Tolerance, trust and truth: what students can learn from Australia’s Voice referendum debates

Share this Article

This article was first published on Geoff’s website and is republished here with permission.

In his recent speech to the National Press Club, Australian National University vice-chancellor Brian Schmidt called on Australia to “enlighten up” – though not in those words. He reflected on how, with topics such as the Voice referendum, “democratic debates generate more heat than light”. Pointing to the way “media circuses” tend to amplify misinformation and polarisation, he called on the media to work with universities on rebuilding “trust and truth”:

“I would argue that democracy cannot function effectively without evidence and knowledge … part of any rounded education must be inculcating values our society supports, such as tolerance, respect, and rationality…”

So (again) let me ask: what does “tolerant and rational” debate look like? How do we recognise those who practise it (and those who don’t)?

My chart below sets out 6 “group tasks”. These align with many campus policies in Australia. They echo longstanding “academic” norms that universities in the UK, the US and other liberal democracies aim to promote. And they recall a hard-won Enlightenment principle: the importance of tolerance in multi-cultural, multi-faith societies.

Australia’s Voice referendum: a case study

Mapping the claims and counterclaims of the Voice debate onto this framework illustrates how wider adoption of “constructive disagreement” norms could help to rebuild cultures of “tolerance, trust and truth” in Australian public discourse.

Chart 1. Six group tasks to support “tolerant and rational” debate

The first task (make it visible) presumes two-way tolerance: for both sides to navigate the moral complexities of any major disagreement, it must first be OK to disagree.

If only one side speaks up, other points of view become less visible – as does the common ground between them. For example: Voice referendum protagonists Noel Pearson (for Yes) and Jacinta Price (for No) both reject “passive welfare” models for Indigenous communities. Both endorse “real” economic participation. Both are critical of “wasteful” government spending. As Pearson puts it, they’re in “furious agreement” on the urgency of the problem. But not on the Voice as a solution.

The second task (keep it rational) is to confirm relevant facts. This helps to focus discussions on the substantive issues – and to combat the spread of misinformation.

For example, many Australians think governments already spend $30-$40 billion a year on Indigenous programs. In fact, less than 20% is spent on Indigenous-specific programs – the rest on support that all Australians rely on, such as health care. That said, the Productivity Commission put total average spending per Indigenous person at twice the non-indigenous average. This reflects greater use of government services, higher remote delivery costs – and in some areas, wasteful spending.

The third task (keep it open-minded) is to understand the reasoning behind competing views of key issues. Here “translation” problems loom when key phrases refer to different contexts or subtexts.

For example: a No camp claim was that the Voice “puts race in the Constitution”. A Yes camp response was: “race is already there”. Parliament (still) has the power to make race-specific laws. However, this fact didn’t address No voter concerns about creating “unequal” constitutional rights for one group. Unlike the call to “equalise” Indigenous rights in the Constitution in the 1967 referendum (which won 90% of all votes), for many the “democratic equality” factor here was unclear. 

The fourth task (keep it fair) is to recognise that on some points your opponent might be right (or partly right).

For example: the Yes camp highlighted how “modest” the proposal was, in that the Voice would have no formal “veto” power. The No camp accepted this; but predicted it would have indirect power to block or delay government decisions. And in media commentary, one Yes side expert (law professor Greg Craven) agreed: he opposed the inclusion of Voice advice to “Executive Government” in case it “meshed” government departments in endless demands and legal challenges.

But then, as there would still be no formal requirement to seek or to follow Voice advice, other law professors on the Yes side (Anne Twomey and George Williams) saw very little risk of any “deluge of legal action”.

But then, Professor Williams also acknowledged that the High Court could direct departments to consider “relevant” Voice advice (not sought but raised anyway); and to “remake” (confirm or vary) prior decisions.

The fifth task (keep it inclusive) is to refrain from abuse – say, by labelling No proponents “racist” or “stupid”.

Of course, racism and ignorance exist in Australia – as in every other liberal democracy. But in polling before October 14, the top three reasons voters gave to choose No didn’t indicate widespread “racism”. They reflected concerns about equal citizen rights (“It divides us”), model design (“There are no details”) and practical benefits “It won’t help Indigenous Australians”. An ANU survey after the referendum (cited below) supported this reading: it didn’t find widespread dislike or indifference to Indigenous disadvantage. Most voters accepted the need to do more – but rejected the model.

Leaving aside the practice of civility as a democratic virtue, simply labelling opposing views on a matter of social policy “deplorable” – or those inclined to support them as “dinosaurs or dickheads” – is no way to widen voter support for a faltering political campaign. Nor were the presumptuous predictions that unless “Yes” prevailed, Australia would be seen as a “small-minded nation“, or a “frightened resentful people … prepared to punish Indigenous people“. Typically, this style of rhetoric creates what psychologists call a “persuasive boomerang”: the disdainful subtext of the messaging backfires, hardening resistance to the cause and its promoters.

The final task (keep it in context) is to protect participants from social sanctions for taking the “wrong” view on a contentious topic. This is particularly important for student learning, where discussions should consider minority views as well as mainstream ones.

A recent student survey indicates that fear of peer criticism on social media is a key reason US students self-censor in class. Creating low-risk forums to work through points of disagreement allows participants to exercise the intellectual virtues of “charity” and “humility” promoted by moral philosophers (and groups such as Heterodox Academy).

