which the law shapes the information environment as it applies to referendum campaigns, the degree to which it helps or hinders Australians to cast an informed vote, and the potential and limits of law reform in helping citizens make considered choices. In seeking to expand the scope of existing debates about referendum voting in an original way, we employ the methods and analytical approaches of both law and political science: two disciplines that, despite being highly complementary, are rarely combined in scholarship on this topic.

Part II examines the idea of informed referendum voting. It discusses why it might be considered valuable, how it differs from informed voting at elections, and suggests three different approaches to informed voting. Part III looks at how legal regulation of referendums affects the capacity of citizens to both form a view on reform proposals, and to cast a vote that is consistent with that view. Part IV considers a variety of evidence, including data from original focus group research, to explore the different ways in which Australians try to meet the challenge of informed voting, the extent to which they succeed, and how they feel about it. Part V suggests a variety of reform measures that might make citizens better able to meet the challenge of informed referendum voting. Part VI concludes the article and considers the natural limits of any attempt, by legal means or otherwise, to improve the incidence of informed referendum voting in Australia.

II WHAT IS THE CHALLENGE OF INFORMED VOTING AT REFERENDUMS?

A Why Informed Voting at Referendums?

Among the most basic assumptions of liberal democracies is the idea that citizens have a stake in the way they are governed and that therefore they will exercise their votes in an informed way. By casting informed votes, citizens act according to their opinions and interests and the collective outcome of voting is legitimate. From this perspective, uninformed voting is a corruption of the type of deliberative and informed decision-making that voting should represent. Following John Stuart Mill, many commentators have made the claim that uninformed voting not only indicates a lack of proper personal development, it also produces poor representatives and bad government. Mill argued that uninformed voters were likely to disadvantage themselves and damage the collective interest of a country. For these reasons, he contended that voting ought only to be available when people showed a willingness, among other things, to become informed about the issues at stake.[13] This type of claim is repeated throughout the literature on voting and citizenship. Ian McAllister, the leading political scientist in Australian electoral research, argues that:

One of the most important requirements for the functioning of representative democracy is the existence of informed and knowledgeable citizens. It is normally considered a pre-requisite to voting in a democracy that citizens have some basic information about how the system operates.[14]

In ordinary elections for political representatives, informed voting could refer to at least three areas of knowledge. First, it could refer to the possession of practical and technical information about the who, when, where and how of voting, as well as the ways in which votes are tallied to determine winners and losers. Second, it might refer to the basis for choosing between different candidates. Informed voting based on a certain level of knowledge and reflection could be contrasted with a voting decision based on uncritical partisan loyalty, ‘gut instinct’, the ‘honest’ face of a candidate, cynicism about politicians, and so on. Third, it might be understood to encompass knowledge about the place of voting and elections within the broader democratic system.

B Elections and Referendums: Similarities and Differences

Each of these types of knowledge has relevance for constitutional referendums as much as for parliamentary elections, although the specifics differ somewhat in each case. In the Australian context, voting in referendums might be assumed to present greater challenges than voting in parliamentary elections. For one thing, constitutional referendums have become rare. Australians aged under 34 will have voted in up to five federal elections and a similar number of state or territory elections but will not have had the opportunity to vote in a single referendum. The 16-year gap without a referendum between 1999 and the present is not unprecedented: a similar gap occurred between 1951 and 1967. Australians voted on proposed constitutional changes on 12 occasions up to 1951 but have only done so seven times since then. The infrequency of Australian votes on constitutional matters, combined with the fact that the issues at stake concern the most basic laws of the Commonwealth and the requirement that any change receive support from a majority of Australian voters as well as voters in a majority of states, might suggest that constitutional referendums require particularly informed consideration by voters.

The practical and technical information required of voters in referendums is, however, arguably less onerous than that for ordinary federal elections. Although federal elections and referendums tend not to be held simultaneously, voters can apply much of their knowledge about the former to the latter. As with federal elections, voting is compulsory. The Australian Electoral Commission (‘AEC’) is responsible for administering the same suite of voting methods – ordinary votes, pre-poll votes, postal votes, absentee votes, mobile polling and overseas votes – as for federal elections. Many of the same polling places are used. The ‘Yes/No’ structure of the ballot paper is simpler than that for either the House of Representatives or Senate ballots. The double majority provision that determines whether or not a proposed change has succeeded introduces some complexity into the count; however, voters in referendums do not confront the strategic considerations involved in preferential voting for either the House of Representatives or Senate. If they want the proposal to succeed, they simply vote ‘Yes’; if not, they vote ‘No’.

The equivalent in a referendum of choosing between different candidates at a parliamentary election is choosing whether or not to support the proposed constitutional amendment. In both cases, an informed vote is often understood as a vote based on understanding the policy issues, including their implications for the voter and other relevant groups (family, neighbourhood, co-workers, city, country and so on). Although the elevated status of constitutional issues might suggest a greater information burden for voters in referendums than exists in ordinary elections, this is not necessarily the case. Constitutional issues are often complex and cause disagreement among learned experts; however, the same is true of the economic and social policy issues that typically confront voters at elections. The texts of referendum questions on the ballot paper usually present the issues in a simplified way. Voters who wish to become informed about the precise changes to the Constitution have to consult the specifics of the Bill passed by the Parliament. Again, this task is not dissimilar to that faced by a voter who is attempting to discover the details behind a party’s policy announcements during an election campaign. In fact, since proposed amendments to the Constitution must be contained in a specific Bill that has been debated and passed by the Parliament, it may be easier for voters to become informed about the fine-grained details than is the case for the policy announcements that parties deliberately keep vague throughout election campaigns.

Moreover, a vote on a referendum question typically requires voters to consider whether or not they support change on a single constitutional issue, regardless of their views on a range of other constitutional issues.[15] (If more than one constitutional issue is simultaneously put to the voters, each will have its own question.) By contrast, informed voters in House of Representatives and Senate elections must compare their own positions on a range of salient issues with those of the available candidates, then rank the candidates according to the relative proximity of their overall bundle of policy positions to the voter’s own positions. This requires voters to make more complex judgments than they do when considering the single issue presented in a referendum question.

The third requirement for informed voting mentioned above, an understanding of the place of voting and elections within the broader democratic system, also presents different challenges for citizens participating in ordinary elections and in referendums. Voters may well make an informed assessment that voting in a parliamentary election is a far less effective method of democratic political participation than other traditional forms of political engagement, such as direct petitioning of parliamentary or government representatives on specific issues, or than newer forms of engagement designed to mobilise popular opinion, such as protests, boycotts and social media campaigns. Writing from a United States’ perspective, Russell Dalton argues that:

Voting is a form of action for those with limited skills, resources and motivations – the simplicity of voting explains why more people vote than any other single political activity. As political skills and resources expand, citizens realize the limits of voting as a means of political influence ... and participate through individualized, direct, and more policy-focused methods.[16]

On this interpretation, informed citizens may reasonably pay little attention to voting for their parliamentary representatives. The same reasoning cannot be applied to referendums. In contrast to the multiple means for influencing government policy or parliamentary legislation that exist regardless of the outcome of an election, the referendum is the sole avenue for constitutional change available to most Australian citizens. Citizens may campaign for proposed constitutional changes to be put to referendum in the first place, but once a referendum vote has been won or lost, that effectively ends the matter. The losers have no alternatives but to accept the result or begin campaigning for another referendum.

The possibility that such a campaign might be unsuccessful implies a further difference between ordinary elections and referendums that informed citizens must take into account. They know that the result of any ordinary election lasts for a limited period before another election must occur and the result of the previous election can be overturned. Referendums do not follow such a timetable. Multiple referendums on the same issue are possible;[17] however, there is no guarantee that a referendum proposal that is put once and fails will ever be considered again.[18] Informed voters at a referendum who oppose the constitutional status quo but who are not entirely satisfied with the proposed alternative must therefore make difficult contextual judgments about whether a more acceptable alternative will eventually reach the referendum stage sometime in the future and whether voting for or against the current proposal is likely to increase or reduce such a prospect.

C Different Conceptions of Informed Voting at Referendums

Political science research suggests that there are three ways in which we might conceptualise the behaviour of an informed voter at a referendum. In the conceptualisation that probably first comes to mind, informed voters are independent thinkers who take care to inform themselves about the details of the referendum proposal and the array of contending public arguments about its merits, using them to determine which way they will vote. Although the information costs involved in taking such an approach may be lower than they are for an ordinary election (see Part II(B) above), a voter may still consider those costs too high, particularly given the unlikelihood that her vote will be pivotal to the outcome. In this situation, voters may choose what Anthony Downs famously termed ‘rational ignorance’, keeping their information costs as low as possible.[19]

Drawing on Downs, Arthur Lupia and Richard Johnston have suggested an alternative to the independent thinker approach to informed voting. In this alternative, voters make use of ‘information short cuts’. Instead of working out their position on a referendum question from scratch, voters rely on following the position of trusted information sources that have proved good predictors of their positions on past political issues.[20] These sources would include family members, friends, interest groups, news media outlets and political parties. Lupia and Johnston accept that most citizens will be ‘ignorant of political facts’ but argue that ‘competent performance does not require that voters be fully informed about all of the minute details of a ballot question’.[21] For them, ‘citizens can be competent without detailed knowledge’ if the information short cuts get them to the same decision they would have made when fully informed.[22] Political parties, for example, often take positions on referendum questions, making party identification a natural information short cut for many voters.[23] One obvious issue that is raised by this second approach to informed voting is whether parties and other opinion leaders have an interest in distorting or misrepresenting the issues at stake in the heat of a referendum campaign.[24]

A third possible conceptualisation of informed voting in referendums occurs when – as is usually the case in Australia – the referendum question is proposed by the government of the day. In this situation, voters may cast a vote based on their overall retrospective assessments of how well the incumbents have governed. This type of retrospective voting provides an information short cut in ordinary elections. Voters who do not know much about the details of government policies draw on more immediately experienced information about how they and their communities are being affected by those policies.[25] Retrospective voting theory is usually applied to judgments about incumbent governments or legislators rather than referendums.[26] It may seem inappropriate to extend retrospective voting to referendums; however, voters may well apply

a retrospective information short cut in referendum voting. If the government

of the day seems to be doing a good job in general, then its constitutional proposals are also likely to be reasonable, whereas if it has shown itself

incapable of governing well, why should voters expect its constitutional proposals to be sound?[27] Some European studies have suggested that voters do punish governments that have been poor economic managers by voting against referendum proposals.[28]

The implicit or explicit choice confronting citizens is therefore not just whether or not to become informed about the question at hand, but which approaches to becoming informed they will adopt. Individuals may combine elements of all three approaches by, for example, reading the proposal, talking with friends and identifying the proposal with the government of the day. These different approaches have different implications for anyone whose goal is to increase informed voting in referendums, which are considered later in this article.

