What the record reveals of the chances of Indigenous recognition

Recent opinion polls have given a boost to the campaign for Indigenous constitutional recognition. Prime Minister Tony Abbott has flagged a possible referendum date in 2017. At the same time, many proposals for the form of constitutional change are in competition. None as yet commands general support among opinion leaders. This must be a matter of major concern.

If the referendum goes ahead, any proposal will have to win a majority of votes across the nation plus majorities in a majority of states. As is well known, the referendum record is marked by failures: 44 proposals (in 19 referendums) have delivered only eight successes.

The government’s Expert Panel on Constitutional Recognition of Indigenous Australians was deeply troubled by this fact. Its 2012 report prioritised strategies for success. However, like many supporters today, it paid only modest attention to the referendum record.

The panel concluded that bipartisanship, a well-designed campaign and a large-scale public education program could deliver success. Sadly, such optimism is questionable. An assessment of the chances needs to be tempered with realism.

A referendum history dominated by rejection

What does the record tell us? First, to succeed, a proposal needs soaring levels of support and virtually no organised opposition. Any significant degree of opposition will almost certainly mean defeat.

Parliamentary Education Office

Bipartisanship is essential, but it will not be enough. Four referendum questions (1919, 1926, 1937, 1967) were backed by both major parties but were defeated. Political opposition, including from minority parties or independent MPs, is likely both to indicate and encourage voter opposition.

Parliament must pass a Constitution Alteration Bill before a referendum proceeds. Members of parliament are entitled to prepare the official “Yes” and “No” cases that reflect their vote.

The cases are then printed in pamphlets and sent to voters. These serve as an important indicator. If no-one in parliament votes against the bill, there will be no No case. Every referendum question in the past with only an official Yes case has succeeded.

The likelihood of this happening for Indigenous recognition looks slim, at least at present. Government Senator Cory Bernardi has already indicated that he may trigger a campaign against any proposal.

Only one referendum with a No case – the 1946 social services referendum – has ever succeeded. That referendum was the narrowest of the eight successes, but it offers a glimmer of historical hope, standing as the single victory in the face of divided political opinion. It should therefore be the focus of attention.

The 1967 referendum is an unreliable precedent

Instead, the spectacularly successful 1967 referendum, which attracted more than 90% of the national vote and support in all states, has consistently been treated as a precedent.

A repeat of what happened in 1967 is unlikely. That referendum has been poorly understood. The specific question put to the voters on the ballot paper – Do you approve the proposed law for the alteration of the Constitution entitled: “An Act to alter the Constitution so as to omit certain words relating to the people of the Aboriginal race in any state and so that Aboriginals are to be counted in reckoning the population?” – bore only a roundabout resemblance to the actual alterations to be made to the Constitution, and almost no resemblance to what people seem to think happened.

The referendum did two things, having little in common with today’s recognition aspirations. It gave the Commonwealth the constitutional power to make “special laws” (beneficial or adverse) for the people of the Aboriginal “race” and it deleted a section of the Constitution (s 127) that had excluded Aborigines from “reckoning” the people of the Commonwealth.

The purpose of this particular section (admittedly not obvious from its words) was to guide the calculation of the numbers of parliamentary representatives and per capita financial entitlements to which each state was entitled.

The erroneous framing of the 1967 referendum as giving Indigenous Australians equal rights surely helped it succeed.AAP/National Library of Australia

However, many myths surrounded the referendum, and these have endured. Among others, it is said that it gave “equal rights” to Aboriginal Australians. It did not. No rights were added or altered.

Specifically, it did not give Aboriginal people the right to vote. By 1967, they already voted in all states and in the Commonwealth – they voted in the referendum.

The deletion of section 127, it is also said, meant that Aborigines could be counted in the census. It did not. They have always been counted, in every census.

The belief that the 1967 referendum was about indigenous citizenship and equal rights appears, however, to have helped the campaign. A similar narrative will be hard to achieve for Indigenous recognition, unless the question is meticulously framed to avoid the impression of favouring one sector’s rights over others.

Proposing rights is no guarantee of success

Even then, a rights proposal cannot be certain of success. Indeed, the worst referendum defeat on record was the 1988 “rights and freedoms” proposal. It attracted only 30.79% of the national vote. This result alone should caution against any proposal to insert an anti-race discrimination provision in the Constitution.

The 1988 referendum and several others also sound warning bells against including multiple proposals in a single question. Voters are unlikely to warm to an “all or nothing” proposal.

A large-scale education campaign will not necessarily help. Unprecedented public funding was allocated for the 1999 republic referendum: nearly A$17 million for the official referendum pamphlets, an additional $4.5 million for public information and a further $15 million for campaign advertising. Voters were possibly better informed than in any previous referendum. Their answer was still resoundingly negative.

In any case, the education strategy confuses information with opinion. Constitutional knowledge provides little guidance about the desirability of constitutional alteration. Constitutional lawyers have never been unanimous on the merits of referendum proposals. Knowledge may lead some voters to reject a proposal.

We don’t know much about what Australians know or think about the Constitution. There is little recent data.

Polls conducted in 1987 and 1994 revealed a very low level of knowledge about the Constitution. These are frequently cited and appear to encourage the case for more education. However, they are now seriously out-of-date. A modern referendum campaign needs much better data and cannot rely on small, unfocused samples of opinion.

Social media: a new wildcard

In assessing the prospects for success, one critical variable, never trialled before, must be confronted. Although the last referendum, in 1999, took place in the internet era, it pre-dated interactivity, social networking, mass emailing, blogging, tweeting and texting. These new media, as the Expert Panel noted, create huge opportunities for educating and harnessing support and are of particular significance to young voters.

But there is another side to the picture. Social media also offer formidable avenues for opposition. Campaigns, furthermore, can be anonymous and allow things to be said that would be unacceptable if openly expressed. The disproportionately negative comments attached to online opinion pieces on Indigenous recognition provide a stark illustration of the latter.

A defeat for Indigenous constitutional recognition would be disastrous and demoralising. Believing that worthy proposals will attract support because they are worthy and “talking up” the chances of success are profoundly misguided.

Supporters must look long and hard at the referendum record. If not, they will hand the outcome over to the opposition.

Helen Irving does not work for, consult to, own shares in or receive funding from any company or organisation that would benefit from this article, and has no relevant affiliations.