This presumes the two-way tolerance mentioned earlier – an essential precondition for open, honest debate in any social context. Instead of “scoring points” to win the debate, participants can gain by exchanges that allow for uncertainty and offer room to reflect. This makes it easier to self-correct and perhaps reconcile viewpoints – or at the very least, “agree to disagree”.

Conclusion – and next steps?

As we know, the call to establish an Indigenous Voice in the Constitution won many hearts and minds. But in polling, voter support declined steadily in the months leading up to October. In the end, Yes came to grief with six in ten Australians voting No.

Chart 2 Sources: Simon Jackman and ABC News (ABC 8 October). Resolve Political Monitor (Age, 9 October). Evershed and Nicholas (Guardian, 25 September). Adrian Beaumont (Conversation, 25 September)

A sad irony of what became a very partisan referendum debate is that on key aspects there was no fundamental disagreement. As prime minister Anthony Albanese told Parliament in August:

“Both sides of parliament are saying they support constitutional recognition, and both sides of parliament are saying they support a legislated voice”.

In the Age/Sydney Morning Herald on October 14, politics journalist David Crowe explained the result – and a flawed Yes campaign – this way:

“Beware the sweeping assertions that this makes Australia a backward and racist country … (or) that there was no way to avoid a partisan fight. Of course there was a way. One path was to propose Indigenous recognition in the Constitution without a Voice – an option that many Liberals endorsed and Dutton said he would support. Another was to put two questions on the same day: one for recognition, one for the Voice. Indigenous leaders … were not willing to compromise. They insisted the Voice was practical recognition, chaining the two concepts together. This has sunk them both…“

In his Press Club speech Professor Schmidt referred to new ANU research on the factors that led to the referendum result. That research confirmed that most No voters didn’t oppose “the idea of constitutional recognition in general”, and that a “symbolic” form of recognition may have succeeded. However, voters “didn’t want division and remain sceptical of rights for some Australians that are not held by others … They did not see the Voice model put to them as the right approach to remedy … disadvantage”. (See Notes below.)

Despite the referendum setback, Indigenous leaders still want a national body to co-ordinate advice to policy makers. In time, a legislated voice-type body – or scaled-up local and regional bodies – seems feasible, although without the “symbolic” recognition a constitutional Voice body would offer.

The failure of the Yes campaign, despite wide political support for better efforts to Close the Gap, may be seen as a “teachable moment” in Australian political history.

As Professor Schmidt suggests, universities and media groups have key roles to play in addressing the often-toxic dynamics of democratic debates. They can practice and promote more constructive modes of discourse that examine substantive issues, and so better inform the public on many urgent debates on challenging topics.

For this, a key premise is the one Voltaire prescribed to end the religious wars of 18th century Europe. The point of tolerance is not to reach full and final agreement, but to maintain social space for good-faith disagreement. Without this, societal cultures of “trust and truth” give way to factionalism and fanaticism.

Discussion Paper Chart 21 Image source: Wikimedia Commons


1. The “six tasks” at Chart 1 are inspired by my attempt earlier this year to translate Heterodox Academy guidance into a simple presentation slide for students, to show how to debate difficult topics. As the School of Athens imagery here suggests, there are many overlaps between “keeping it academic” and “keeping it democratic”.

2. Here’s an excerpt from the ANU survey research findings (Nicholas Biddle et al):

“There is no evidence in the data to suggest that Australians are against the idea of constitutional recognition in general. Of those who were willing to give an opinion as to what their vote would have been if the referendum was just about recognition those who would vote yes outnumber those who would vote no by a margin of almost five-to-one. The vote also did not signal a lack of support for reconciliation, for the importance of Aboriginal and Torres Strait Islanders having a voice/say in matters that affect them, truth-telling processes, or for a lack of pride in First Nations cultures. All of these notions were supported in our survey by around eight-in-ten Australians … The data suggests that Australians voted no because they didn’t want division and remain sceptical of rights for some Australians that are not held by others. The data suggests that Australians think that Aboriginal and Torres Strait Islander Australians continue to suffer levels of disadvantage that is both caused by past government policies and that justified extra government assistance. They did not see the Voice model put to them as the right approach to remedy that disadvantage … Australian voters are … not uniquely conservative. Australia’s referendum experience can be explained in part by reference to what appears to be three explanatory factors in other constitutional referendums, namely the need for a double majority, the use of compulsory voting, and the absence of bipartisanship … (and) specific characteristics of the 2023 Voice referendum … meant a yes vote was going to be a challenge. First, celebrity endorsement(s) usually depress the yes vote … The use of business and celebrity endorsements in Australia in 2023 may have provided voters a cue that the elites and the rich and famous were in favour of the Voice and, by implication, that this was against their interests. Second, in contrast to the 1967 referendum to include Indigenous Australians in the census—which was carried by a record majority—the Voice was seen as creating rights for small minority that were not held by others…”

3. For a more detailed discussion of the Voice referendum, with further links to reports and commentary, see my September post, Six ways to re-enlighten” Australia’s Voice referendum debates.”

Share this article

Got something to add? Join the discussion and comment below.


  • Geoff Sharrock

    Dr Geoff Sharrock has a PhD in university leadership and management. He has worked at the University of Melbourne's Centre for the Study of Higher Education as a program leader, and at Monash University as a higher education policy adviser. In consulting work he has advised universities on internal change management, and on their policy responses to the 2019 French Review. In 2022 he ran a project on free inquiry and viewpoint diversity in Australian higher education, funded by Heterodox Academy. He is an Honorary Senior Fellow with the University of Melbourne's School of Psychological Sciences.

Follow Us

Join our Newsletter