Politics is the rare sport where the amateur is better than the professional.
— Lawrence Lessig, professor at Harvard Law School, interviewed in 2011
Suppose that you lived in a world thirsting for cheap and non-polluting energy sources. Imagine, further, that we had, at several points in history, known how to produce energy in a way that was simultaneously safe, clean and affordable.
Unfortunately for us, this knowledge was mostly forgotten over the centuries and, thus, this technology had disappeared from mainstream use. We found ourselves forced to rely on dirtier and more dangerous ways to obtain energy.
In some remote parts of the world, though, this technology had been revived and was currently being used successfully. There was also a vibrant community of academics and practitioners who, for several decades, had been working out the details of how to best use this technology and the ways in which it could be further improved.
What I will argue in this chapter (and, more broadly, throughout this book) is that the little-known practice of citizen deliberation similarly has the potential to help us address one of the fundamental challenges facing us today. It might not be quite on the scale of cold fusion—but, luckily for us, it is not quackery, either.
♦ ♦ ♦
So what is citizen deliberation and how can we make use of it? The fundamental idea is a radically simple one. A group of ordinary citizens is tasked with collectively deciding on a policy matter. They consult with experts, listen to advocates representing different interest groups and, with the assistance of skilled facilitators, engage in careful, reasoned group discussions in which they explore the issues at hand. Throughout the entire process, the citizen panel is autonomous and its actions self-directed: it decides on, for example, the information it needs to gather from external sources, which experts or advocates to interview and what questions to ask them. A professional administrative and research staff assists the citizens in these duties.
After an adequate deliberation period, the group makes a collective decision on the topic by taking a vote and then issues a public statement. Its decision can be integrated into our existing political structures in several different ways, some of which we will review at the end of this chapter. 
Now, how does one go about selecting ordinary citizens to participate in these deliberative panels? One doesn't. Citizens are recruited from the electorate at random—the same way they get called up for jury duty in Anglo-Saxon (and other) countries—and then are appointed for a single, non-renewable term.
This may seem crazy at first. For many, it conjures up images of raucous popular assemblies where spirits run wild, and only those who shout the loudest get heard. Readers who followed the debate about healthcare reform in the US might remember such sorry scenes from the “town hall meetings” held in the summer of 2009, many of which quickly degenerated into little more than shouting matches. Rest assured that what I propose in this chapter bears no relation to that.
Although most of us have never heard of citizen deliberation, the use of a lottery to select citizens for political duty dates back to ancient Athens, where it was an established practice. The Greeks understood that choosing individuals from the citizenry at random is the only way to defend against the different forms of corruption that plague a professional political class. By entrusting power to a panel of citizens drawn by lot and having them serve a single, non-renewable term, most of the problems described in the second chapter of this book can be avoided. Free from the pressure of seeking reelection, on the one hand, and from the biases inherent in being part of a powerful elite, on the other, randomly chosen citizens are able to pursue what best serves the public interest. 
♦ ♦ ♦
With an understanding of the rationale for randomly recruiting citizens, the question that immediately comes to mind is whether ordinary citizens have what it takes. Can they possibly be smart enough? To the surprise of many, this turns out to be an unfounded concern. Over the last three decades, countless citizen panels have been convened all over the world, and experience tells us that, if the process is set up in the right way, citizen panels are perfectly capable of analyzing and deciding on complex policy matters.
Skeptical? Let us survey what some leading scholars in the field of citizen deliberation have written on the topic:
♦ ♦ ♦
Remember that, in these panels, citizens are not left to discuss and decide on policy matters based just on their prior knowledge of the topics involved.
Instead, citizen panels function by summoning policy and scientific experts to provide them with testimonies and vital information on the topic being discussed. The panel questions these experts, and their explanations and advice are pitted against that of experts who hold different views. In what constitutes a key part of the deliberative process, citizen panelists (assisted by trained facilitators) critically assess the evidence presented to them. Thus, the decisions they reach are based on a comprehensive, rigorous understanding of the issue(s) before them.
It's worth pointing out that, in their initial lack of knowledge and understanding of specific policy matters, members of citizen panels are not that different from our elected representatives. For any piece of legislation being considered in parliament, most of its members will not come close to being “experts” on that particular topic, either.
♦ ♦ ♦
But even if we believe that ordinary citizens have the skills required for the job, isn't it dangerous to randomly recruit the members of these citizen panels? Our first instinct is probably to draw upon our personal reservoir of prejudices and envision a panel made up of particularly unsavory characters who are handed the power to formulate and/or review public policy! How can that possibly be a good idea? 
Fortunately, this fear is misplaced, as simple calculations can attest. As long we accept that a substantial majority of our fellow citizens are decent, honest people, then the probability of drawing a citizen panel dominated by unsavory characters is abysmally low. Consider the hypothetical case of citizen panels made up of 24 participants. If you believe that one out of every ten people is a bad apple, then the chances of half or more of the participants in that panel being rotten are practically zero. This remains true under even more-pessimistic worldviews: for those who guess that one out of every five people (a whopping 20% of the population—reason enough to hardly ever leave home) is not to be trusted, the chance of them making up half or more of the people on that 24-seat panel remains at 0.10%.