III LEGAL REGULATION AND INFORMED REFERENDUM VOTING

While the casting of an informed (or uninformed) referendum vote is to some extent a matter of individual choice, the capacity of electors to reach a considered view about a referendum proposal is strongly influenced by the type and amount of information available, and the ease of its accessibility. It is in this respect that legal regulation of referendum practice plays a critical part in informed voting. The law helps to shape the information environment in which referendums take place, and in doing so it can both help and hinder citizens in meeting the challenge of casting an informed vote. This Part explores the different ways in which constitutional and statutory rules affect informed referendum voting, and the extent to which the existing legal framework is adequate. It considers the impact of legal regulation on two aspects of informed voting: first, the formation of a preference about a referendum proposal; and second, the casting of a vote in a way that reflects that preference. We find that the law largely leaves Australians to their own devices when it comes to making their minds up about substantive issues, while providing a supportive framework to assist electors to cast valid votes.

A Forming a Preference on Referendum Proposals

1 Parliamentary Deliberation and Referendum Timing

One concrete way in which the law affects the ability of citizens to develop an informed view on referendum proposals is by prescribing procedures and timeframes for the conduct of referendums. For instance, the Constitution provides that proposals for constitutional amendment ‘must be passed by an absolute majority of each House of the Parliament’.[29] This requirement triggers a series of events that foster careful consideration of the proposed change: the drafting of a Bill containing the precise wording of the amendment; the possible referral of that Bill to a specialist committee; and finally, parliamentary debate. These events in turn provide a focus for public deliberation. As Stephen Tierney notes, ‘[i]f nothing else the very time it can take to pass such a law creates space for debate’ and parliamentary deliberations ‘may well help to inform public discussion’.[30]

Further, the Constitution requires that a proposed amendment be put to the voters ‘not less than two nor more than six months’ after its passage through Parliament.[31] According to legal scholar William Harrison Moore, these times were set down ‘to afford sufficient time for the electors to inform themselves of the issue and to prevent undue delay’.[32] In other words, the prescribed timeframes were put in place to prevent governments from holding ‘snap’ referendums, and to help ensure that people could ‘cast a vote on the basis of full information and free from the heat of the moment’.[33] The requirement that at least two months elapse between passage of the Bill and the referendum will be especially important to informed voting where the proposed change is complex or unfamiliar.

These constitutional rules provide a platform for informed voting, but there are limits to their effectiveness. In 2013, for example, the Labor Government used its numbers to impose a time limit on Senate debate about its proposal to recognise local government in the Constitution.[34] And providing for a minimum timeframe does not in itself ensure the deliberative quality of referendum debate – there is, for instance, still ample scope for misinformation to be spread in the heat of a campaign. The different ways in which the law addresses the challenges of information provision and misleading statements are set out below.

2 Distribution of Official Pamphlet

The most direct way in which the law seeks to support citizens in casting an informed vote is through a requirement that an official pamphlet be distributed in advance of the referendum. Section 11(1) of the Referendum (Machinery Provisions) Act 1984 (Cth) (‘Referendum Act’) requires the Electoral Commissioner to send an official pamphlet to each household within 14 days of polling day.[35] That same provision sets down rules about the content of the pamphlet: it is to contain arguments for and against the proposed constitutional amendment, each being no more than 2000 words in length; the arguments themselves are to be authorised by a majority of the members of Parliament who voted for or against the amendment in Parliament; and, a statement showing the proposed textual changes to the Constitution must be included. Parliament retains discretion to not send a pamphlet to voters (this occurred in 1919, 1926 and 1928). Where the proposed amendment receives unanimous parliamentary approval, the pamphlet is sent out without a ‘No’ case, as happened with the proposals concerning Aboriginal peoples (1967) and retirement of federal judges (1977).[36]

Commonwealth expenditure on the production and distribution of the information pamphlet was first authorised by Parliament in 1912. Proponents viewed it as a means of providing voters with basic facts about referendum proposals and of helping them make an informed choice, particularly in light of the misconceptions and misrepresentation that were seen to have characterised Australia’s first three referendum campaigns in 1906, 1910 and 1911.[37] Prime Minister Andrew Fisher said that there ‘can be nothing worse for a country than to expect the people in it to vote for or against the alteration of their Constitution without knowing what they are doing’.[38] Opposition Leader Alfred Deakin agreed, saying that ‘[i]t is our duty, when we ask the electors to vote for or against momentous proposals of this kind, to give them the best material we have in order that they may form an independent judgment’.[39]

The high ideals that accompanied the introduction of the official pamphlet now seem naïve. In practice, the information value of the pamphlet has been undermined by the prevalence of adversarial and misleading statements, and a general failure to convey basic facts about the proposed amendments and their likely impact. Saunders has remarked that ‘[t]he primary purpose of the yes/no case is to sway votes, not to provide understanding’;[40] others have suggested that the pamphlet ‘provides very little assistance to a voter who genuinely seeks to understand the issues’.[41] Three examples from past pamphlets illustrate the point. In 1974, the ‘No’ case suggested that ‘[d]emocracy could not survive’ under a ‘deceitful’ proposal to determine the average size of electorates by population, rather than the number of electors.[42] In 1988, the ‘No’ case claimed that a proposal to extend freedom of religion would threaten ‘the future of State Aid for independent schools’ and ‘open the way to extreme sects and practices’.[43] One of the key messages of the ‘No’ case in that pamphlet was: ‘Don’t risk the rights you already have. Don’t vote mistakes into the Constitution. Vote “No”’.[44] And in 1999, the ‘No’ case stated simply: ‘Don’t Know? – Vote “NO”’.[45]

If the official pamphlet is flawed, there is nonetheless a general acceptance that it should be retained. A 2009 parliamentary inquiry into Australia’s referendum machinery laws conceded that the pamphlet was falling short of its potential, but concluded that it was ‘an important communication and democratic tool through which the government can provide electors with informed debate’ on constitutional issues.[46] A similar message was communicated in our focus group interviews, where participants expressed broad support for the pamphlet, provided that it was impartial. Nonetheless, others said that the pamphlet would likely go ‘straight in the bin’. A survey conducted the day before the 1999 referendum suggests many voters have similar feelings: 51 per cent of people reported that they had received and read at least some of the pamphlet, while a further 32 per cent said they had received the pamphlet but not read any of it.[47] All of this suggests that, if the official pamphlet is to be retained, it should be the subject of a ‘rethink’.[48] We canvass some ideas for pamphlet redesign in Part V.

3 Restrictions on Federal Expenditure

The legal regulation of federal referendum expenditure has an impact on the capacity of citizens to cast an informed vote. It does this by limiting the capacity of the federal government to spend money on educational initiatives.

Legislation imposes strict limits on Commonwealth expenditure during referendum campaigns. Section 11(4) of the Referendum Act provides that the Commonwealth ‘shall not expend money in respect of the presentation of the argument in favour of, or the argument against, a proposed law’ unless that spending is in relation to:

• the preparation, printing and distribution of the official ‘Yes/No’ information pamphlet, its translation into other languages and its adaptation for the visually impaired: sections 11(4)(a), (aa), (ab), (ac);

• the ‘provision by the Electoral Commission of other information relating to, or relating to the effect of, the proposed law’: section 11(4)(b); or

• the salaries and allowances of members of Parliament, their staff and public servants: section 11(4)(c).

Section 11(4) thus confines Commonwealth spending on referendum advocacy to the production and distribution of the official pamphlet, and ancillary activities. It effectively channels all federal spending on the presentation of arguments into one outlet: the official pamphlet.

The purpose of these expenditure limits is to ensure even-handedness in federal referendum spending. They were introduced in 1984, at the instigation of non-government members of Parliament, following a move by the Hawke Government to give extra funding to the ‘Yes’ campaign in the lead up to a planned referendum.[49]

This commitment to spending neutrality affects the ability of electors to cast an informed vote by circumscribing the types of information available to voters during referendum campaigns. Section 11(4) prevents the Commonwealth from spending money to promote referendum arguments via mass media outlets such as television, radio and newspapers, even if it wishes to do so in an even-handed manner. The expenditure limits further pose a barrier to government spending on education campaigns, as such spending will be vulnerable to challenge where

any information materials produced could be perceived as crossing the fine

line between neutral information and ‘argument’.[50] Problematically, there are no equivalent restrictions on expenditure by state and territory governments, political parties, interest groups or individuals. This asymmetry is most significant with respect to state governments, which have on occasion been willing to fund campaigns opposing referendum proposals.[51]

There is a growing view that the Referendum Act’s spending rules are overly strict and act as an impediment to public education about referendum proposals. In 2009, the House of Representatives Standing Committee on Legal and Constitutional Affairs (‘House Standing Committee’) concluded that section 11(4) ‘severely restricts the way in which the Government can engage with electors on issues of constitutional change’ and recommended that existing spending limits be lifted.[52]

Recent practice also suggests that those limits are unsuited to a modern campaign environment. In both 1999 and 2013, the two most recent occasions

on which Parliament voted to put a constitutional amendment to a

referendum, legislators voted to suspend the operation of section 11(4) for the duration of the campaign. In 1999, the Howard Government used its new-found spending freedom to allocate $7.5 million each to ‘Yes’ and ‘No’ campaign committees to make the case for and against republic reform to the community, and to establish a $4.5 million neutral public education campaign conducted by a panel of experts.[53] In 2013, the Gillard Government allocated $10.5 million for the promotion of ‘Yes’ and ‘No’ arguments, and earmarked $11.6 million for a civics education campaign.[54] There is therefore a strong case for revisiting the expenditure limits currently imposed upon federal governments, and we consider some options for reform in Part V.

4 Implied Freedom of Political Communication

More broadly, the objective of informed referendum voting is supported by the Constitution’s protection of an implied freedom of political communication. That freedom, first endorsed by the High Court in 1992,[55] operates to protect communications about ‘political or government matters’ and applies to referendums as well as elections.[56] The implied freedom requires the free flow of information at all times, not just during campaign periods, and encompasses communication between voters and their representatives, between voters and election candidates, and amongst voters themselves.[57] In a series of decisions the Court has cited the goal of informed voting as a key rationale for the

protection of free political discussion. In Lange, the Court said that freedom of political communication ‘enables the people to exercise a free and informed choice as electors’.[58] Specifically in relation to referendums, Deane and Toohey JJ remarked in Nationwide News that ‘[a]n ability to vote intelligently can exist only if ... the content of a proposed law submitted for the decision of the people at a referendum can be communicated to the voter’.[59]

The implied freedom shapes the information environment of referendum campaigns by imposing limits on the capacity of the Parliament and government to curtail freedom of political discussion. The Court will strike down legislative or executive action that restricts that freedom in a manner that is disproportionate to the attainment of a legitimate objective.[60] Thus, the Parliament may not ‘deny the electors access to information that might be relevant to the vote they cast in a referendum’, unless it has a reasonable justification.[61] In ACTV, for example, a Commonwealth law that banned the broadcast of political advertisements during referendum (and election) campaigns was found to be invalid.[62] In this way, the implied freedom helps to preserve an environment in which informed voting can take place.

