Australia should run a citizens’ assembly to inform the referendum on an Indigenous Voice to Parliament
Referendums ask people to give their views on complex questions off the top of their head and with a bit more time and information, they often change their mind
Prime Minister Albanese has pledged a referendum on a Voice for First Nations people, a “body enshrined in the Constitution that would enable Aboriginal and Torres Strait Islander people to provide advice to the Parliament on policies and projects that impact their lives.” For non-Australian readers, a referendum returning a double majority of both voters and states is required to enact constitutional change in Australia. Only eight of 44 referendums in Australia’s history have been successful, the most recent of these successes occurring in the 1970s.
But if the framers of the Constitution opted for this then nascent method as the ultimate democratic safeguard, it’s not serving this democratic purpose. We know from years of international experience that referendums don’t typically elicit people’s informed views. The government, or civil society if government won’t, should run a citizens’ assembly in the lead-up to the referendum on a Voice to inform people’s vote. The assembly would produce a one-page statement of facts and arguments for and against the government’s reform proposal, in the nature of the “Citizens’ Statement” produced by the Citizens’ Initiative Review process that has been run in several US states in the lead-up to a ballot initiative. The assembly process would have the features of a standard mini-public: amongst others, random selection, presentation of a range of balanced evidence and testimony, and lengthy participant deliberation.
Voters should pay attention to the findings of a citizens’ assembly
Political scientists agree that referendums are messy and dangerous tools. When asked whether they were a good idea, Michael Marsh, a political scientist at Trinity College Dublin said “The simple answer is almost never.” (I’m stealing this gem from a New York Times article I’d recommend, explaining “Why Referendums Aren’t as Democratic as They Seem). Without time or resources to pore through and fact-check all the available arguments on complex issues, voters take problematic “short-cuts.” They look to authority figures they trust, fit the choice within a familiar narrative or based it on some set of issues or events that aren’t related to the subject of the referendum. These narratives have already been playing out in relation to the Voice. And it’s hard to dispute that the issue of a Voice to Parliament raises plenty of complexity, making those narratives all the more appealing.
But if everyone cast a perfectly informed vote, the idea of a say for every Australian in whether the constitution should be changed is a worthy one. What if you could combine this democratic ideal with time, information and deliberation? The citizens’ assembly is the closest tool we have for this, using random selection (or stratified random selection) to generate representativeness as far as possible. Of course, there’s no substitute for a direct vote, but substitution isn’t and can’t be what’s on the table – the proposal is to run an assembly in the lead-up to the referendum as a way of informing the referendum result.
When people are given time and resources to understand and deliberate on an issue, they routinely change their initial view. Professor James Fishkin has conducted deliberative polling experiments over 100 times in 29 countries on issues ranging from unification of Korea and South Korea to social policies affect Roma communities and controversial electoral reform proposals in the UK. There are consistently dramatic, statistically significant changes in people’s views before and after deliberation. In fact, while Australian voters rejected a constitutional amendment to make Australia a republic 55-45 in 1999, just weeks earlier, 73 percent of voters in a little-known deliberative poll voted Yes. This represented a massive 16 percentage point increase on the pre-deliberation poll. These opinion changes are of course not restricted to deliberative polls but are replicated across hundreds of mini-public processes that have been run globally.
At least some voters are likely to pay attention to the findings of a citizens’ assembly
But what’s the point if no-one trusts or listens to the results of a citizens’ assembly? In the Australian Republic example above, the deliberative poll clearly didn’t have much impact.
Let’s start with the evidence we have from other international examples of mini-publics run in the lead-up to a direct democratic vote (referendum or ballot initiative). Do mini-publics influence the way the public votes? The evidence is pretty scant and academics perhaps unsurprisingly say “it depends.” For example, analysing three Citizens’ Initiative Review processes run in Oregon, California and Massachusetts, they find that two of these processes significantly influenced voting choices but the latter didn’t. All of the studies we have explore different and certainly not all aspects of what mattered to gaining voter trust.
What we don’t have much evidence on in particular is the importance of investing in advocacy of assembly results, especially advocacy that is led by participants themselves, supported by leaders from civil society and ideally also from a politically diverse group of elites; perhaps surprisingly, it has not been a big feature of many processes run to date. The potential benefit of this investment seems intuitively obvious. The fundamental difference between this kind of advocacy and other advocacy campaigns? For understandable, politicians and partisan groups will inevitably find it hard to reach beyond their partisan audience.
Even without “advocacy” of the process, around 60% of an Australian sample presented with a short description of mini-publics said they would probably or definitely support the idea. Thirty percent were on the fence, saying they needed a little more information/evidence that it works. What if in advocating the results of an assembly, Australians were given this information and evidence? Proposals developed by two Canadian Citizens’ Assemblies on electoral reform were rejected by the citizenry and the authors hypothesize that this was “mainly because elite signals and media attention were lacking.”
The bottom line? Whether or not voters listen depends on whether a good or bad job is done of sharing findings with the general public, and not enough attention has been paid to this critical role in processes run to date.
Inevitably not every Australian would pay attention to the results, even with a strong advocacy campaign. But even if a small number of us cast our vote on the basis of better information we learned from an assembly, the process seems to me a worthy investment in strengthening our democracy.
I had previously had the same thought regarding a citizen jury on this topic. You have expressed the case exceptionally well. Have you considered how to reach a wider audience with your pieces?