Democracy without Politicians: Could it Work in Northern Ireland? More Events at QUB’s Spring Festival

GarryThe Spring Festival of the Institute for the Study of Conflict Transformation and Social Justice at Queen’s continued yesterday with a discussion on ‘Randomocracy: An Alternative to Elections and Voting,’ led by Prof John Garry.

Some Slugger readers may have seen the short video produced as part of Garry’s research project, which summarizes the result of his experiment of asking random citizens to deliberate on the flags issue.

Garry showed the video at the event and then provided further information about the research methods and rationale.

A number of events still remain in the Festival, including lectures by First Minister Arlene Foster and former Taoiseach Bertie Ahern. (Registrations for the Arlene Foster event have closed; while registrations for the Bertie Ahern event close on Friday.)

Garry opened his remarks by pointing out that the idea of random citizens making decisions might seem ‘crazy’ to us today. But for the ancient Greeks, randomness was the norm and elections were considered ‘bonkers.’

He cited the example of the Constitutional Convention in the Republic of Ireland, where 66 randomly selected citizens and 34 politicians took part in a deliberative process – one result of which was a recommendation to hold a referendum on gay marriage. That referendum was held and passed overwhelmingly last year.

‘From the perspective of 1980s Ireland, that would have seemed science fiction-esque,’ Garry said.

This is how Garry described his Northern Ireland-based experiment:

Taking the issue of flag display as an example, what would happen if a representative sample of citizens became informed about the issue of flag display, if they considered the arguments from all perspectives and seriously reflected on the issue? Would their views become more centrist and compromise-oriented or perhaps more hardline? Would there be an increase in the chances of reaching a resolution on the issue?

To find out, we conducted a real world experiment on a representative sample of over 1000 Northern Ireland citizens. In order to inform citizens about the background to the issue of flag display we created a short film clip which provided an overview of the issue. In order to ensure that citizens heard a balanced set of arguments on all sides of the issue we created a further film clip providing this balanced set of perspectives. In order to encourage citizens as much as possible to really think about where different people are coming from on this contentious issue, we asked citizens to imagine that they were having a conversation with someone (from the ‘other’ community) about it in which the various arguments on the issue were being debated.

Citizens were put into various groups: a control group that got no information, a group that got information, a group that got information and the summary of arguments, a group that got information and imagined dialogue, and a group that got information, the summary of arguments, and imagined dialogue.

Those at the event had the opportunity to watch the videos used in the experiment.

Those who got the information, the summary of arguments and imagined dialogue were the most likely to come to a ‘compromise’ position (flying the flag on designated days), at 59%. This differed the most from the control group, where 48% of citizens chose the compromise position.

Garry said:

‘What these findings suggest is that when citizens engage in informed reflection on the flag display issue their views do change, and furthermore their views change in the direction of the compromise option.’

However, Garry was quick to point out that his experiment was not designed to manipulate people into ‘compromise’:

‘The point is not to shuffle people to a compromise position. The process is the point – it is the process of deliberation that matters, not the outcome.’

He noted that his experiment was rather unique among other studies of ‘deliberative’ democracy, where people are usually encouraged to talk with each other as a form of reflection. Garry specifically chose internal, ‘imagined dialogue’ because he felt that external dialogue would ‘contaminate’ the randomness of the sample, undermining his statistical ability to infer to the wider population.

In addition to the sample of citizens who participated in the experiment, Garry conducted surveys of the general public and MLAs to gauge support of ‘deliberative democracy responses to resolving contentious issues.’ Although it could not be assumed that either of these groups would be familiar with the idea of a ‘Citizens Assembly’, 65% of the general public said this would be a ‘good idea.’ Results are summarized below:

Question wording: On some important issues – such as flag display and the issue of welfare reform – the political parties in Northern Ireland find it very hard to agree with each other, and this leads to political crises. When such a crisis happens, there may be a number of ways to try and resolve it. Please tell me to what extent you think the following approaches is a good idea or a bad idea.

% Saying Good Idea

General PublicMLAs
Get the British Government to come up with a solution427
Get the British and Irish governments working together to come up with a solution5426
Get the British Government and the Northern Ireland parties to come up with a solution6661
Get the British and Irish governments and the Northern Ireland parties to come up with a solution6467
Get someone from outside Britain and Ireland, such as a politician or diplomat from the US, to chair talks between the Northern Ireland parties and come up with a solution2838
Hold a referendum on the issue so that the people can directly decide6125
Have an immediate election to try and resolve the issue257
Get a cross section of ordinary citizens on a Citizens Assembly to learn about the issue, listen to a presentation of the main arguments and then reach a decision on the issue6517


Although Garry was quick to say that the point of his talk was not to argue that deliberative democracy is necessarily a good idea or would quickly resolve contentious issues, he did highlight some of its potential advantages:

  • Representative voice of the people
  • Addresses problem of apathy
  • Addresses problem of bringing young people into politics
  • More women in politics
  • Evidence suggests deliberation leads to a greater chance of compromise
  • Evidence suggests public supports the idea
  • Evidence suggests that MLAs do not see deliberative democracy as making a final decision – but could have value in making recommendations

So, democracy without politicians – could it work in Northern Ireland?