B Casting an Informed Vote

Legal regulation has a role to play in helping to ensure that voters, having formed a preference, are able to cast a ballot that gives expression to it. The law assists this aspect of informed voting in four ways. First, it empowers the federal government to fund the AEC to provide ‘information relating to, or relating to the effect of, the proposed law’.[63] This enables the AEC to promote awareness about the fact of the referendum and the method of voting.[64] For example, in 1999 the AEC conducted an extensive public information campaign, the aims of which were ‘to ensure all eligible electors were informed and understood what was required of them to fully participate in the referendum and to advise them of the range of services to which they had access’.[65] Through advertising and other means, the campaign provided information to electors on ‘how, when, and where to enrol and vote; how and when to vote using services such as pre-poll and postal voting; [and] how to correctly complete the referendum ballot papers’.[66] Advertising materials were translated into 25 languages for ethnic media, and 20 Indigenous languages.

Second, the law makes it unlawful to mislead voters as to the manner in which they cast their referendum ballot. It is unlawful for a person, during the campaign period, to ‘print, publish or distribute, or cause, permit or authorize to be printed, published or distributed, any matter or thing that is likely to mislead or deceive an elector in relation to the casting of a vote at the referendum’.[67] This provision has been interpreted narrowly, such that it does not apply to statements that misrepresent the substance of a referendum proposal. It applies, instead, to statements that might mislead a voter about the process of casting their vote.[68] The sort of material likely to be captured includes false instructions on how to vote (eg, an advertisement instructing voters to use a tick or a cross on their ballot), false claims about when or where to vote (eg, ‘the referendum is on Sunday’) or a statement that falsely claims that a certain group has advocated a ‘Yes’ or ‘No’ vote.[69] The provision is thus concerned with protecting voters from interference as they seek to give effect to a previously formed preference, rather than with preventing false or misleading claims that might influence the formation of that preference.

Third, the law plays a role in setting down rules about the wording of referendum questions. Legislation requires that ballot papers, and the questions presented on them, follow a particular format: the long title of the referendum Bill must be set out, and underneath voters are asked to indicate whether they approve the proposed alteration.[70] This effectively gives Parliament complete control over the wording of referendum questions, as it is Parliament that determines the title of the Bill. This obviously creates the potential for ‘loaded’ or misleading questions to be put to the people in a way that could undermine informed voting. In 1988, for example, voters were asked to approve a proposed law ‘[t]o alter the Constitution to provide for fair and democratic parliamentary elections throughout Australia’, notwithstanding the fact that the actual proposal was aimed at achieving the far more narrow objective of ‘one vote, one value’.[71]

Fourth, the law sets down rules about the manner in which electors’ preferences should be recorded. Legislation requires that voters write either ‘Yes’ or ‘No’ in a single box that appears on the ballot paper alongside the question: ‘Do you approve this proposed alteration?’[72] This system has been in place since 1965.[73] The extent to which it assists informed voting by establishing a simple procedure for recording a preference as a valid vote is open to debate. It is arguably more complicated than its predecessors. Initially, voters were asked to put an ‘✘’ next to the words ‘Yes’ or ‘No’; in 1926, the rules changed and required that electors write ‘1’ and ‘2’ next to ‘Yes’ and ‘No’.[74] One advantage of the current rules, even if they are less straightforward, is that they eliminate the possibility of donkey voting.[75] The law also permits some flexibility in how voters record their intention, provided that it is clear.[76] For instance, some words and symbols that carry the same meaning as ‘yes’ or ‘no’ (such as ‘definitely’, ‘never’, or ‘✓’) will be accepted as a clear expression of the voter’s intention.[77]

The legal regulation of Australian referendums largely places the onus on voters to become informed about the issues at stake. As in ordinary elections, much of the emphasis is on helping Australians to fulfil their voting obligations and assisting them to cast valid votes. On the substantive issues at hand, the law largely leaves voters to their own devices. They will usually receive the official pamphlet, which usually contains both ‘Yes’ and ‘No’ cases, which may vary in quality. Beyond that, voters are expected to navigate a relatively unregulated and potentially lopsided constitutional debate before casting their votes. Part IV of this article argues that in this environment most Australians will not act as independently informed voters but will seek out trusted information short cuts to help them vote.

IV EVIDENCE ABOUT INFORMED VOTING AT AUSTRALIAN REFERENDUMS

One of the consequences of the rarity of Australian referendums since the 1970s is that there is still little systematic evidence on the question of how voters approach them. The 1999 republic and preamble referendum is the only Australian instance that has been accompanied by a major social scientific survey.[78] The analysis that follows draws on that 1999 research, along with other survey studies of Australians’ constitutional knowledge, observations drawn from case studies of specific referendums and qualitative interview material from two sets of focus groups on informed voting and referendums.[79] The first set of four focus groups took place in August 2008 and involved 24 students at the University of New South Wales, most of whom were under 30 years old.[80] The second set of four focus groups, involving a total of 34 participants, took place in October 2014 in two metropolitan locations and one rural location and covered enrolled voters aged between 18 and 70 years.[81]

The combination of quantitative and qualitative research strengthens the analysis that follows.[82] Surveys reveal the broad percentages and types of Australians who respond in particular ways to the fixed questionnaire items about referendums and related political issues with which they are presented. The qualitative focus group material cannot do that, since the numbers of participants are too small to extrapolate to the Australian electorate. The value of the dynamic focus group material is that it allows more exploratory and fine-grained analysis of the ways in which Australians approach referendums. This inductive element of the research is crucial if we are to improve the currently limited state of knowledge on the topic.

A Information about the Mechanics of Voting

As noted earlier, the mechanics of voting in Australian referendums present few issues for voters who have voted before in ordinary elections. First-time voters are likely to rely on a range of formal and informal information sources about the when, where and how of voting. They often leave this to the last minute, as in the case of this young voter from the 2014 rural focus group:

I had friends and like my family ... because I had no idea how to do it. ... because it was my first time ... And none of [my friends] went up with me because they’d all done it already ... so they were like, ‘Just go up. There’s like a place up the road there you can go and do it’. And I was like, oh man. So I just kind of winged it, walked in, called my dad on the way.[83]

A consistent concern among participants in the 2014 focus groups was whether they had allocated their preferences properly in federal and state elections.[84] The simpler ‘Yes/No’ structure of the referendum ballot is likely to increase confidence among voters that they have recorded their choice correctly and avoided an informal vote. A comparison of the 1998 federal election and 1999 referendum shows that while turnout was approximately 95 per cent on both occasions, less than one per cent of electors voted informally on either referendum question (0.86 and 0.95 per cent respectively), while 3.8 per cent voted informally for the House of Representatives and 3.2 per cent for the Senate.[85]

B Information and the Importance of Voting in Referendums

Despite the theoretical reasons discussed earlier that might lead voters to give special attention to constitutional referendums, there is little evidence that Australian voters do this in practice. Some participants in the 2014 focus groups thought that referendums had a special status and that this would influence their information-gathering practices. For one, the decisions made at referendums could have more longstanding consequences than choices made at elections:

I mean for me I think that it’s more important than an election, mainly because you know that in three years’ time there could potentially be another change, whereas with a constitutional change this is a change that’s going to affect my children and my grandchildren potentially and for everybody ...[86]

Other participants noted that governments and policies change regularly – whatever happens ‘you can fix it in four years’ – whereas constitutional change is infrequent, and is ‘set in stone for a long time’. Referendums were variously described in the youngest focus group as ‘major’, ‘a big deal’ and ‘extremely huge’.

In contrast to these views, the young participants in the 2008 focus groups revealed a strong view that issues of constitutional reform, when compared to other policy issues, had little ‘direct effect’ on people’s everyday lives and were therefore given a low priority. For participants, policy issues were ‘a lot more important’ than constitutional issues because they ‘affect us directly’. One participant thought that constitutional issues were ‘like little tiny details that aren’t really going to affect anyone really at all’; another thought the policy issues were ‘more significant just because they really do affect the day to day lives of people’, while constitutional issues were more akin to ‘moral concepts which can be used to define the nation’.[87]

The suggestion from the focus groups is that while some voters pay more attention to constitutional issues than to ordinary policy issues, others give them less attention. A comparison of results from the Australian Election Study, 1998[88] and the 1999 Referendum Survey[89] suggests that voters were no more or less likely to care about the result of the referendum (71 per cent) than the election result (74 per cent), or to have an interest in the campaigns (approximately 37 per cent had a ‘good deal’ of interest in each case, with 41 and 42 per cent expressing ‘some’ interest in the referendum and election campaigns respectively). Reported media use was also remarkably similar throughout both campaigns. Thirty-two per cent followed the 1998 election campaign ‘a good deal’ via television, compared with 21 per cent via newspapers and 17 per cent via radio. The comparable 1998 referendum figures were 28, 24 and 17 per cent. Asked if they had ‘discuss[ed] politics with others’ during the 1998 election campaign, 84 per cent said that they had, slightly more than the proportion in the 1999 survey (78 per cent) who said they had discussed the referendum with their family, friends or others at least some of the time. Finally, two-thirds of voters (67 per cent in 1998 and 66 per cent in 1999) claimed they ‘definitely’ would have voted if it were not compulsory. There is nothing in these figures to suggest that, taken as a group, Australian voters believe that they treat referendums as special votes that require additional information.