Better yet: we can actually make the odds of unsavory types taking control of a citizen panel as low as we wish. It does come at a cost, but it can be achieved in either of two different ways.
The first option is to make the panel larger. By having more citizens on the panel, the proportion of bad apples sitting on it will tend to be closer to the proportion in the general population (1/10 or 1/5, respectively, in the examples above). The fact that deliberation works best in relatively small groups (experienced facilitators say that panels shouldn't have more than 25 participants) does not pose a significant obstacle since larger panels can be—and frequently are—subdivided into multiple smaller work-groups. 
The second option is to increase the percentage of votes required for the panel to pass a decision. For example, if we demand a “supermajority” of at least 60% (rather than just 50% plus one vote), then the likelihood of there being enough unsavory types on the panel for them to effectively control it will also become smaller.
Taken together, these two parameters give us good reason to trust that the nightmare scenarios that we all involuntarily conjure when we think of a random grouping of our fellow citizens will stay that way—more a portrayal of our deep-seated fears about the society we live in than a situation likely to occur.
Does this mean we can be absolutely sure that such a system would never produce a panel dominated by people we should rightly be concerned about? No. But electoral politics likewise provides no assurances of that, as a cursory glance at twentieth-century history books will attest. (We will revisit this topic in the conclusion.)
♦ ♦ ♦
Exploring the implications of randomness a bit further, there are some other things we can say about the composition of citizen panels recruited in this way. In particular, we know that participants will tend to be as smart, educated and wise as the average citizen. The panels' demographic make-up will also tend to mirror that of the general population in terms of gender, age, race, occupation and socioeconomic status.  And the same will be true of political views and attitudes: the people who comprise citizen panels will tend to be from different political persuasions in the same proportions as we find them in the general population.
Compare this—the prospect of policy being formulated and/or reviewed by groups of citizens who mirror the make-up of the general population—with our current system. As mentioned earlier, professional politicians are very different from the rest of us in several important ways. The political class is disproportionately made up of white males from a narrow range of professional backgrounds who are significantly wealthier than the average citizen they are supposed to represent. Citizen panels, on the other hand, would constitute a true cross-section of our society. Statistically speaking, such citizen panels would be much more “representative” of the general citizenry than our elected representatives ever have been.
♦ ♦ ♦
So, where and for what purpose has citizen deliberation been used in recent times? Let's survey two different real-world cases. 
The first of these took place in the Canadian province of British Columbia. In 2004, the provincial government randomly recruited 160 citizens to form a “citizens' assembly,” which was asked to investigate how the provincial voting system should be reformed. The government promised up-front that the assembly's proposals would be put up for a vote in a referendum and, if approved by the population, implemented.
Over the course of eleven weekends, these citizens learned about different electoral systems, consulted with experts and eventually decided to propose a voting system based largely on what is known as “Single Transferable Vote” (STV).  Dating back to the early 19th century, STV is a well-known alternative to the two most common voting systems and is currently used for national elections in Ireland, Australia and Malta. In the words of David Farrell (professor of politics at University College Dublin), STV is the voting system that “politicians, given a choice, would probably least like to see introduced but which voters, given a choice, should choose.”
When the referendum took place, the changes proposed by the citizens' assembly were approved by a majority of 57% of the population. However, the results failed to fulfill all the strict requirements the government had imposed for them to be considered binding, and, disappointingly, the reforms were never implemented.
Another famous case of citizen panels wielding real power is that of the Citizens' Initiative Review process in the US state of Oregon. Referendums have a long history in that state, dating back to 1902. More than a century later, in 2010, Oregon public officials agreed to introduce a substantial innovation in the way ballot measures are handled.
What they did was to add a deliberative “layer” to the process. Before a referendum is held, a panel of 24 randomly chosen citizens now deliberates for a number of days on the measure being proposed. After interviewing advocates on both sides and consulting scientific experts who provide them with in-depth information on the topic, these citizens carefully analyze the question before them and conclude their work by issuing a public statement.
Written in everyday language and not more than two or three pages long, this document includes the panel's key findings about the choice facing the electorate; short group statements by the panelists who support and oppose the ballot measure; and any additional considerations that the panel collectively deems relevant. How many panelists support and oppose the measure at the end of the deliberation period is clearly indicated; the same information is provided for each of the panel's key findings, allowing readers to easily gauge how convincing the panel found each of them once the group had carefully researched the issue.
The full statement is then included in the “voter's pamphlet” that all registered voters in Oregon receive in the mail before a referendum. Recently conducted research by John Gastil and colleagues at Pennsylvania State University (quoted earlier) indicates that these statements not only succeed in making voters more knowledgeable about ballot measures, but also substantially influence the voting behavior of those who read them.
The cases of British Columbia and Oregon clearly refute the establishment's warnings of the “tremendous risks” inherent in the kind of “democratic experimentalism” advocated in this book. As these tales make clear, there is no reason to believe that bringing citizen deliberation into the core of our political systems would somehow wreak havoc on society. And I am not alone in my views. Consider, for example, that the Oregon legislature itself recently chose—after a successful one-year trial period—to pass a law making the Citizen Initiative Review a permanent feature of the way referendums are conducted in that state. All available evidence suggests that citizen deliberation is a safe and promising way to democratize our political institutions.