Read a previous guest post on Randomocracy by John Garry.

  • Kevin Breslin

    Isn’t there a referendum coming up?

  • Croiteir

    And so the Marxists have come up with another idea to remove the power to select your ruler, this time it is randomocracy.

    Erik Olin Wright, one of the proponents of this system, under his Real Utopias Project, wants to use it to forward a marxist policy as the knows that the ends which he wants to deliver will never be achieved by electoral politics were his ideas have been debated and rejected. So, in order to bring his utopia – and he does call this this and related strategies Real Utopia – he wants to remove the debate element of democracy.

    To show this his article ends with a brief discussion on the problem of transformation which requires, he says, an ‘understanding of how strategies of transformation have long-term prospects for eroding capitalist power relations and building up socialist alternatives’ (Wright 2013: 20).

    These strategies of transformation he divides into three broad categories:
    ruptural, interstitial and symbiotic. Wright explains that ‘ruptural strategies are
    most closely associated with revolutionary socialism and communism, interstitial
    strategies with some strands of anarchism, and symbiotic strategies with
    social democracy’

    He wants to intertwine the interstitial and the symbiotic in order to deliver the outcome envisaged by those who use the ruptural method.

    The radical left, Wright concludes, should broaden its view of power, particularly political power, and embrace interstitial strategies as these strategies show that ‘another world is possible by building it in the spaces available, and then pushing against the state and public policy to expand these spaces’ .

    The undermining of representative democracy by marxists is no longer conducted by means of activists, it is doe by trying to erode real debate, debate that Marxists have lost.

  • Brian O’Neill

    Under that logic do we have a Marxist judiciary system? As you know jury panels are selected at random.

  • Croiteir

    hardly Brian

  • chrisjones2

    In my experience many people in Belfast are random …but some are even more random than others

    The problem is though if the random person comes up with a great solution but the people causing the problem dont agree its a great solution what happens then?

  • Harlequin

    Nonsense comment if you’d read the piece or watched the video. Strange that you took the time to randomly copy and paste from Conor McCabe’s article in the Irish Journal of Sociology.

  • Croiteir

    I used the argument to back my point. Nothing g strange about it. It is a Marxist answer to failure. Even Marxists know that it is a nonsense

  • MainlandUlsterman

    I have some experience of deliberative workshops, including running a few. My own personal view is (not speaking for anyone else here including any employers or clients), don’t expect people to be able to make decisions for government this way – they are great for flushing out the issues and seeing what people’s thought processes around them are, to add into the policy-maker’s mix of considerations, but they are not suited to unmediated decision-making.

    Consulting the public creatively makes a lot of sense, big fan of it. But passing the subsequent research analysis process and policy-design to the public also is pretty impractical and I think raises all sorts of democratic issues. It also raises big issues with how government should go about listening to the public intelligently and not just adopt two-dimensional, literal interpretations of what people say and do in these sessions. Intelligent analysis pulls out what people mean, not just what they say. It notices apparent contradictions in people’s views, for example, and attempts to resolve them, so we get what their real point was.

    Other big issues:
    (1) sample design / recruitment criteria – any sampling for qualitative research (my specialism) cannot be representative, we talk about qual sampling as aiming to be ‘indicative’, which is different – we can only ever be selective in qual sampling, which is fine for our qual research purposes because of the nature of what comes out (meaningless without interpretation or ‘sense-making’) but becomes problematic if being used to ‘represent’ wider public in a straightforward way. You need quantitative sampling for that – bigger numbers and a wholly different, statistical approach. Usually the two are done in a complementary way; neither can really replace the other;
    (2) linked to this, it is really bad practice to pull out percentages on small sample sizes. Qualitative research (going in depth with small samples) is not about measuring. It’s a mind-bender at first if you’re not used to qual as it’s such an instinct to ask ‘how many people said that?’ but you have to not think like that with qual.

    Point (2) is the thing that most trips up non-qual researchers (including senior clients and experienced users of research) when using qual research. Qualitative research (and deliberative workshops being small numbers, in-depth and ‘participant-led’, generally work to qual paradigms) developed largely out of methods and ideas used in psychology and anthropology. It is good at understanding and describing what is going on with people beneath the surface and in multiple strands and at building hypotheses of how people think and behave. But it can’t measure how many people do thing x over thing y, or have opinion a over opinion b. You need other tools for that, involving getting a much larger number of people and presenting them with distinct choices. On that, you can then count and give percentages.

    What I’m seeing here is a kind of halfway house approach which tries to do quant things with qual samples and methods. And it’s presenting results unmediated as if they reveal real views and choices of the public. But maybe I need to hear more about the methodology before making a definitive judgment! I’m just very wary of the way it’s being suggested public consultation should be done here.