C Deciding Whether To Vote ‘Yes’ or ‘No’

Only a minority of Australian voters appear to have the detailed knowledge required to make independently informed judgments on constitutional referendum issues. Six survey items assessing political knowledge have been included in the Australian Election Study series since 1996.[90] The results have been consistent over time.[91] In 2013, for example, correct answers ranged from 73 per cent of respondents knowing that ‘Australia became a federation in 1901’ to 23 per cent knowing that ‘no one may stand for Federal Parliament unless they pay a deposit’. The mean correct response over all six items was 41 per cent. Perhaps of greatest interest to this article, only 33 per cent of respondents recognised that ‘the Constitution can only be changed by the High Court’ was an incorrect statement, with 30 per cent thinking it true and 37 per cent unsure.[92] A similar lack of knowledge about Australian constitutional provisions and conventions has been recorded in surveys dating back to the early 1980s on matters including the existence of a written Constitution, the names of the two Houses of Parliament, who can serve as a Cabinet minister, how prime ministers are chosen, the appointment and powers of the Governor-General, the meaning of federalism, the role of the High Court and the non-existence of an Australian Bill of Rights.[93]

Letters and submissions to government consultations and anecdotal evidence from specific referendums also paint a picture of widespread constitutional ignorance.[94] Australia’s most celebrated referendum success, the ‘Aboriginal’ question in April 1967 that gained the support of 90.77 per cent of voters and all six states, is still widely misunderstood to have given Indigenous people citizenship and/or the vote.[95] According to one of the leading proponents of the 1967 vote, this misperception emerged during the referendum campaign itself:

The average voter at the time of the referendum in May may not have known very much about the Aboriginal question and probably had the idea, based on a good Australian tradition, that the Aboriginal people of this country had not had a fair go. Most voters probably had vague ideas that the Aboriginals had not full citizenship rights and might not be able to vote. Doubtless some voters were misinformed on many matters, but most were directed by their consciences and sense of social justice to take the view that something ought to have been done to better the lot of the Aboriginal people.[96]

Several points can be made to clarify this generally gloomy picture of constitutional ignorance. First, constitutional issues that are most distant from voters’ directly experienced or regularly observed experiences of politics are likely to be particularly difficult for them to comprehend. The survey research cited above indicates that Australians know much less about processes from which they are excluded and which rarely feature on television (Cabinet business, parliamentary proceedings outside question time etc) than they do about political processes in which they have directly taken part (voting) or have seen repeatedly highlighted on the television news (leadership challenges).[97] Second, as rational choice theory would predict, the most informed citizens are those whose resources lower the information costs of acquiring additional knowledge.[98] Voters with higher levels of formal education, those born in Australia, older voters and those already interested in politics have more constitutional knowledge than those who lack such advantages.[99]

Third, many voters are well aware of the information challenges presented by referendums and their own lack of information. Participants in the 2014 focus groups highlighted the problem that the key information demanded by constitutional referendums was hard to grasp. While agreeing that voters ‘have to be aware of what the Constitution actually says in the first place’ and ‘have to have an understanding of what it means’, participants noted that such matters ‘aren’t really that straightforward’. As a result, the authorities had to make a special effort to ‘fully explain’ what is changing and to ‘break it down into layman’s terms’.[100] In a similar vein, respondents to the 1999 Referendum Survey were asked whether, come polling day, they felt they ‘knew enough about the issues’ or alternatively would ‘have liked to have had more information about them’. Although approximately half said that they knew enough, 28 per cent said they would have liked to have a ‘little more’ information and 20 per cent would have liked a ‘lot more’ before being asked to cast their ballots.[101] Despite the considerable efforts made by the authorities in the lead-up to the referendum, adequate knowledge still seemed beyond the reach of around half the electorate.[102]

D Looking for Information Short Cuts?

There is evidence to suggest that many Australian voters respond to their limited information about constitutional matters and the apparently high cost of gaining that knowledge themselves by adopting the short cuts suggested by

Lupia and Johnston.[103] Although voters may make their minds up earlier for referendums than they do for ordinary elections,[104] opinion polls suggest that the referendum campaign period seems to matter, with the undecided segment of the electorate usually shrinking and some net change between intended ‘Yes’ and ‘No’ votes occurring over the campaign period.[105]

As might be expected, given the comparatively strong levels of party identification among Australian voters, political parties and their leaders are a key source of information cues in referendums. At the electorate level, the results of the September 1951 referendum on ‘Powers To Deal with Communists and Communism’ were highly correlated with party support at the federal election five months earlier.[106] Electorate-level and individual-level survey analysis of voting on the 1999 ‘republic’ question also shows strong party influences, even though key figures in the Labor and Liberal Parties took different positions on the issue.[107] Differences in state voting patterns over different referendums suggest that state as well as national party leaders act as cues for voters.[108] The impact of party cues underlies the longstanding assertion that bipartisan support is necessary (but not sufficient) for popular acceptance of a referendum proposal.[109]

The argument that voters rely on parties and other opinion leaders as information short cuts in referendums does not presuppose that those opinion leaders will confine themselves to rational appeals. In their analysis of the 1951 referendum, Murray Goot and Sean Scalmer note that the major party protagonists mobilised voters through a mix of reason, emotion and prejudice:

The campaign strategies of the two leaders were much as the contest demanded: each played to their strengths and sought to exploit their opponent’s weaknesses. If this meant adopting positions at odds with previous positions, talking past one another, introducing ‘extraneous’ matter, ‘scaremongering’ and so on, these were the turns the contest took. Menzies’ campaign was not an exercise in pure ‘reason’. Evatt’s campaign was not based solely on a defence of civil liberties. The decision of Evatt and of Menzies not to take each other’s arguments head-on could have been anticipated. This is how campaigns are typically conducted.[110]

Similarly, in leading the 1967 ‘Aboriginals’ referendum campaign, the Federal Council for the Advancement of Aborigines and Torres Strait Islanders (‘FCAATSI’) deliberately but misleadingly framed the proposal as one that would give ‘Aborigines full citizenship rights’.[111] It is little wonder that many voters believed that voting and other citizenship rights were at stake,[112] despite the more accurate presentation of the proposed changes in the official ‘Yes’ case.[113]

Such examples of misleading and emotional appeals may not trouble defenders of the information short cut approach to referendum voting, as long as the short cuts are consistent with the specifics of the proposals. It seems unlikely, for example, that voters in 1967 who supported full Aboriginal citizenship would have voted down the referendum proposal if they knew that it meant including Aboriginal people in population counts and allowing the federal parliament to make laws for them.

E Retrospective Voting

There is no direct evidence that Australians oppose referendum questions because they believe the government of the day has failed them in other areas. Nonetheless, there is some evidence that governments fear this response. In 1951, Prime Minister Robert Menzies worried about the effects of Opposition criticism of his Government’s economic management on the referendum result.[114] More recently, 56 per cent of respondents in the 1999 Referendum Survey agreed or strongly agreed that the republic referendum was ‘a distraction from Australia’s real problems’.[115] A number of participants in the 2008 focus groups expressed similar views about the republic proposal and a range of other constitutional changes.[116] The recent relaunching of the Australian Republican Movement produced a similar response.[117] Governments that are perceived as handling economic and other policy areas poorly may at the very least struggle to get some Australians to take a serious interest in constitutional change and may provoke backlashes against their referendum proposals.

Taken together, the evidence in this Part suggests that Australians mostly handle the mechanics of referendum voting capably, do not treat referendums as particularly special opportunities to vote, lack the information needed to allow them to cast an independently informed referendum vote, follow information short cuts provided by parties and other opinion leaders, and may use referendums to punish poorly performing governments that propose constitutional changes. The implications of these points for better informing voters in referendums are discussed below.

V REFORM IDEAS

It is apparent that Australians often struggle to meet the challenge of casting an independently informed vote at referendums. The question therefore naturally arises as to what might be done to help voters better meet that challenge. Answering that question necessarily involves acknowledging the limits of such an enterprise. As Craven observes, ‘there is no obvious means by which twenty million Australians may be brought from a standing start to rapid constitutional literacy’.[118] The objective must not be to produce an electorate of constitutional experts – instead, it should be to better equip interested voters to cast an informed vote on referendum propositions that are put before them.

Any attempt to assist voters to cast informed referendum votes must be multifaceted. Writing broadly about the topic of political disenchantment, Colin Hay usefully distinguishes between demand and supply side explanations and responses.[119] Demand side explanations direct responses to changing deficiencies within the citizenry; supply side explanations point to rectifying problems in the political goods on offer and the way they are marketed. Translated into the referendum context, demand side approaches might focus on improving political information among voters through formal education curricula, information campaigns and community engagement programs. Supply side approaches might include suggestions for the better conduct of elite politics during referendum campaigns and better reporting by the news media.[120]

Legal regulation has an important role to play here. As outlined in Part III, some elements of existing regulation pose a barrier to informed voting and are in need of reform. However, as is apparent from Hay’s analysis, law reform will only form part of any response to the challenge of informed referendum voting. Also important will be institutional innovations, such as deliberative forums, and improved civics education. Any attempt to expand informed voting must also be mindful of Craven’s warning and accept the probability that the challenge of informed voting will never be fully met.

A The Referendum Campaign

One approach to meeting the challenge of informed referendum voting would be to improve the quality of information that is made available to voters. While this is a worthy goal, it is difficult to execute because of the sheer volume and diversity of information sources to which voters are subjected over the course of a campaign. Our focus group research affirms the common sense notion that voters will access the same information sources for referendums as they do for elections: for example, news reports in the print and broadcast media, websites and social media. Many of these sources are beyond straightforward regulatory control, even if that were considered desirable, and so our analysis here focuses primarily on possible legal and institutional reforms.

