Whether, and to what extent you think a crowd can be wise has implications for the kinds of engagement you might advocate.
Democratic theory has tended to take a pretty dim view of people and their ability to make decisions. Many political philosophers believe that people are at best uninformed and at worst, ignorant and incompetent. This view is a common justification for our system of representative democracy – people can’t be trusted to make decisions so this responsibility should fall to those who have the expertise, knowledge or intelligence to do so.
Think back to what Edmund Burke said on the subject in his speech to the Electors of Bristol in 1774, “Your representative owes you, not his industry only, but his judgement; and he betrays, instead of serving you, if he sacrifices it to your opinion.” He reminds us that “government and legislation are matters of reason and judgement, and not of inclination”. Others, like the journalist Charles Mackay, whose book on economic bubbles and crashes, Extraordinary Popular Delusions and the Madness of Crowds, had an even more damning view of the crowd’s capacity to exercise either judgement or reason.
The thing is, if you believe that ‘the crowd’ isn’t wise then there isn’t much point in encouraging participation – these sorts of activities can only ever be tokenistic or a way of legitimising the decisions taken by others.
There are then those political philosophers who effectively argue that citizens’ incompetence doesn’t matter. They argue that the aggregation of views – through voting – eliminates ‘noise’ which enables you to arrive at optimal decisions. The larger the group, the better its decisions will be. The corollary of this view is that political decision making should involve mass participation and regular referenda – something akin to the Swiss model.
Another standpoint is to say that there is wisdom within crowds – it’s just that it’s domain specific, unevenly distributed and quite hard to transfer. This idea was put forward by Friedrich Hayek in his seminal 1945 essay on The Use of Knowledge in Society in which he argues that:
Hayek argued that it was for this reason that central planning couldn’t work since no central planner could ever aggregate all the knowledge distributed across society to make good decisions.
More recently, Eric Von Hippel built on these foundations by introducing the concept of information stickiness; information is ‘sticky’ if it is costly to move from one place to another. One type of information that is frequently ‘sticky’ is information about users’ needs and preferences. This helps to account for why manufacturers tend to develop innovations which are incremental - meeting already identified needs - and why so many organisations are engaging users in their innovation processes: if knowledge about needs and tools for developing new solutions can be co-located in the same place (i.e. the user) then the cost of transferring sticky information is eliminated.
These assumptions about the distributed nature of knowledge underpin both concepts of open innovation and collective intelligence. The latter was popularised by James Surowiecki in his book The Wisdom of Crowds, in which he argued that crowdsourcing is a superior method for, among other things, sampling and forecasting. Essentially, he describes the phenomenon of aggregating information in groups, where the information it aggregates doesn’t have to be perfect and you don’t need smart participants to get smart aggregate decisions. The concept of open innovation has similar theoretical foundations and is based on the idea that a single organisation can’t contain all the knowledge and skills required to develop new products and services and should source these ideas externally.
If one subscribes to the view that knowledge is widely distributed across society, then the task for policymakers is to tap into this expertise, which then has implications for the kind of engagement that’s necessary - it could mean a greater focus on crowdsourcing or collaboration with small groups of expert citizens rather than, for example, mass voting or polling.
There is growing evidence on how crowdsourcing can be used by governments to solve clearly defined technical, scientific or informational problems. Evidently there are significant needs and opportunities for governments to better engage citizens to solve these types of problems.
There’s also a growing body of evidence on how digital tools can be used to support and promote collective intelligence.
Nesta’s recent research on the subject has examined how innovative patient organisations are working as collectives to assemble and analyse information involved in healthcare, and in particular in managing long term conditions. Some of these patient organisations are already supporting the development of peer relationships, driving landmark research programmes, sharing skills and unlocking the energy and expertise of patients. Indeed, our research demonstrates that where citizens are highly motivated regarding specific issues they can and do self-organise to access, interpret and distribute large amounts of complex information and take decisive action in innovative campaigns.
But what about problems which are normative or values based? Can the tools and principles of open innovation be applied to democratic institutions such as parliaments and political parties which are arenas for contestation about the public good, and not simply marketplaces for ideas?
For example, experts can tell you how to build a nuclear power station but they can’t really tell you whether you should build power stations since that isn’t a purely technical question. In these cases, it’s not entirely straightforward what a ‘good decision’ might look like. If there is no such thing as an objectively correct answer then why not open it up to the crowd – especially where there is significant public appetite? If you take the Hayekian view, the crowd are more likely to come to an optimal decision than a group of elected representatives.
However, is the aggregation of votes really the best mechanism for getting a smart answer? As our ongoing research suggests, in some cases, it’s just as useful to understand the plurality of opinions and relative priorities as it is to understand the majority view. So, for example, if you simply ask people how a city should spend its infrastructure budget you will probably get a list of ideas and lots of comments without really any understanding of people’s relative priorities. However, if you structure a participatory budgeting process to enable people to vote and comment on their favourite ideas, and rank their priorities, then public officials will have far greater information on which to make decisions.
For some questions, there are no straightforward yes or no answers. Where the question is particularly complex, it might be as useful to know why people vote in a particular direction, as much as whether they vote yes or no. In some cases, a completely legitimate answer might be ‘maybe, it depends’. One good example of this is the recent referendum on the UK’s membership of the EU. Even though a majority of the voting public voted to leave the EU, it’s not at all clear why we voted to leave. The yes/no vote didn’t give an indication of people’s relative priorities in terms of trade, controls on immigration, sovereignty, public spending or the myriad other issues discussed during the campaign. There is currently no consensus on what brexit means and there is no mandate for one type of brexit over another since the referendum didn’t ask the public what it might want from a new kind of relationship with the EU.
So, the critical task for public officials is to have greater clarity over the purpose of engagement - in order to better understand which methods of engagement should be used and what kinds of groups should be targeted.
At the same time, the central question for researchers is when and how to tap into collective intelligence: what tools and approaches can be used when we’re looking at arenas which are often sites of contestation? Should this input be limited to providing information and expertise to be used by public officials or representatives, or should these distributed experts exercise some decision making power too? And when we’re dealing with value based judgements when should we rely on large scale voting as a mechanism for making ‘smarter’ decisions and when are deliberative forms of engagement more appropriate? These are all issues we’re exploring as part of our ongoing programme of work on democratic innovations.
 This kind of information is sticky for other reasons: it might be highly contextualised or require the use of existing skills or knowledge.