♦ ♦ ♦
So, how might we incorporate citizen deliberation into existing political systems? In the short-term, a reasonable goal would be to import the deliberative apparatus of the Citizen Initiative Review process found in Oregon. Later in this book, we will see how it can be fruitfully combined with referendums that increase our control over the political class. Most countries already have referendums in some form: adding a citizen review panel to the process—and ensuring that all registered voters had easy access to the panel's conclusions—would serve two important purposes. First, it would contribute to more-informed decision-making on the part of the electorate, as the evidence from Oregon shows. Second, it would help familiarize the general population with citizen deliberation, thus paving the way for its broader use.
Acquainting citizens with the process of citizen deliberation matters. It is my personal experience that, even in a context of total disillusionment with, and often outright hostility towards, the political class, citizens will be extremely skeptical of having other “ordinary people” actively engaged in politics. Those chanting in protest outside a besieged parliament might nod in agreement upon hearing a proposal to summarily jail those in power, yet will most often balk at the notion of their fellow citizens being directly involved in policy matters.
This state of affairs is sad but understandable. When it comes to politics, ordinary citizens have, for too long, been reduced to the role of mere spectators. Thus, suggesting that things could be different will inevitably seem alien to most. Combined with our tendency to harbor nearly pathological levels of distrust towards anyone outside our immediate social circle, this creates substantial challenges for anyone advocating greater citizen involvement in policy-making. (We will revisit this topic in the conclusion.)
In the long term, we should aim higher and look at some of the more ambitious proposals that have been advanced. The main idea is to have a large citizen panel—ranging in size from, say, 50 participants to the few hundreds typical of a lower house of parliament—constitute a separate legislative chamber. If this sounds strange, think for a moment about the tremendous potential demonstrated by citizen deliberation in both British Columbia and Oregon over the last decade. After giving it some thought, you might find yourself wondering instead: if ad hoc citizen panels work that well, why not try to tap into this source of reasoned, public-spirited decision-making on a more permanent basis? This is the rationale behind those bolder proposals.
A “citizens' chamber” would be picked at random from the general population. Its members would take a legally-protected leave from their regular jobs and receive an adequate wage for their work. Throughout their time in office, they would be assisted by staff who would not only act as facilitators, but would also support them in their various duties. Each citizens' chamber would serve a single, non-renewable term that should be sufficiently long for its members to get acquainted with their new role, but also short enough to prevent them from becoming too accustomed to the continued exercise of power.
This citizens' chamber would review the measures passed by elected representatives. With a sufficiently large supermajority, it would be able to block (or demand amendments to) the decisions made by the traditional elected chamber. In situations of irreconcilable disagreement, a deadlock would be resolved through a referendum: the public would decide which of the two chambers of parliament to side with. 
Obviously, any plan of this sort faces a long wait before seeing the light of the day. In the meantime, we can make significant inroads by promoting the establishment of citizens' chambers at the city level. The recent popularity of participatory budgeting in cities all around the globe provides more than just grounds for optimism: it could also be used to convincingly argue that the (perceived) distance separating us from this kind of solution is not that great after all.
11. ^ It should be clear from the outset, though, that these citizen panels should have more power than merely producing “recommendations” for the benefit of the government and/or the state bureaucracy.
12. ^ In the wake of the Occupy movement, some argued that “popular assemblies,” in which all citizens who wished to do so would be able to freely participate in the decision-making process, could also help us avoid those problems. However, they are plagued by their own serious difficulties. First, popular assemblies do not scale to a large society. Second, they are vulnerable to manipulation by powerful interests who are able to more effectively organize and sponsor the participation of their own supporters. Third, when all are invited to talk, often only the most motivated—and most extreme—voices will make the effort to be heard. Although it might seem paradoxical at first, a random few are preferable to the unfiltered many when it comes to representing the totality of the public.
13. ^ You might notice that this section does not define who, exactly, these “unsavory” types might be. This is done on purpose. This section's argument and simple math apply regardless of the particular prejudices (or ideological preferences) of each reader.
14. ^ See, for an extreme example, the several “Citizen Summits” organized by AmericaSpeaks in which thousands participated.
15. ^ A common sampling method known as stratified random sampling can be used to ensure that all panels effectively mirror the general population across such key demographic traits.
16. ^ Note that this is in addition to many hundreds of “citizen juries” organized by governments the world over in recent decades. Such citizen juries function in a way similar to that described above, yet are typically devoid of any real political power. Instead, governments use them either as “beefed-up” focus groups (to determine which decisions might be acceptable to the public) or, at worst, to help legitimize decisions which, in fact, have already been made.
17. ^ We will revisit the topic of voting systems in the next chapter.
18. ^ For an example of a more nuanced proposal, see Terrill Bouricius' article Democracy Through Multi-Body Sortition: Athenian Lessons for the Modern Day.