In terms of law reform, efforts should be focused on amending expenditure rules to better foster information provision, and on redesigning the official pamphlet. On expenditure, we share the view of the House Standing Committee that the current limits on Commonwealth referendum spending impede genuine government efforts to educate and engage voters on constitutional change. As such, they should be removed. This would free the federal government to implement public information initiatives (such as the ‘neutral’ campaign of 1999) without fear of being in ‘technical breach’ of the law.[121] It would also permit the federal government, where appropriate, to establish publicly funded ‘Yes’ and ‘No’ campaign committees. As the 1999 experience showed, these are useful devices for enabling partisans on both sides of a debate to promote their arguments to the community, and an effective means of raising public awareness about the cases for and against change. Public funding would also potentially reduce any resource imbalances between the ‘Yes’ and ‘No’ cases, making a lopsided information campaign less likely.[122]

As to the official pamphlet, it should be redesigned with the goal of making it a source of basic, accurate information about referendum proposals, rather than the adversarial and confusing document that it has become.[123] As outlined in Part III, the pamphlet currently contains arguments for and against the proposed amendment (as authorised by parliamentarians) and a copy of the textual alterations. In addition, it should contain a clear, ‘plain English’ explanation of the relevant parts of the Constitution and the effect of the amendment. Consideration should be given to transferring responsibility for drafting pamphlet content from parliamentarians to a neutral body, or at least requiring that the information presented be vetted for accuracy and intelligibility. In 2009, the House Standing Committee recommended that parliamentarians retain the role

of ‘authorising’ the content of the ‘Yes’ and ‘No’ arguments, as it is they who

are ultimately accountable to the people for the referendum proposal.[124] The experience of other jurisdictions, however, shows that entrusting at least the preparation of referendum information to a neutral body is workable and effective. The ‘hybrid’ model employed in California is worth considering in this regard: there, the referendum information pamphlet sent out to voters contains an official summary prepared by the Attorney-General; a copy of arguments for and rebuttals against the measure, prepared by parliamentarians; and, an impartial analysis prepared by the Legislative Analyst, a public servant.[125]

Some minor changes to the distribution of the pamphlet would also better equip voters to make an informed choice at the ballot box. The legislative requirement that the pamphlets be posted to voters ‘not later than 14 days before’[126] polling day is inappropriate given that debates about constitutional reform usually start many months, if not years, in advance of a vote. Sending the pamphlet out so late diminishes its potential impact, as many voters will have already formed a preference by the time they receive it (see Part IV(D) above). The Referendum Act should therefore be changed to allow governments to distribute the pamphlet well in advance of the referendum. Another worthwhile change would be to amend section 11 of the Referendum Act to permit parliamentarians to authorise a ‘No’ case in situations where the referendum Bill is passed unanimously. As noted, an official case against some referendum proposals put in 1967 and 1977 was not included in the pamphlet due to their unanimous approval by members of Parliament. The suggested amendment would ensure that the pamphlet always presents arguments on both sides to voters, irrespective of the preferences of parliamentarians.[127] Finally, the very idea of mailing a pamphlet may be archaic. The 2014 focus group participants indicated that they were likely to access information using their own preferred methods, including digital formats.[128] The law is gradually adapting to such preferences: the pamphlet could now be published online, and distributed via email.[129]

If changes to expenditure limits and pamphlet production are worthwhile, neither addresses the larger concern that the ‘politicised, boisterous and noisy’[130] campaigns that surround referendums do little to encourage informed voting. Nor do they address the common suspicion among our 2014 focus group participants that most information sources are biased and require an independent ‘filter’.[131] One potential reform in this respect would be to extend the existing prohibition on misleading or deceiving voters, which covers statements about the manner in which electors cast their vote, to public statements about the substance of referendum proposals. This would address the fact that campaign protagonists can currently make exaggerated or misleading claims and face no legal sanction. However, this approach would be highly problematic and should not be pursued. It would risk having a ‘chilling’ effect on referendum advocacy, and it would be unclear who should be the arbiter of the truth of public utterances, or how to evaluate opinions, promises and predictions as opposed to statements of fact.[132]

An alternative approach would be to require that official ‘Yes’ and ‘No’ committees meet minimum standards of objectivity, accountability and fairness in the way that they spend public money.[133] An independent entity, such as a Referendum Panel, could review the accuracy of factual statements made by the committees, and issue instructions to withdraw, amend or retract those statements where it found them to be inaccurate, deceptive or misleading. Some material, such as campaign pamphlets or advertising, could be subject to Panel scrutiny before being approved for release into the public domain. The Panel’s oversight could be limited to statements of fact, and should not extend to expressions of opinion, which are properly understood as a feature of robust debate and disagreement.

Finally, existing practices around the wording of the referendum question should be improved to ensure that voters are faced with a balanced question. As a matter of good practice, Parliament should refer the wording of a proposed question to public research, so that it can be tested for clarity and fairness. This is done as a matter of course in the United Kingdom (‘UK’), where the Electoral Commission has a statutory obligation to consider the wording of a proposed referendum question and publish a statement on its intelligibility.[134] In recent years, the UK government has altered the wording of referendum questions (on the alternative vote, Scottish independence and membership of the European Union) in response to Commission advice.[135] In 2014, for instance, the question ‘Do you agree that Scotland should be an independent country?’ was changed to ‘Should Scotland be an independent country?’ after Commission advice that the revised wording was more neutral.[136] In the Australian context, introducing a similar practice would not alter the fact that Parliament has final say over the wording of the question.

B Before the Referendum Campaign

Efforts to foster informed voting should not be confined to the conduct of the referendum campaign. Attention should also be given to how citizen capacity might be increased in the months and years prior to polling day. One way in which this might be done is by expanding opportunities for public input and deliberation. The underlying claim here is that participation provides a means for citizens to learn about proposals for amendment and the wider constitutional system, and thus puts them in a better position to cast an informed vote. As Saunders has stated, ‘written material cannot be the sole answer. Active engagement offers an alternative’.[137]

The existing constitutional and legal framework is not especially supportive of active engagement. The constitutional requirement that a referendum be held gives citizens no formal role to play in the ‘issue framing’ stage of a referendum, in which the reform proposal is developed and debated,[138] and referendum machinery legislation focuses on information provision. There is ample room, however, for public participation to be promoted as a matter of practice. We should not expect this to occur spontaneously. Instead, special mechanisms, or ‘structured participation’, are needed to effectively involve and educate citizens.[139]

Community consultations are one mechanism by which public input on constitutional issues can be encouraged. These typically take the form of public meetings in which interested individuals can offer opinions on reform proposals and ask questions of experts. Examples include the consultations undertaken by expert panels established to inquire into the constitutional recognition of Aboriginal and Torres Strait Islander peoples, and of local government, and the meetings held by the Republic Advisory Committee in 1993.[140] As a device for enhancing informed voting, however, consultations have their limitations. Inside the meeting, participants are generally asked to express pre-existing opinions, and may come away from the meeting knowing little more about the issue than when they entered. Outside the meeting, few people may know that the consultations are taking place at all, and the ability to reach a significant audience is easily undermined by poor resourcing or weak government commitment.[141] It is also the case that public meetings attract a self-selected group of participants, many from interest groups, who do not represent a cross-section of the population.

More promising are deliberative forums, such as citizens’ assemblies and some constitutional conventions. These forums typically bring together a group of citizens to deliberate about reform proposal. Over the course of the event, they read about the proposals, have access to experts, and debate the issues in small and large groups. This structured approach can lead to significant improvements in participant understanding of constitutional reform proposals.[142] For example, delegates at the 1999 Deliberative Poll on the republic left with improved knowledge on a number of issues, including the role of the president under the proposed model (92 per cent of delegates had knowledge of this after the weekend, as opposed to 40 per cent beforehand).[143] Participants in a series of local deliberative forums run by the Constitutional Centenary Foundation in the period 1997–98 also reported substantial increases in constitutional knowledge.[144] These results are consistent with research conducted on larger deliberative events, such as citizens’ assemblies, on both constitutional and non-constitutional matters.[145]

Deliberative forums therefore offer a way for citizens to learn about constitutional issues, and also make a structured contribution to a reform process. A significant challenge in this respect is ensuring that some of the educative benefits being enjoyed by participants are available to the wider public. Publicity has an important role to play here. Constitutional conventions, for example, can be a major national event – the 1998 convention on the republic, although not deliberative, demonstrated this. Broad exposure on television and other media can help to draw attention to the constitutional issue being discussed, and substantially promote awareness and understanding among the general public. Williams and Hume have suggested that a constitutional convention be held at least once every decade to debate proposals for reform, and provide a new means for community engagement in constitutional change.[146]

Civics education programs form a second pre-campaign method for bolstering the capacity of citizens to cast informed votes. In 2009, the House Standing Committee recommended that the federal government ‘develop and implement a national civics education program to enhance the engagement of the Australian public in democratic processes and to improve knowledge and understanding of the Australian Constitution’.[147] In making this recommendation, the Committee acknowledged the need for any civics education initiative to extend beyond schools to the general public.

The extent to which a national civics education program of this kind would have an impact on informed voting is open to question. On the one hand, there is clearly scope for voters to improve their general understanding of Australia’s constitutional arrangements. On the other, citizens may be best placed to learn about their constitutional system in the context of a specific referendum campaign when actual questions are being put to them, rather than in the abstract.[148] It is also the case that a great deal of information is already available – online, and through non-government organisations and universities – for the citizen who is motivated enough to seek it out. This was a point that came up often in the 2014 focus group interviews,[149] and was acknowledged by the House Standing Committee in its 2009 Report. The Committee noted the extensive educational activities that the AEC already performs, and the role of organisations such as the Constitutional Education Fund Australia in providing information to Australians about their constitutional system.[150]

VI CONCLUSION

Despite what campaigners for ‘Yes’ or ‘No’ cases might believe, the success or failure of a referendum proposal does not signal the presence or absence of informed voting. As noted earlier in this article, the same outcome may result from an electorate that is highly informed or largely ignorant. The suggestions that we have made in this article for increasing informed voting at referendums do not imply that, were they introduced, constitutional change would be any easier or harder in Australia than it is now. Instead, what we have suggested is that the quality of the process by which individual voters reach their decisions and an overall result is achieved could be improved. At present, Australians mostly handle the mechanics of referendum voting capably but they do not consider referendums to be particularly important decision opportunities. They often lack the information they need to cast an independently informed referendum vote and rely instead on information short cuts provided by parties and other opinion leaders.

A multifaceted approach that takes into account the ways in which voters address referendum issues might improve this situation. We have, for example, suggested that the official referendum pamphlet arrives too late to affect the thinking of many voters, does not always present them with clear information, and may be delivered in a hardcopy form that many of them will ignore. More broadly, adequately funded ‘Yes’ and ‘No’ committee campaigns, reviewed by a Referendum Panel, might provide the sort of unbiased information that voters seek. As well as reducing the information costs for voters wanting information, these campaigns could also act as checks against misleading claims by political parties and the other opinion leaders to whom voters generally turn for referendum information short cuts.

We have also suggested that any attempt to increase informed voting that begins with the formal campaign is starting too late. Information campaigns should certainly continue right up until the day of the referendum, since some voters will still be making up their minds at that point, but efforts to increase information should begin in the period during which any constitutional change is first debated, through community forums and targeted civics education programs.

As noted earlier in this article, some of these measures would require regulatory changes. The extent to which any of them would improve informed voting is an open question. In large-scale elections, the costs for individuals of becoming more informed are often not matched by the benefits of casting a more informed vote, leading most voters to remain relatively uninformed. Although this situation can be improved by lowering the costs of information to individuals, there are still strong limits to the electoral information it is rational for voters to acquire. On an empirical level, Australian and international research suggests that increased voter education programs and general educational levels might yield only small improvements in political knowledge.[151]

Some commentators argue that the complex and abstract nature of constitutional issues poses an inevitable additional barrier to informed voting. Greg Craven perhaps represents this view most colourfully when he writes that ‘[s]aying the Australian Constitution does not have a strong hold upon our popular imagination is like saying fish survive better in water than on land ... Our Constitution is regarded as, in a word, dull: in three words, very, very dull’.[152] More soberly, Craven has argued that ‘most Australians simply are not interested in issues of constitutional reform, and despite the enthusiasms of their betters, will resist all attempts to impose such an enthusiasm upon them’.[153] There is force in this observation, but our arguments in this article suggest that it presents an overly simplified picture of constitutional engagement. The decision-making required by informed voting at a referendum may actually be less complex than that required in an election for parliamentary representatives. Moreover, some constitutional issues are more interesting than others, particularly where they are linked to core political values such as citizenship, democracy and rights.

A realistic goal for those seeking more informed voting at referendums is not an Australian electorate in which every voter is fully informed. Some Australians, to adapt the words of the Review Panel quoted at the beginning of this article, will never be ready for a referendum. It does not follow from this observation that no improvement could be expected to the level of informed voting if reforms were made, particularly if those reforms took careful account of the information demands that referendum choices make on voters and the ways in which voters currently deal with those demands.


[*] Lecturer, UNSW Law School, and Director, Referendums Project, Gilbert + Tobin Centre of Public Law.

[**] Professor, Department of Government and International Relations, University of Sydney. The authors thank the anonymous referees for their helpful comments. The research undertaken for this article was partly funded by the Electoral Regulation Research Network Research Collaborative Initiative, ‘The Challenge of Informed Voting in the 21st Century’.

[1] John Anderson, Tanya Hosch and Richard Eccles, ‘Final Report of the Aboriginal and Torres Strait Islander Peoples Act of Recognition Review Panel’ (Report, Aboriginal and Torres Strait Islander Peoples Act of Recognition Review Panel, September 2014) 25 (‘Act of Recognition Report’).

[2] George Williams, ‘Recognising Indigenous Peoples in the Australian Constitution: What the Constitution Should Say and How the Referendum Can be Won’ (2011) 5(1) Land, Rights, Laws: Issues of Native Title 1, 13–14.

[3] Expert Panel on Constitutional Recognition of Indigenous Australians, Recognising Aboriginal and Torres Strait Islander Peoples in the Constitution (Report, January 2012) 227. See also Review Panel, which recommended that the federal government support an expanded public awareness and education strategy: Act of Recognition Report, above n 1, 8.

[4] Official Report of the National Australasian Convention Debates, Sydney, 8 April 1891, 894 (Sir Samuel Griffith). This view was not universally shared among the framers; responding to Sir Griffith, Alfred Deakin argued that referendum proposals could be presented in a ‘short, succinct’ way, allowing voters to cast their judgment ‘with certainty and with knowledge’: at 895.

[5] L F Crisp, Australian National Government (Longman Cheshire, 4th ed, 1978) 51. See also Parker’s account of evidence before the Royal Commission on the Constitution where, citing the problem of a misinformed public, one witness remarked: ‘I think that the confidence formerly expressed in the referendum has largely disappeared’: Commonwealth, Royal Commission on the Constitution of the Commonwealth, Minutes of Evidence: Part 5 (Final), Sydney, 9 March 1928, 1331 (William Arthur Holman), quoted in R S Parker, ‘The People and the Constitution’ in Geoffrey Sawer et al (eds), Federalism in Australia (Cheshire, 1949) 135, 160.

[6] See discussion in Part IV below.

[7] Crisp, above n 5, 51; Don Aitkin, ‘Australia’ in David Butler and Austin Ranney (eds), Referendums: A Comparative Study of Practice and Theory (American Enterprise Institute for Public Policy Research, 1978) 123, 131.

[8] Ian McAllister, ‘Elections without Cues: The 1999 Australian Republic Referendum’ (2001) 36 Australian Journal of Political Science 247, 266.

[9] Brian Galligan, ‘Amending Constitutions through the Referendum Device’ in Matthew Mendelsohn and Andrew Parkin (eds), Referendum Democracy: Citizens, Elites and Deliberation in Referendum Campaigns (Palgrave, 2001) 109, 123; Brian Galligan, ‘The Republican Referendum’ (1999) 43(10) Quadrant 46, 50, 52. See also the discussion in Cheryl Saunders, ‘The Parliament as Partner: A Century of Constitutional Review’ (Research Paper No 3, Parliamentary Library, Parliament of Australia, 2000) 26–7.

[10] See generally Saunders, ‘The Parliament as Partner’, above n 9, 26–7.

[11] Galligan, ‘Amending Constitutions’, above n 9, 123.

[12] Ibid.

[13] John Stuart Mill, ‘Representative Government’ in H B Acton (ed), Utilitarianism, Liberty, Representative Government (J M Dent & Sons, 1910, 1972 ed) 171. The publications that have echoed Mill’s misgivings are too voluminous to list here. An indication of the continuing power of such concerns can be found in recent British debates about the extent and causes of 21st century political disengagement. Major contributions include Gerry Stoker, Why Politics Matters: Making Democracy Work (Palgrave Macmillan, 2006); Colin Hay, Why We Hate Politics (Polity Press, 2007); David Marsh, Therese O’Toole and Su Jones, Young People and Politics in the UK: Apathy or Alienation? (Palgrave Macmillan, 2007); Matthew Flinders, Defending Politics: Why Democracy Matters in the Twenty-First Century (Oxford University Press, 2012).

[14] Ian McAllister, The Australian Voter: 50 Years of Change (UNSW Press, 2011) 56. See also at 71.

[15] The unsuccessful 1944 ‘Fourteen Powers’ referendum is perhaps the most obvious exception to this statement: see Constitutional Alteration (Post-war Reconstruction and Democratic Rights) 1944 (Cth).

[16] Russell J Dalton, The Good Citizen: How a Younger Generation Is Reshaping American Politics (CQ Press, 2009) 67–8. For similar arguments in Australia, see Ariadne Vromen, ‘“People Try To Put Us Down ... ”: Participatory Citizenship of “Generation X”’ (2003) 38 Australian Journal of Political Science 79; Ariadne Vromen, ‘Traversing Time and Gender: Australian Young People’s Participation’ (2003) 6 Journal of Youth Studies 277.

[17] Proposals to require simultaneous House of Representatives and Senate elections, for example, were put and lost three times between 1974 and 1984: Constitutional Alteration (Simultaneous Elections) 1974 (Cth); Constitutional Alteration (Simultaneous Elections) 1977 (Cth); Constitutional Alteration (Terms of Senators) 1984 (Cth).

[18] Ron Levy, ‘“Deliberative Voting”: Reforming Constitutional Referendum Democracy’ [2013] Public Law 555, 561–2.

[19] Anthony Downs, An Economic Theory of Democracy (Harper, 1957) 244 ff. For a recent elaboration and refinement, see César Martinelli, ‘Rational Ignorance and Voting Behavior’ (2007) 35 International Journal of Game Theory 315, 315.

[20] Arthur Lupia and Richard Johnston, ‘Are Voters To Blame? Voter Competence and Elite Maneuvers in Referendums’ in Matthew Mendelsohn and Andrew Parkin (eds), Referendum Democracy: Citizens, Elites and Deliberation in Referendum Campaigns (Palgrave, 2001) 191, 196. For a critique of this kind of approach, see Michael X Delli Carpini and Scott Keeter, What Americans Know about Politics and Why It Matters (Yale University Press, 1996) 51–5.

[21] Lupia and Johnston, above n 20, 192.

[22] Ibid 195. Lupia and Johnston present some survey evidence suggesting that this is the case: at 195 ff.

[23] On party identification as an information short cut in ordinary elections, see McAllister, The Australian Voter, above n 14, 33–4; Russell J Dalton, David M Farrell and Ian McAllister, Political Parties and Democratic Linkage: How Parties Organize Democracy (Oxford University Press, 2011) 148–53. The argument that there is an informed decision to adopt a party loyalty in the first place is slightly undermined by the common argument that most longstanding party loyalties are initially developed when children emulate their parents. See, eg, McAllister, The Australian Voter, above n 14, 43–5.

[24] Brian Galligan, A Federal Republic: Australia’s Constitutional System of Government (Cambridge University Press, 1995) 130–1; Galligan, ‘Amending Constitutions’, above n 9, 114–15; Scott Bennett, ‘The Politics of Constitutional Amendment’ (Research Paper No 11, Parliamentary Library, Parliament of Australia, 2003) 17; John Higley and Ian McAllister, ‘Elite Division and Voter Confusion: Australia’s Republic Referendum in 1999’ (2002) 41 European Journal of Political Research 845, 845.

[25] Morris P Fiorina, ‘Economic Retrospective Voting in American National Elections: A Micro-Analysis’ (1978) 22 American Journal of Political Science 426; Morris P Fiorina, Retrospective Voting in American National Elections (Yale University Press, 1981).

[26] The recent survey by Healy and Malhotra, for example, does not mention voting in referendums: see Andrew Healy and Neil Malhotra, ‘Retrospective Voting Reconsidered’ (2013) 16 Annual Review of Political Science 285.

[27] Levy makes a similar argument in terms of citizens’ level of willingness to trust governing elites: Ron Levy, ‘Breaking the Constitutional Deadlock: Lessons from Deliberative Experiments in Constitutional Change’ [2010] MelbULawRw 28; (2010) 34 Melbourne University Law Review 805, 829–30.

[28] See, eg, Gilles Ivaldi, ‘Beyond France’s 2005 Referendum on the European Constitutional Treaty: Second-Order Model, Anti-establishment Attitudes and the End of the Alternative European Utopia’ (2006) 29 West European Politics 47; Nicholas Bornstein and Philippe Thalmann, ‘“I Pay Enough Taxes Already!” Applying Economic Voting Models to Environmental Referendums’ (2008) 89 Social Science Quarterly 1336; Nuria Font, ‘The Domestic Politics of the EU in the Constitutional Treaty Referendums’ (2008) 9 Perspectives on European Politics and Society 301.

[29] Constitution s 128.

[30] Stephen Tierney, ‘Using Electoral Law To Construct a Deliberative Referendum: Moving Beyond the Democratic Paradox’ (2013) 12 Election Law Journal 508, 519.

[31] Constitution s 128.

[32] W Harrison Moore, The Constitution of the Commonwealth of Australia (Sweet & Maxwell, 2nd ed, 1910) 600, quoted in George Williams and David Hume, People Power: The History and Future of the Referendum in Australia (UNSW Press, 2010) 50.

[33] Williams and Hume, above n 32, 50.

[34] Anne Twomey, ‘Is the Government Gagging the Referendum Debate?’, The Conversation (online), 21 June 2013 <https://theconversation.com/is-the-government-gagging-the-referendum-debate-15404>.

[35] Before 2013, the Electoral Commissioner was required to send a pamphlet to every elector. This made the production and distribution of pamphlets an expensive undertaking; in 1999, for example, it cost $17 million: Williams and Hume, above n 32, 87.

[36] Ibid 72.

[37] Alfred Deakin said during parliamentary debate: ‘I cannot believe that any of us have been satisfied with the amount of information conveyed to the people in regard to the previous referenda. ... a very large section [of the population] remained very imperfectly informed’: Commonwealth, Parliamentary Debates, House of Representatives, 16 December 1912, 7155. See also at 7154 (William Hughes).

[38] He added: ‘I have no doubt at all that the case will be put from both sides impersonally, and free from any suggestion of bias or misleading on the one side or the other’: ibid 7156.

[39] Ibid 7155.

[40] Cheryl Saunders, ‘The Australian Experience with Constitutional Review’ (1994) 66(3) Australian Quarterly 49, 56.

[41] Marian Sawer, Norman Abjorensen and Philip Larkin, Australia: The State of Democracy (Federation Press, 2009) 22. For further criticism see Williams and Hume, above n 32, 261–3; Lynette Lenaz-Hoare, ‘The History of the “Yes/No” Case in Federal Referendums, and a Suggestion for the Future’ in Constitutional Amendment Sub-committee, Australian Constitutional Convention 1984: Report to Standing Committee (June 1984) 85.

[42] Australian Electoral Commission, Official Pamphlet [Untitled] (Official Referendum Pamphlet, 1974) 10, accessed via Australian Electoral Commission, Australian Referendums 1906–1999 (CD-ROM, 2000).

[43] Australian Electoral Commision, Yes or No? (Official Referendum Pamphlet, 1988) 25, accessed via Australian Electoral Commission, Australian Referendums 1906–1999 (CD-ROM, 2000).

[44] Ibid 26.

[45] Australian Electoral Commission, Yes/No Referendum ’99: Your Official Referendum Pamphlet (Offical Referendum Pamphlet, 1999) 9.

[46] House of Representatives Standing Committee on Legal and Constitutional Affairs, Parliament of Australia, A Time for Change: Yes/No? – Inquiry into the Machinery of Referendums (2009) 54 (‘A Time for Change’). See also Williams and Hume, who acknowledge that the pamphlet has a role to play in educating voters: Williams and Hume, above n 32, 261.

[47] Fifteen per cent said that they had not received the pamphlet, and two per cent were unsure: Australian Electoral Commission, Submission No 24 to House of Representatives Standing Committee on Legal and Constitutional Affairs, Parliament of Australia, Inquiry into the Machinery of Referendums, 9 October 2009, attachment C, 2.

[48] Williams and Hume, above n 32, 261.

[49] Commonwealth, Parliamentary Debates, House of Representatives, 29 May 1984, 2358 (Raymond Steele Hall).

[50] See Reith v Morling (1988) 83 ALR 667. The finding in this case also signals to the AEC that its own educational initiatives – which are the subject of federal expenditure under Referendum Act s 11(4)(b) – will be vulnerable to challenge should they stray beyond promoting awareness about the fact of the referendum and the method of voting: see Graeme Orr, ‘The Conduct of Referenda and Plebiscites in Australia: A Legal Perspective’ (2000) 11 Public Law Review 117, 123, 127–8.

[51] See generally Williams and Hume, above n 32, 160–1, 175.

[52] A Time for Change, above n 46, 64 [5.46], 65 (Recommendation 11).

[53] Williams and Hume, above n 32, 64.

[54] Australian Government, Budget Measures 2013–14 (Budget Paper No 2, 14 May 2013) 246 <http://www.budget.gov.au/2013-14/content/bp2/download/BP2_consolidated.pdf> Anthony Albanese, ‘Funding Provided To Promote Public Debate about Constitutional Change’ (Media Release, 17 June 2013) <http://anthonyalbanese.com.au/funding-provided-to-promote-public-debate-about-constitutional-change> . Controversially, the promotional funding was allocated in proportion to the votes cast in Parliament on the referendum Bill, with the consequence that $10 million (or 95 per cent of the available funds) was given to the ‘Yes’ campaign.

[55] Nationwide News Pty Ltd v Wills [1992] HCA 46; (1992) 177 CLR 1 (‘Nationwide News’); Australian Capital Television Pty Ltd v Commonwealth [No 2] [1992] HCA 45; (1992) 177 CLR 106 (‘ACTV’).

[56] Lange v Commonwealth [1997] HCA 25; (1997) 189 CLR 520, 560–1 (The Court) (‘Lange’).

[57] Ibid.

[58] Ibid 560; affirmed, most recently, in McCloy v New South Wales [2015] HCA 34; (2015) 325 ALR 15, 18–19 [2], 27 [42] (French CJ, Kiefel, Bell and Keane JJ), 44–5 [118] (Gageler J), 84–5 [303] (Gordon J) (‘McCloy’).

[59] [1992] HCA 46; (1992) 177 CLR 1, 72.

[60] Lange [1997] HCA 25; (1997) 189 CLR 520, 560 (The Court); McCloy [2015] HCA 34; (2015) 325 ALR 15, 18–19 [2], 36–8 [80]–[89] (French CJ, Kiefel, Bell and Keane JJ).

[61] Lange [1997] HCA 25; (1997) 189 CLR 520, 561 (The Court).

[62] [1992] HCA 45; (1992) 177 CLR 106.

[63] Referendum Act s 11(4)(b).

[64] Orr, ‘The Conduct of Referenda’, above n 50, 123, 127–8.

[65] Australian Electoral Commission, Public Information Campaign (20 January 2011) <http://www.aec.gov.

au/elections/referendums/1999_Referendum_Reports_Statistics/Public_Information_Campaign.htm>.

[66] Ibid.

[67] Referendum Act s 122(1).

[68] Evans v Crichton-Browne [1981] HCA 14; (1981) 147 CLR 169, 204, 208 (The Court). The Court was interpreting the equivalent provisions in Commonwealth Electoral Act 1918 (Cth) s 161(e) prior to the Referendum Act. See also the commentary in Orr, ‘The Conduct of Referenda’, above n 50, 124.

[69] Orr, ‘The Conduct of Referenda’, above n 50, 124; Graeme Orr, The Law of Politics: Elections, Parties and Money in Australia (Federation Press, 2010) 148–9.

[70] Referendum Act s 25, sch 1.

[71] Constitutional Alteration (Fair Elections) 1988 (Cth); Williams and Hume, above n 32, 44.

[72] Referendum Act s 24, sch 1.

[73] Orr, ‘The Conduct of Referenda’, above n 50, 126.

[74] Williams and Hume, above n 32, 52.

[75] Orr, ‘The Conduct of Referenda’, above n 50, 126.

[76] Referendum Act s 93(8).

[77] Australian Electoral Commission, Australian Electoral Commission Scrutineers Handbook (Handbook, August 2015) 62 <http://www.aec.gov.au/elections/candidates/files/scrutineers-handbook.pdf> . See also the discussion in Benwell v Electoral Commissioner [1999] FCA 1532.

[78] David John Gow, Clive Bean and Ian McAllister, Australian Constitutional Referendum Survey, 1999 (31 January 2000) Australian Data Archive <https://www.ada.edu.au/social-science/01018-pub> (‘1999 Referendum Survey’).

[79] Some commentators have suggested that exit polls could be used to understand more about how voters approach referendums. See Michael Coper, ‘Judicial Review and the Politics of Constitutional Amendment’ in Rosalind Dixon and George Williams (eds), The High Court, the Constitution and Australian Politics (Cambridge University Press, 2015) 38, 41. This suggestion has some merit; however, exit polls have important limitations, including their high cost and need for brevity relative to other forms of survey delivery: see D James Greiner and Kevin M Quinn, ‘Long Live the Exit Poll’ (2012) 141(4) Daedalus 9. In Australia, any exit poll would have to be supplemented with other survey forms so as to capture the significant minority of voters who are now likely to cast postal and pre-poll ballots, as provided for by Referendum Act pts IV, IVA.

[80] For more details, see Paul Kildea, ‘A Great Big Verandah’? Citizen Participation in the Process of Constitutional Reform, 1991–1999 (PhD Thesis, University of New South Wales, 2011) <http://unsworks.unsw.edu.au/fapi/datastream/unsworks:10347/SOURCE02?view=true> .

[81] For more details, see Rodney Smith et al, ‘The Challenge of Informed Voting in the 21st Century’ (Research Report, Electoral Regulation Research Network, April 2015) <http://www.law.unimelb.

edu.au/files/dmfile/TheChallengeofInformedVotingFinalReport2.pdf>

[82] For discussion on the strengths and limitations of the two general approaches, see Ariadne Vromen, ‘Debating Methods: Rediscovering Qualitative Approaches’ in David Marsh and Gerry Stoker (eds), Theory and Methods in Political Science (Palgrave Macmillan, 3rd ed, 2010) 249; Peter John, ‘Quantitative Methods’ in David Marsh and Gerry Stoker (eds), Theory and Methods in Political Science (Palgrave Macmillan, 3rd ed, 2010) 267.

[83] Smith et al, above n 81, 57.

[84] Ibid 54–5.

[85] Australian Electoral Commission, 1999 Referendum Report and Statistics: Key Results (24 January 2011) <http://www.aec.gov.au/Elections/referendums/1999_Referendum_Reports_Statistics/Key_Results.htm> Australian Electoral Commission, 1998 Federal Election Voter Turnout by Division (27 January 2011) <http://www.aec.gov.au/Elections/federal_elections/1998/voter_turnout.htm> Australian Electoral Commission, 1999 Referendum Report and Statistics: Informal Voting (25 October 2011) <http://www.aec.gov.au/elections/referendums/1999_Referendum_Reports_Statistics/informal_voting.htm> Australian Electoral Commission, Informality (%) House of Representatives and Senate (13 July 2015) <http://www.aec.gov.au/Voting/Informal_Voting/summary.htm> .

[86] Smith et al, above n 81, 66.

[87] For more details, see Kildea, A Great Big Verandah, above n 80, 183–203.

[88] Clive Bean, David Gow and Ian McAllister, Australian Election Study, 1998 (5 February 1999) Australian Data Archive <https://www.ada.edu.au/social-science/01001> (‘1998 Election Study’).

[89] 1999 Referendum Survey, above n 78.

[90] McAllister, The Australian Voter, above n 14, 66.

[91] Ibid 65–71.

[92] Clive Bean et al, Australian Election Study, 2013 (16 January 2014) Australian Data Archive <https://www.ada.edu.au/social-science/01259>.

[93] See, eg, Senate Standing Committee on Employment, Education and Training, Parliament of Australia, Active Citizenship Revisited (1991); Constitutional Commission, ‘Bulletin’, (Newsletter No 5, September 1987); Civics Expert Group, Whereas the People: Civics and Citizenship Education (Australian Government Publishing Service, 1994) 142–50; 1999 Referendum Survey, above n 78; Consultative Group on Constitutional Change, Resolving Deadlocks: The Public Response (Report, 2004) 29; Roy Morgan Research,Anti-terrorism Legislation Community Survey’ (Report, Amnesty International, 2006).

[94] See, eg, Kildea, A Great Big Verandah, above n 80, 294.

[95] In 2014, the misinformed included two government ministers with responsibility for Indigenous affairs, the federal Country Liberal Party Minister for Indigenous Affairs Nigel Scullion and the Queensland Liberal National Party Minister for Aboriginal and Torres Strait Islander and Multicultural Affairs Glen Elmes: see Joanna Heath, ‘Referendum Delay Likely’, Australian Financial Review (Sydney), 16 September 2014, 5; Kimberley Vlasic, ‘A Day To Recall Past Injustices’, Cairns Post (Cairns), 26 May 2014, 8.

[96] Commonwealth, Parliamentary Debates, House of Representatives, 23 August 1967, 384 (Gordon Bryant), quoted in Bain Attwood and Andrew Markus, The 1967 Referendum: Race, Power and the Australian Constitution (Aboriginal Studies Press, 2nd ed, 2007) 58–9.

[97] Rodney Smith, Australian Political Culture (Pearson Education, 2001) 30. This distinction seems to emerge before Australians reach voting age: see Ariadne Vromen, ‘Paul Keating Is the Prime Minister, but Who Delivers the Mail? A Study of Political Knowledge Amongst Young People’ (1995) 30 Australian Journal of Political Science 74.

[98] Martinelli, above n 19.

[99] Ian McAllister, ‘Civic Education and Political Knowledge in Australia’ (1998) 33 Australian Journal of Political Science 7, 14; Kildea, A Great Big Verandah, above n 80, 293–9; McAllister, The Australian Voter, above n 14, 69–70; Smith et al, above n 81, 42–3.

[100] Smith et al, above n 81, 66.

[101] 1999 Referendum Survey, above n 78.

[102] See Kildea, A Great Big Verandah, above n 80, 263–314.

[103] Lupia and Johnston, above n 20.

[104] McAllister, ‘Elections without Cues’, above n 8, 252.

[105] Murray Goot and Terence W Beed, ‘The Referenda: Pollsters and Predictions’ (1977) 12(2) Politics 86.

[106] Murray Goot and Sean Scalmer, ‘Party Leaders, the Media, and Political Persuasion: The Campaigns of Evatt and Menzies on the Referendum to Protect Australia from Communism’ (2013) 44 Australian Historical Studies 71, 84–5.

[107] McAllister, ‘Elections without Cues’, above n 8, 256–8.

[108] Campbell Sharman and Janette Stuart, ‘Patterns of State Voting in National Referendums’ (1981) 16 Politics 261.

[109] See John McMillan, Gareth Evans and Haddon Storey, Australia’s Constitution: Time for Change? (Law Foundation of NSW and Allen & Unwin, 1983) 34; Campbell Sharman, ‘The Referendum Results and Their Context’ in Brian Galligan and J R Nethercote (eds), The Constitutional Commission and the 1988 Referendums (Centre for Research on Federal Financial Relations and Royal Australian Institute of Public Administration (ACT Division), 1989) 105, 109; Brian Galligan, ‘The 1988 Referendums in Perspective’ in Brian Galligan and J R Nethercote (eds), The Constitutional Commission and the 1988 Referendums (Centre for Research on Federal Financial Relations and Royal Australian Institute of Public Administration (ACT Division), 1989) 118, 130.

[110] Goot and Scalmer, above n 106, 86.

[111] Attwood and Markus, The 1967 Referendum, above n 96, 47. See also Bain Attwood and Andrew Markus, ‘(The) 1967 (Referendum) and All That: Narrative and Myth, Aborigines and Australia’ (1998) 29 Australian Historical Studies 267. One commentator has argued that the lack of a ‘No’ case contributed to misconceptions about the referendum question by allowing FCAATSI’s claims to go largely unchallenged: Russell McGregor, ‘An Absent Negative: The 1967 Referendum’ (2008) 5 History Australia 44.

[112] For examples, see Jennifer Clark, Aborigines and Activism: Race, Aborigines & the Coming of the Sixties to Australia (University of Western Australia Press, 2008) 199.

[113] Attwood and Markus, The 1967 Referendum, above n 96, 50.

[114] Goot and Scalmer, above n 106, 84.

[115] 1999 Referendum Survey, above n 78.

[116] Kildea, A Great Big Verandah, above n 80, 185–92.

[117] ABC News, ‘Front-Bench Anger over Hockey’s Push for a Republic’, AM, 27 August 2015 (Peta Donald) <http://www.abc.net.au/am/content/2015/s4300866.htm> .

[118] Greg Craven, ‘Referenda, Plebiscites and Sundry Parliamentary Impedimenta’ (2005) 20(1) Australasian Parliamentary Review 79, 83.

[119] However, Hay also cautions against too heavy a reliance on the distinction: Hay, above n 13, 158–60.

[120] Stoker, above n 13, chs 9–11; Henry Milner, The Internet Generation: Engaged Citizens or Political Dropouts (Tufts University Press, 2010) chs 6–10.

[121] This fear influenced the Howard Government’s decision to seek the suspension of expenditure limits in 1999: Commonwealth, Parliamentary Debates, House of Representatives, 11 March 1999, 3762 (Daryl Williams, Attorney-General).

[122] Of course, this notion of a level playing field would be undermined should public funding be allocated unequally between the ‘Yes’ and ‘No’ campaigns, as occurred in 2013: see sources referred to in above n 54.

[123] This paragraph draws on suggestions made in Paul Kildea and George Williams, ‘Reworking Australia’s Referendum Machinery’ [2010] AltLawJl 6; (2010) 35 Alternative Law Journal 22.

[124] A Time for Change, above n 46, 57–8.

[125] Cal Elections Code §§ 9040–96 (West 1994). In NSW, an information pamphlet containing summaries of the ‘Yes/No’ cases is ‘prepared by public servants’, and then ‘vetted’ by constitutional lawyers and other experts ‘to ensure its fairness’: Anne Twomey, The Constitution of New South Wales (Federation Press, 2004) 320.

[126] Referendum Act s 11(1).

[127] This reform was recommended by A Time for Change, above n 46, 58 (Recommendation 5).

[128] Smith et al, above n 81, 62–3.

[129] Referendum Act ss 11(2C), (4)(ac).

[130] Galligan, A Federal Republic, above n 24, 130.

[131] Smith et al, above n 81, 59–60.

[132] Such issues also arise with respect to proposals to introduce so-called ‘truth in political advertising’ laws during election campaigns: Anne Twomey, ‘Truth in Political Advertising’ (2010) 21 Public Law Review 232.

[133] See further Kildea and Williams, above n 123; Williams and Hume, above n 32, 243–4.

[134] Political Parties, Referendums and Elections Act 2000 (UK) c 41, ss 104(1), (2).

[135] Electoral Commission, Referendum on the UK Parliamentary Voting System: Report of Views of the Electoral Commission on the Proposed Referendum Question (Report, 2010) 24; Electoral Commission, ‘Scottish Referendum’ (8 February 2016) <http://www.electoralcommission.org.uk/find-information-by-subject/elections-and-referendums/past-elections-and-referendums/referendums/scottish-referendum> Electoral Commission, ‘EU referendum question assessment’ (8 February 2016) <http://www.electoral

commission.org.uk/find-information-by-subject/elections-and-referendums/upcoming-elections-and-referendums/eu-referendum/eu-referendum-question-assessment>.

[136] Electoral Commission, ‘Scottish Referendum’, above n 135.

[137] Cheryl Saunders, ‘Ways and Means of Constitutional Reform’ in Clement Macintyre and John Williams (eds), Peace, Order and Good Government: State Constitutional and Parliamentary Reform (Wakefield Press, 2003) 10, 11.

[138] Cheryl Saunders, ‘How Important Was the Convention?’ (1998) 21 University of New South Wales Law Journal 868, 869.

[139] Helen Irving, ‘The People and Their Conventions’ in Michael Coper and George Williams (eds), Power, Parliament and the People (Federation Press, 1997) 113, 124–5. See also Williams, who argues that Australians are likely to overcome resistance to constitutional change if they are given more meaningful opportunities to participate in the reform process: George Williams, ‘Thawing the Frozen Continent’ (2008) 19 Griffith Review 9, 18.

[140] See Paul Kildea, ‘Expert Panels, Public Engagement and Constitutional Reform’ (2014) 25 Public Law Review 33, 36–8, 42.

[141] Ibid 45–6.

[142] See Levy, ‘Deliberative Voting’, above n 18, 568–9; Levy, ‘Breaking the Constitutional Deadlock’, above n 27, 808.

[143] Issues Deliberation Australia, Australia Deliberates: A Republic – Yes or No? (Final Report, 1999) 18–19.

[144] Paul Kildea, ‘A Little More Conversation? Assessing the Capacity of Citizens To Deliberate about Constitutional Reform in Australia' (2013) 22 Griffith Law Review 291, 302–3.

[145] Graham Smith, Democratic Innovations: Designing Institutions for Citizen Participation (Cambridge University Press, 2009) 94–101.

[146] Williams and Hume, above n 32, 242–3.

[147] A Time for Change, above n 46, 60 (Recommendation 6). Similarly, a 2004 Senate inquiry into republic reform recommended the establishment of an ongoing education and awareness program, overseen by a dedicated standing parliamentary committee: Senate Legal and Constitutional References Committee, Parliament of Australia, The Road to a Republic (2004) 134 (Recommendations 1 and 2).

[148] Evidence to Senate Legal and Constitutional References Committee, Parliament of Australia, Canberra, 13 April 2004, 34 (John Warhurst).

[149] Smith et al, above n 81, 56–9.

[150] A Time for Change, above n 46, 59–60.

[151] McAllister, The Australian Voter, above n 14, 68–71.

[152] Greg Craven, Conversations with the Constitution: Not Just a Piece of Paper (UNSW Press, 2004) 9.

[153] Craven, ‘Referenda, Plebiscites and Sundry Parliamentary Impedimenta’, above n 118, 83.